FUEL PUMP WITH INJECTION PRESSURE CHAMBER

FOR A JET ENGINE AFTERBURNING SYSTEM

Alexandru Nicolae TUDOSIE
University of Craiova, Faculty of Electrical Engineering, Avionics Department,
e-mail: atudosie@elth.ucv.ro, antudosie@yahoo.com
Abstract This paper deals with a fuel pump with
injection chamber, used as main fuel supplying system
for an afterburning system, which can be able to serve
on a jet engine (single-jet or twin-jet). The fuel system
consists of: fuel pump with actuator, pressure ratio
transducer, fuel injection chamber and on/off switching
system. The author has established the mathematical
model of this fuel system, which means the non-linear
form, the linearised form as well as the non-dimensional
form, based on the small perturbations method. This
mathematical model, after the Laplace transformation
applying, represents the systems description as
controlled object, as well as the basic elements of the
block-diagram with transfer functions building.
Furthermore, the author has realized some studies
about the systems stability and quality; the conditions
for the systems stability leads to some interesting
results, concerning the criteria for the injection
chambers parts choice. Systems quality studies consist
of the system step response for the main controlled
parameters, offering an image about the systems
behavior, as well as about its operability and possibility
of interconnection with other systems. The paper is
based on some similar works of the author, as well as on
similar works presented in the reference list, about
aircraft jet engines and engines main parts as
controlled object(s).
Keywords: Fuel, Pump, Injection, Afterburning, Jet-
engine.
1. INTRODUCTION
For the aircraft jet engines, the most efficient thrust
augmentation method is the afterburning, in spite of
its huge fuel consumption.
The gas-dynamic principles, as well as the equations
of the afterburning system are presented in [2], [3]
and [6]. Meanwhile, the afterburning system as
controlled object is depicted in [7], [9] and [11]; a
possibility for the afterburners fuel pump automatic
control was presented by the author in [14] and a
similar simplified system in [11].
The purpose of this paper is to identify the
afterburning fuel pump in fig.1 as controlled object,
by describing it through its mathematical model as
well as through its transfer function and performing
some stability and quality studies, which can be
extended for a whole class of such fuel systems,
useful for different afterburners.
The fuel system in fig. 1 consists of the following
main parts: I-fuel pump; II-pumps actuator; III-
injection pressure chamber; IV-engines gas turbine
pressure ratio transducer; V-afterburning on/off
switching system.
One can observe that the fuel flow-rate depends on
the pumps rotor speed as well as on the pumps plate
position (cline angle); fuels pressure is kept constant
by the valves in the injection chamber, so the
controlled parameter is the injection fuel flow-rate
, as well as the injection pressure . The control
parameters are the pressures before and after the
engines gas turbine, that means the turbines total
pressure ratio . The system must assure the co-
relation between the turbines pressure ratio and the
injected fuel flow-rate during the engines operation,
no matter the flight regime is.
i
Q
i
p
*
T
o
The pump is permanently turned round by the
engines shaft, but the afterburning operates only by
command, so, when the afterburning is switched-off,
the fuel pump only re-circulates the fuel, without
injection and burning. This manner of operating is
realized by the on/off switch valve and its command
equipment (V): when the afterburning is switched-off
the electro-magnet 41 moves the slide-valve 42 to the
right, so the 38 actuators B-chamber is supplied with
high pressure fuel. Consequently, the spring 40
moves the piston 39 and the slide valve 37 to the left,
so the injection pressure chamber (III) supplying
main pipe 32 is cut-off (so the afterburners injectors
become inactive). Meanwhile, the pump actuators B-
chamber is supplementary discharged, so the piston
23 moves to the right and its rod 22 reduces the
mobile plates angle to the minimum value, which
assures only the fuel re-circulation through the
systems components. When the afterburning is
switched-on, the mobile plates angle is established
by the actuators piston 23 positioning, commanded
by the pressure balance in the actuators command
block.
2. SYSTEMS MATHEMATICAL MODEL
The studied systems mathematical model consists of
the motion equation for each of its parts. The non-
linear equation will be transformed, in order to bring
them to an acceptable form for further studies and
229
____________________________________________________________
Annals of the University of Craiova, Electrical Engineering series, No. 34, 2010; ISSN 1842-4805
n
1
2
3
4 8
7
A B
I
5
A
B
Fuel tank
0 0 < < y
0 0 < < z
0
0
<
<
u
0 0 > >z
r
0
0
<
<
u
r
0 0 > >x
p
2
*
p
4
*
6
4
20
21 22
II
25
23
24
26
9
19 18
12
13
16
14 15
III
11
10
Q
i
30
31
27
28
29
IV
V
36
35
33
34
32
37 38 39
40
41
42
p
4
*
p
R
p
C
p
i
0 0 > >x
17
Figure1: Fuel systems constructive and operating schematic
I-fuel pump; II-pumps actuator; III-injection pressure chamber; IV-engines gas turbine pressure ratio transducer; V-after-
burning on/off switching system; 1-pump rotor; 2-pump cradle; 3-mobile plate; 4-plunjers, 5-lid; 6-bearing; 7,8-limitators; 9-
fuel filter; 10-drossel; 11-hemispherical valve; 12,16-adjustment bolts; 13,15-springs; 14-slide-valve;17-injection pipe; 18-
pump pressure chamber; 19-constant pressure valve; 20-actuators supplying block; 21-slide-valve; 22-actuators rod; 23-
actuators piston; 24-actuators spring; 25-pump supplying pipe; 26-main discharging pipe; 27-transducers elastic mem-
brane; 28- transducers rod; 29- transducers spring; 30-variable fluidic resistance; 31-drossel; 32-main fuel pipe; 33-actu-
ators supplementary discharging pipe; 34-switching systems supplying pipe; 35, 42-on/off-valves; 36, 40-springs; 37-
switching systems main slide valve; 38- switching systems actuator; 39-piston; 41-electro-magnet.
simulations.
2.1. Systems non-linear equations
The system non-linear motion equations are
a) fuel pump equations
, (1) ) , ( y n Q Q
p p
=
, (2)
a p C
Q Q Q =
where is the pumps fuel flow rate,
p
Q n rotor
speed, actuators rod displacement, y
a
Q com-
mand fuel flow rate, effective fuel flow rate;
C
Q
b) fuel pump actuator equations
t
y y
S
t
p
V Q
s
B
B
B B
d
) ( d
d
d
0
+
+ = | , (3)
t
y y
S
t
p
V Q
s
A
A
A A
d
) ( d
d
d
0

= | , (4)
,
2
a A a a A
p p x b Q =

(5)
,
2
0
p p x b Q
B a a B
=

(6)
( ) ( ) + + + + =
s s a B B A A
y y
t
y y
t
m S p S p
d
d
d
d
2
2

( )
s ea
y y k + + , (7)
where
B A
Q Q , actuators chambers fuel flow rates,
through the slide valves slots; fuel pressure
in actuators chambers; actuator cham-

B A
p p ,

0 0
,
B A
V V
230
____________________________________________________________
Annals of the University of Craiova, Electrical Engineering series, No. 34, 2010; ISSN 1842-4805
bers volumes;
B A
S S , pump actuator pistons
surfaces areas; | fuels isotherm compressibility
co-efficient;
a
flow rate co-efficient;
a
b slots
width; actuators rod displacement; y
a
p supply-
ing pressure; low pressure fuel circuit pressure;
slide valves displacement; actuator piston+
rod ensembles mass; actuator spring elastic
constant;

0
p
x
a
m

ea
k
viscous friction co-efficient; fuel
density; time; t
c) pressure ratio transducer equations
( ) ) (
d
d
d
d
2
2
*
4 r et t R m
x x k
t
x
t
x
m p p S + + + = , (8)
*
2
1
2
p
S
S
p
R
= , (9)
where transducer membrane surface area;
gas pressure after the engines turbine; air
pressure after the engines compressor (considered
proportional to the pressure before the turbine , so
the turbine pressure ratio can be determined by

m
S

*
4
p
*
2
p
*
3
p
*
T
o
*
4
*
2
*
4
*
3 *
p
p
p
p
T
~ = o ); corrected pressure, given by
(9), because of the critical flow regime through the
31 and 31 fluidic resistances; effective areas
of the fluidic resistances 30 and 31; transducers
rigid center+rod ensembles mass; transducer
springs elastic constant; springs displacement
preset (pre-tensioning);

R
p
*
2
p

2 1
, S S

t
m

et
k

r
x
d) injection pressure chamber equations
,
2
10 10 i C C
p p z d Q =

t (10)
,
2
f i b i
p p bu Q =

(11)
t
u u
S
t
p
V Q Q
s
D
i
i i C
d
) ( d
d
d
0
+
+ = | , (12)
) (
4
2
10
z z k p
d
r e C
+ =
t
, (13)
, (14) (12) ) (
r ei i D
u u k p S + =
where fuel pressure in pumps chamber;
fuel injection pressure; pressure in the af-
terburner chamber; injection fuel flow rate;
injection chamber volume;

C
p

i
p
f
p

i
Q

0 i
V
10
,
b
flow rate
co-efficient of the injection chamber
orifices; injection drossel diameter;
10
d b
injection slot width; 13 and 15 springs
elastic constants;

ei e
k k ,
u z, 13 and 15 springs dis-
placement
r r
u z , 13 and 15 springs displacement
preset (pre-tensioning); 14 slide-valve frontal
area amount.

D
S
2.2. Linear equation system
The above determined non-linear equation system is
difficult to be used for further studies, so it can be
linearized, using the small perturbation method,
considering formally any variable or parameter X as
X X X A + =
0
and
0
X
X
X
A
= , where AX parameters
deviation,
0
X steady state regimes value and X
non-dimensional deviation.
Introducing the new form of each parameter into the
above mentioned equation system and separating the
steady state terms, one obtains a new for of the
system, as follows
y k n k Q
y n p
A + A = A , (15)
a p C
Q Q Q A A = A , (16)
y
t
S p
t
V Q
B B B B
A + A = A
d
d
d
d
0
| , (17)
y
t
S p
t
V Q
A A A A
A A = A
d
d
d
d
0
| , (18)
A Ap Ax A
p k x k Q A + A = A , (19)
B Ap Ax B
p k x k Q A A = A , (20)
y
t
T
t
T p S p S
a a B B A A
A
|
|
.
|

\
|
+ + = A A 1
d
d
d
d
2
2
2

, (21)
( ) x
t
T
t
T p p S
t t R m
A
|
|
.
|

\
|
+ + = A A 1
d
d
d
d
2
2
2 *
4
, (22)
*
2
1
2
p
S
S
p
R
A = A , (23)
(24) ( ,
i C Cp Cz C
p p k z k Q A A + A = A )
,
i ip iu i
p k i k Q A + A = A (25)
u
t
S p
t
V Q Q
D i i i C
A + A = A A
d
d
d
d
0
| , (26)
) (
4
2
10
z z
d
k
p
r
e
C
A + A = A
t
, (27)
) (
r ei i D
u u k p S A + A = A , (28)
where , ,
0 0
|
|
.
|

\
|
c
c
=
|
|
.
|

\
|
c
c
=
y
Q
k
n
Q
k
p
y
p
n

0
2
i
b iu
p
b k = ,
231
____________________________________________________________
Annals of the University of Craiova, Electrical Engineering series, No. 34, 2010; ISSN 1842-4805

0
0
2
1
i
b ip
p
bu k = ,
( )
,
2
0 0
10 10

t
i C
Cz
p p
d k

=
( )
,
2
1
0 0
0 10 10

t
i C
Cp
p p
z d k

0
2
A
a a Ax
p
b k = ,
,
2
1
0
0

A
a a Ap
p
x b k = , ,
ea
a
ea
a
a
k
T
k
m
T

= =
et
t
et
t
t
k
T
k
m
T

= = , . (29)
2.3. Non-dimensional linear equation system
Using some appropriate chosen amplifying terms, the
above-determined mathematical model can be
transformed in a non-dimensional one. After applying
the Laplace transformer, one obtains the non-
dimensional mathematical model, as follows
y k n k Q Q
py pn p C
+ = = , (30)
where one has used the observation that the
command fuel flow rate is very small comparative
to , so ;
a
Q
C
Q
p C
Q Q ~
( )
x
k
p p
a a SA
B A
t +
=
s
1
, (31)
( ) (
B A SA a a
p p k y T T = + + 1 s s
2 2

). (32)
i pi ui i
p k u k Q + = , (33)
i zi C zc C z
p k p k Q z k + = , (34)
r rp i up
u k p k u = , (35)
( ) ( )u k p k z k p
u u C ic iz i i
1 s 1 s + + = + t t , (36)
( )
* 2 2
1 s s
T t t
k x T T o
o
= + + . (37)
The used annotations in the above system, based on
the annotation presented in (29)-system, are as
follows:
,
1
,
1
0
0
0
0
|
|
.
|

\
|
c
c
=
|
|
.
|

\
|
c
c
=
y
Q
Q
k
n
Q
Q
k
p
p
py
p
p
pn
, ,
4
, ,
0
0
0
2
10
0
0
0
0
0
p
Cz
z
C
e
pz
i
i ip
pi
i
iu
ui
Q
z k
k
p d
z k
k
Q
p k
k
Q
u k
k = = = =
t
( )
, , ,
0
0
0
0
0
0
Cp ip i
C Cp
ic
p
i Cp
zi
p
C Cp
zc
k k p
p k
k
Q
p k
k
Q
p k
k
+
= = =
( )
, , ,
4
0
0 0
0
2
10
0
Cp ip i
i Cz
iz
Cp ip
i
i
C
r e
prz
k k p
z k
k
k k
V
p d
z k
k
+
=
+
= =
|
t
t
( )
, , ,
0 0
0
0
0
p p x
y S
k
S
k k p
u k
k
a a
A
a
iu
D
u
Cp ip i
iu
u

= =
+
=

t t
( )
. ,
2
,
1
2
*
40
0
0
0
0
S k
S S p
k
p p S
y k
u k
p S
k
et
m
a A
ea
a
ei
i D
up
=

= =
o
(38)
Based on some practical observation, one can make
some supplementary hypothesis: a) the fuel is a non-
compressible fluid ( 0 = | ), so the terms containing it
become null ( ) 0 =
i
t ; b) the inertial effects are very
small, as well as the viscous friction, so the terms
containing and
t a
m m ,
.
are becoming null ( = =
a a
T T
) 0 = = =
t t
T T . Consequently, eq. (31) and (32) give
x y
a a
t +
=
s
1
(39)
and from eq. (36) and eq (37) one obtains
( )u k p k z k p
u u C ic iz i
1 s + + = t , (40)
*
T
k x o
o
= . (41)
Meanwhile, the preset operation are realized during
the engine or pump ground tests, so the terms
containing and (as well as
r
u
r
z
r
u and
r
z ) become
null and shall be eliminated.
The new, simplified, mathematical model is given by
the equations (30),(39),(33),(34),(35),(40) and (41).
Based on it, the systems block diagram with transfer
functions is built and depicted in fig. 2. (This block
diagram contains the terms
r
u and
r
z too).
pn
k
_
pz
k
pi
k
s+1
u
t k
u
( )
n
+
_
_
i
p
C
p
C
p
C
p
u
i
Q
1
z k
z z
iz
k
z
_
prz
k
_
+
zr
ic
k
zc
k
C
p
_
_
+
s+1
i
t
1 i
p
i
p
i
p
i
p
zi
k
+
up
k
_
rp
k
_
+
u
r
u u
ui
k
_
+
+
p
Q
C
Q
py
k
y
s+
a
tt
1
a

x
o
k
o
T
-
_
+
+
p
Q
Figure2: Fuel system's block diagram with transfer functions
232
____________________________________________________________
Annals of the University of Craiova, Electrical Engineering series, No. 34, 2010; ISSN 1842-4805
2.4. Systems transfer functions
The block diagram with transfer functions in fig.2
shows that the system has two inputs ( n and
*
T
o ) and
one output (
i
Q ). Using the above-determined
mathematical models equations, one can obtain an
equivalent form:
( )
*
s
s
T
a a
py
pn i f f
k k
n k Q o
t
t
o
+
+ = + , (42)
where
a
t and
a
are given in (38) and
f
t and
f
are:
( )
( )( )
u
pi up ui pz ic iz
pz zc z up u
f
k k k k k k
k k k k k
t t
+ +
+
= , (43)
-2000

( )( ) ( )
( )( )
pi up ui pz ic iz
pz ic iz zi up u pz zc z
f
k k k k k k
k k k k k k k k k
+ +
+ + +
=
1
. (44)
Systems transfer functions expressions are:
( ) ,
s
s
a a
pn
n
k
H
t +
= (45)
( )
( )( )
.
s s
s
f f a a
py
k k
H
t t
o
o
+ +
= (46)
3. SYSTEM STABILITY, QUALITY AND
CONCLUSIONS
3.1. System stability
Taking into account the transfer functions
expressions (see (45) and (46)), one can affirm
(according the algebraic stability criteria) that for the
systems stability the conditions must be:
, 0 >
a
(47
/
)
. 0 >
f
(47
//
)
While the first condition is always realized,
( )
, 0
2
0
0
>

=
p p S
y k
a A
ea
a
(48
/
)
because all the involved terms are strictly positives,
the second condition

( )( ) ( )
( )( )
0
1
>
+ +
+ + +
pi up ui pz ic iz
pz ic iz zi up u pz zc z
k k k k k k
k k k k k k k k k
, (48
//
)
leads to a relation between the injection pressure
chamber springs elastic constants, represented by a
hyperbolic curve (see dashed curve in fig. 3), which
separates the stability domain from the instability
domain. The stability domain is the one above the
curve, on the right side of the abscissa axis. One can
observe that, if the spring 15 and the valve 14 are
absent , there is a minimum value for the
elastic constant of the 13-spring, which means the
lowest possible elastic constant value for system
stability,
0 =
ei
k
e
k
( )
min e
k .
3.2. System quality
As one can see in fig. 1 and 2, the studied system has
two input parameters: a) engines operating regime -
(given by the engines speed n); b) gas turbines
pressure ratio ( )
*
T
o . So, the system should operate in
case of any changing affecting one or both of these
input parameters.
A study concerning the system quality was realized
using the co-efficient values for a pump of a VK-1F
jet engine, by analyzing its step response, based on
the simulated block-diagram in fig. 2 (systems
response for step input for one or for both above-
mentioned parameters). As output, one has
considered the fuel flow rate
i
Q as main parameter,
as well as the pump injection pressure
i
p as
secondary parameter.
Output parameters behavior is presented by the
graphics in fig. 4 and 5; the situation in fig. 4.a has as
input gas turbines pressure ratio, for constant engine
speed; meanwhile the situation in fig. 4.b has as input
the engines regime (step acceleration or step
throttles repositioning). Systems behavior for both
input parameters step input is depicted in fig. 5 .
One can observe (in each figure, from 4 to 5) that all
the output parameters are asymptotic stables, so the
system is a stable-one. All output parameters are
stabilizing at their new values with static errors, so
the system is a static-one; however, the static errors
are acceptable, having values smaller than 1%. When
both of the input parameters have step variations, the
effects are overlapping, so systems behavior is the
one in fig. 5.Concerning the stabilization time, both
for
i
Q and for
i
p its value is somewhere around
4 5 . 3 s , which is an acceptable value for usual fuel
systems (pumps).
0
2000
4000
6000
8000
10000
12000
00 -3000 -2000 -1000 0 1000 2000 3000 4000 5000
STABILE
UNSTABLE
k
e
[N/m]
k
ei
[N/m]
-40
(k
e
)
min
Figure 3: Systems stability domain
233
____________________________________________________________
Annals of the University of Craiova, Electrical Engineering series, No. 34, 2010; ISSN 1842-4805
This kind of fuel systems (pumps with plunjers and
mobile plate) can be used both for basic jet engines
(single jet, single or double spool) and for their
afterburning systems. The interconnection engine-
afterburner must be carefully realized, in order to
keep them in the stability area of operability.
References
[1] Abraham, R. H. Complex dynamical systems.
Aerial Press, Santa Cruz, California, 1986.
[2] Ciobotea, V. Aircraft engines theory. Publisher
of Military Technical Academy, Bucharest, 1978.
[3] Hill, P. G., Peterson, C. Mechanics and Thermo-
dynamics of Propulsion. Addyson-Wesley Pub.,
New York, 1993.
[4] Lungu, R. Flight apparatus automation. Publisher
Universitaria, Craiova, 2000.
[5] Mattingly, J. D. Elements of gas turbine
propulsion. McGraw-Hill, New York, 1996.
[6] Pimsner, V. Air-reactor engines. Processes and
characteristics, Ed. Didactica si Pedagogica.
Bucuresti, 1984.
[7] Stoenciu, D. Aircraft engine automation. Aircraft
engines as controlled objects. Publisher of
Military Technical Academy, Bucharest, 1977.
[8] Stoenciu, D. Aircraft engine automation. Catalog
of automation schemes. Publisher of Military
Technical Academy, Bucharest, 1986.
[9] Stoicescu, M., Rotaru, C. Turbo-jet engines.
Characteristics and control methods. Publisher of
Military Technical Academy, Bucharest, 1999.
[10] Tudosie, A., Jet engine rotation speed hydro-
mechanical automatic control system, Pro-
ceedings of the Scientific Session 25 Years of
High Education in Arad, Arad, 30-31 october,
1997, section 8, pp. 177-184.
[11] Tudosie, A. Aerospace propulsion systems
automation. Inprint of Univ. of Craiova, 2005.
[12] Tudosie, A. N., Jet Engines Speed Controller
with Constant Pressure Chamber, Proceedings of
the WSEAS International Conference on
Automation and Information ICAI08, Bucharest,
June 26-28 2008, pp. 229-234.
[13] Tudosie, A. N., Jet engines rotation speed
control based on the fuels injection differential
pressures control, Annals of the University of
Craiova Electrical Engineering Series, no. 32,
Craiova, 2008, pp. 231-238, ISSN 1842-4805.
[14] Tudosie, A. N., Mathematical Model of a Jet
engine Afterburning System, Proceedings of the
WSEAS International Conference on Automation
and Information ICAI10, Iasi, June 13-15 2010,
pp. 159-164.
[15] Zelezny, Z. Simulation of Hidro-Mechanical
Control System of Turbofan Engine. Proceedings
of the XXVII-th Scientific Session, Military
Technical Academy, Bucharest, 1997, pp. 17-24.
x 10
-3
0
1
2
3
4
5
6
7
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
t [s]
p
i
p
i
Q
i
Q
i
Figure 5: System step response for combined step input
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
x 10
-4
0
1
2
3
4
5
6
7
t [s]
p
i
p
i
Q
i
n=0
Q
i
x 10
-3
0
1
2
3
4
5
6
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
t [s]
p
i
p
i
Q
i
Q
i
o =0
T
*
a) step input of
*
T
o and 0 = n ; b) step input of n and 0
*
=
T
o ;
Figure 4: System step response for different step inputs
234
____________________________________________________________
Annals of the University of Craiova, Electrical Engineering series, No. 34, 2010; ISSN 1842-4805