Putra Adnan 2018 J. Phys. Conf. Ser. 1005 012026
Putra Adnan 2018 J. Phys. Conf. Ser. 1005 012026
Abstract. In this paper, an aerodynamic design of a single stage gas generator axial turbine for
small turbojet engine is explained. As per design requirement, the turbine should be able to
deliver power output of 155 kW at 0.8139 kg/s gas mass flow, inlet total temperature of 1200 K
and inlet total pressure of 335330 Pa. The design phase consist of several steps, i.e.:
determination of velocity triangles in 2D plane, 2D blading design and 3D flow analysis at
design point using Computational Fluid Dynamics method. In the determination of velocity
triangles, two conditions are applied: zero inlet swirl (i.e. the gas flow enter the turbine at axial
direction) and constant nozzle angle design (i.e. the inlet and outlet angle of the nozzle blade are
constant from root to tip). The 2D approach in cascade plane is used to specify airfoil type at
root, mean and tip of the blade based on inlet and outlet flow conditions. The 3D approach is
done by simulating the turbine in full configuration to evaluate the overall performance of the
turbine. The observed parameters including axial gap, stagger angle, and tip clearance affect its
output power. Based on analysis results, axial gap and stagger angle are positively correlated
with output power up to a certain point at which the power decreases. Tip clearance, however,
gives inversely correlation with output power.
1. Introduction
Small turbojet engine used for Unmanned Aerial Vehicle (UAV) has being developed in our laboratory.
This turbojet engine consist of three main components, namely compressor, combustor and turbine. In
the design of turbojet engine, a turbine is considered as a component which can stand high temperature
loads of heated air from combustion chamber. The kinetic energy of the turbine in rotation is used to
drive the compressor so that the engine attains idle condition and to produce thrust output. For small
turbojet engine, the use of a single axial turbine is to give simplicity and to provide lower cost in
development and manufacture. In addition, using the simple turbine, the engine is easy in maintenance.
In this study of design axial turbine, there are several previous research that become a basis and
reference, for example D.G Ainley and G.C.R. Matheison mentions in Ref. [1] that the information
obtained from the cascade method (which gives two-dimensional of blade section) is not enough because
the three- dimensional effect are so powerful. Donald E. Holeski and Samuel M. Futral predict turbine
efficiency decreased linearly with increasing rotor tip clearance [2]. James Allen, Ref [3], predict that
axial clearance (axial gap) can affect turbine efficiency and torque produced by the turbine. In this work,
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
the design of the axial turbine of small turbojet is carried out by considering constant nozzle. Two
methods namely parametric cycle analysis (analytical) and computation fluid dynamics (numerical)
approaches are used to obtain the axial turbine design which fulfills the turbine design requirement and
objectives. With this study of design a simple axial turbine, be expected can contribute to
turbomachinery field in develop a high performance turbine with low cost in manufacture and
maintenance.
The following are the brief explanation of the steps from the method that used (analytical method and
analytical method)
2
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Figure 2. Characteristics of typical velocity triangle through turbine blade (left) and Zero Inter Stage
Swirl Velocity Triangle (right) [4]
As shown in the Figure 3, the velocity triangle of each section is changed from tip section to the hub
section of the turbine blade. Changes on the flow angle is significantly can be seen at the tip section to
the mean section of the blade (the flow angle is reversed). The flow angle for mean section to the hub
section is increased from 22.4 degree to 41 degree. The following are the illustration of velocity triangle
of the turbine at the tip, middle (mean), and hub section.
3
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Figure 3. Illustration of velocity triangle of each section of designed turbine blade (tip (left), mean
(middle), and hub (right) [4]
The following are the detail calculation to derive the specific flow angle at each blade section (tip, mean
and hub section). From the velocity triangle, it can be derived the equation to determine flow angle (such
as alpha-2, beta-2 and beta-3. The derivation of flow angle equations can be started by defining other
parameters, namely flow coefficient, 𝜙 and work coefficient, 𝜓 as follows:
𝐶𝑥
𝜙= 𝑈
(1)
Δℎ0
𝜓= 𝑈2
(2)
From equations 1 and 2, by performing some mathematical manipulation, the relation between velocity
and flow angles of the blade can be obtained as follows:
𝑈
𝐶𝑎
= 𝑡𝑎𝑛𝛼2 − 𝑡𝑎𝑛𝛽2 (3)
𝑈
= 𝑡𝑎𝑛𝛽3 − 𝑡𝑎𝑛𝛼3 (4)
𝐶𝑎
And then the equations of the flow angles 𝛼2, 𝛽2 and 𝛽3 can be obtained as follows:
The axial velocity variable, 𝐶𝑎 in Equations 3 and 4 can be derived by taken constant nozzle angle
shown in the following equations:
4
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
𝐶𝑎2
= 𝑐𝑜𝑡𝛼2 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 (9)
𝐶𝑤2
2
𝑑𝐶𝑤2 𝑑𝐶𝑤2 𝐶𝑤2
𝐶𝑤2 𝑐𝑜𝑡 2 𝛼2 𝑑𝑟
+ 𝐶𝑤2 𝑑𝑟
+ 𝑟
𝑑𝐶𝑎2 𝑑𝐶
𝑑𝑟
= 𝑑𝑟𝑤2 𝑐𝑜𝑡𝛼2 (10)
𝑑𝐶 𝐶
(1 + 𝑐𝑜𝑡 2 𝛼2 ) 𝑤2 + 𝑤2 = 0 (11)
𝑑𝑟 𝑟
𝑑𝐶𝑤2 2 𝑑𝑟
𝐶𝑤2
= −𝑠𝑖𝑛 𝛼2 𝑟 (12)
2
𝐶𝑤2 𝑟 𝑠𝑖𝑛 𝛼2 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 (13)
2
𝐶𝑎2 𝑟 𝑠𝑖𝑛 𝛼2 = 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 (14)
2
𝑟 𝑠𝑖𝑛 𝛼2
𝐶𝑎2 𝑟 = 𝐶𝑎2 𝑚 ( 𝑟𝑚 )
𝐶𝑎2 𝑚 = 𝐶𝑎3 𝑚 = 𝐶𝑤2 𝑚 𝑐𝑜𝑡 𝛼2 (15)
𝐶𝑝 Δ𝑇
𝐶𝑤2 𝑚 = (16)
𝑈𝑚
5
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Illustration of the stator (NGV) – rotor configuration both of with airfoil profiled NGV and flat plate
profiled NGV is depicted in Figure 4. In the flat plate profiled NGV configuration, the flat plate is
generated by following the main camber line of the airfoil that previously used, the flat plate thickness
that used in the configuration is 2 mm (will be discussed at the following chapter).
6
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Figure 5. Three-dimensional blade model (left) and full three dimensional designed turbine (right)
2.2.2 Numerical Model
The equation that used in the simulation is Reynolds Averaged Navier-Stokes with K-epsilon turbulence
model. The number of mesh that applied on the computational domain geometry is around 5 million to
6 million mesh. The type of the simulation is quasi-steady simulation. This simulation type uses relative
move of the flow in the domain to simulate the rotational movement of the rotor blade. The rotor is
remain at rest but the flow in the rotor domain move with certain rpm that inputted. This conditions is
known by applying frozen rotor interface.
2.2.3 Boundary Condition
The computational simulation starts with the generation of computational domain in which the
thermodynamic properties such as velocity, pressure and temperature are computed. The domain is
defined with some boundaries such as inlet, shroud, hub, and interfaces as shown in Figure 6. The
boundary conditions that applied in the pre-setting of the simulation steps are inlet (in this boundary
conditions use mass flow to be input value and total temperature), wall (use default setting of the wall
boundary conditions), and outlet (in this boundary use static pressure as input value). The place of the
boundary conditions that applied at the computational domain is depicted in Figure 6. In the simulation,
two type of the convergence criteria i.e., error convergence criteria and residual result convergence
criteria, were used to check the convergence. The final step of the simulation is running numerical
process to get convergence solution
Figure 6. Computational domain (left) and meshing (right) that used in this design study
7
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Figure 7. Error curve (left) and quantities of interest curve (right) from running process
8
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Figure 9. Effect of axial gap and turbine rotor clearance due to turbine power
The effect of tip clearance is examined further for 4 mm axial gap turbine case. Table 3 shows the
performance of the axial turbine with the axial gap of 4 mm for the tip clearances of 0.5 mm (left) and
1.0 mm (right), respectively. As can be seen from Table 3, the output power increases by reducing
clearance.
Table 3. Performance of two type designed turbine, turbine with 4 mm axial gap-0.5 mm rotor
clearance (left) and turbine with 4 mm axial gap-1 mm rotor clearance (right)
9
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
Total Temperature Outlet 1022 deg K Total Temperature Outlet 1023 deg K
Mass Flow 0.8139 Kg/s Mass Flow 0.8139 Kg/s
Power 160196 watt Power 155121 watt
4.2 Case 2: Turbine with Flat Plate Profile that applied at the Stator (NGV) Section
Considering manufacture cost and simplicity, the airfoil-profiled stator blade might possibly be
substituted by a curved flat plate profile which follow the camber line of the airfoil as shown in Figure
10. The curved plate axial turbine is then simulated with the axial gap of 4 mm for two different
clearances, 0.5 mm and 1 mm. The thickness of the plate is set to 2 mm.
Figure 10. Stator profile using airfoil profile (left) and using flat plate (right)
The performances of the designed turbine with curved plate stator blade are presented in Table 4. Based
on the simulation result between two kind of turbine models (turbine with NGV blade is airfoil profiled
and turbine model with flat plat profiled), the output power of turbine model with flat plate profiled
NGV is decreasing around 3 % from the previous NGV model. Although the flat plate NGV model give
lower output power but the reduction is not significant, with simple NGV geometry turbine with flat
plate NGV is still worth to be developed.
Table 4. Performance of turbine with flat plate stator-rotor clearance 0.5 mm (left) and rotor clearance
1 mm (right)
4.3 The Effect of Stagger Angle and Flat Plate Thickness to Turbine’s Output Power
4.3.1 The Effect of Turbine Stagger Angle to Turbine Output Power
The influence of stagger angle on turbine output power is depicted in Figure 11. The simulation results
show the turbine output power are lower than 155 KW for the turbine models with clearance 1.5% and
2%. The turbine power output can be increased by setting up stagger angle of the blades to the right
position. As shown from the simulation, the higher angle of stagger gives more power to the turbine.
10
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
However, the increase in stagger angle that produces higher power output required higher inlet pressure
and decreases turbine efficiency also as shown in Figure 11.
Figure 11. Effect of Stagger angle due to turbine efficiency (left) and turbine power (right)
4.3.2 Flat Plate Thickness Effect
The change of turbine power output due to the thickness of the flat plate is shown in Figure 12. In this
study, the flat plate thickness is varied as 1 mm, 2 mm and 3 mm. From simulation results, the maximum
turbine power output is obtained for the flat plate thickness of 2 mm. The layout of the turbine power
output for various thickness is also depicted in Figure 12.
Figure 12. Three models of designed turbine (left) and effect of flat plate due to turbine power (right)
5. Conclusion
In the design of axial turbine there are several things that affect output power of the turbine such as
stagger angle, axial gap, clearance, and the profile of the airfoil (both the stator and rotor profile). Based
on the simulation results there is a certain distance of axial gap that provide optimal power, and the
output power of the designed turbine is tend to decrease with increasing clearance (distance between
rotor blade and shroud wall). The design model that closest with design point (the design point are :
turbine that can produce 155 KW in power wiht 0.8139 kg/s mass flow at 80000 RPM) is the turbine
model with 20% axial gap, 1 mm clearance between tip rotor blade and shroud wall and still use airfoil
11
5th International Seminar of Aerospace Science and Technology IOP Publishing
IOP Conf. Series: Journal of Physics: Conf. Series 1005 (2018)
1234567890 ‘’“” 012026 doi:10.1088/1742-6596/1005/1/012026
profile at the NGV section. The flat plate NGV model also can achieve the design point with change the
clearance from 1 mm to 0.5 mm or may be can also change the stagger angle of the NGV’s blade.
References
[1] D.G. Ainley and G.C.R. Mathieson, A Method of Performance Estimation for AxialFlow
Turbines, Aeronautical Research Council Reports and Memoranda, 1952
[2] Donald E. Holeski and Samuel M. Futral, Effect of Rotor Tip Clearance on the Performance of
A 5-Inch Single Stage Axial Flow Turbine, National Aeronautics and Space Administration, 1969
[3] James Allen Messegee, Influence of Axial and Radial Clearance on the Performance of a
Turbine Stage with Blunt Edge Non-Twisted Blades,Dudley Knox Library, 1967,
[4] Lewis, Turbomachinery Performance Analysis, Elsevier Science & Technology, 1996
[5] H. Cohen, Gas Turbine Theory, Longman Group Limited, 1996
[6] Jack D. Mattingly, Elements of Propulsion: Gas Turbines and Rockets, AIAA Education Series,
1989
[7] Anderson Jr., John D., Fundamental of Aerodynamics, McGraw-Hill Series
[8] Hartono, Firman, Aero Propulsion Presentation Chapter Axial Turbine
[9] Hartono, Firman, Aircraft Propulsion Chapter Parametric Cycle Analysis
[10] Hartono, Firman, Aircraft Propulsion Chapter Turbomachinery
12