Computational Estimation of Flow through the C-D Supersonic Nozzle and Impulse Turbine Using CFD

25 International Journal for Modern Trends in Science and Technology
International Journal for Modern Trends in Science and Technology
Volume: 02, Issue No: 11, November 2016
http://www.ijmtst.com
ISSN: 2455-3778
Computational Estimation of Flow through the C-D
Supersonic Nozzle and Impulse Turbine Using CFD
Snigdha Panday1
| B. Joga Rao2
| Dr.T.Jayanand Kumar3
1PG Scholar, Department of Mechanical Engineering, GIET (A), Rajahmundry, Andhra Pradesh, India.
2Assistant Professor, Department of Mechanical Engineering,, GIET (A), Rajahmundry, Andhra Pradesh, India.
3Professor & Head, Department of Mechanical Engineering,, GIET (A), Rajahmundry, Andhra Pradesh, India.
To Cite this Article
Snigdha Panday, B. Joga Rao, T.Jayanand Kumar, Computational Estimation of Flow through the C-D Supersonic Nozzle
and Impulse Turbine Using CFD, International Journal for Modern Trends in Science and Technology, Vol. 02, Issue 11,
2016, pp. 24-32.
In this paper, CFD analysis of flow within, Convergent – Divergent rectangular super sonic nozzle and
super sonic impulse turbine with partial admission have been performed. The analysis has been performed
according to shape of a super sonic nozzle and length of axial clearance and the objective is to investigate the
effect of nozzle-rotor interaction on turbine’s performance. It is found that nozzle-rotor interaction losses are
largely dependent on axial clearance, which affects the flow within nozzle and the extent of flow expansion.
Therefore selecting appropriate length of axial clearance can decrease nozzle-rotor interaction losses. The
work is carried in two stages:1) Modeling and analysis of flow for rectangular convergent divergent super
sonic nozzle. 2) Prediction of optimal axial gap between the nozzle and rotor blades by allowing the above
nozzle flow.
In the present work, using a finite volume commercial code, ANSYS FLUENT 14.5, carries out flow through
the convergent divergent nozzle study. The nozzle geometry is modeled and grid is generated using
ANSYS14.5 Software. Computational results are in good agreement with the experimental ones.
KEYWORDS: Convergent – Divergent Nozzle,Nozzle-rotor interactions , Optimal axial clearence, Modeling
using ANSYS FLUENT 14.5
Copyright © 2016 International Journal for Modern Trends in Science and Technology
All rights reserved.
The work is carried in two stages:
1. Modeling and analysis of flow for
rectangular convergent divergent
supersonic nozzle.
2. Prediction of optimal axial gap between the
nozzle and rotor blades by allowing the
above nozzle flow.
In the present work, using a finite volume
commercial code, ANSYS FLUENT 14.5, carries out
flow through the convergent divergent nozzle
study. The nozzle geometry is modeled and grid is
generated using ANSYS14.5 Software.
Computational results are in good agreement with
the experimental ones.
The objective of present work is
1. Modeling and meshing of nozzle geometry.
2. Validate the CFD results of nozzle flow both
theoretically and experimentally.
3. Modeling and meshing of nozzle and
turbine blades is considered as a case of
partial admission type.
4. Validate the CFD results of nozzle and
turbine blades with experimental data.
ABSTRACT

Snigdha Panday, B. Joga Rao, T.Jayanand Kumar : Computational Estimation of Flow through the C-D Supersonic
Nozzle and Impulse Turbine Using CFD
This thesis aims to predict the following:
 Estimation of velocity at nozzle exit as a
case of supersonic (or) not.
 Estimation of nozzle and turbine rotor gap
under static condition of rotor.
 Flow visualization.
I. FLUID GOVERNING EQUATION
The fundamental equations of fluid
dynamics are based on the following universal law
of conversation. They are:
1) Conservation of mass.
2) Conservation of momentum.
3) Conservation of energy.
Continuity Equation:
where „‟ is the density, „V‟ is the fluid velocity. For
an incompressible flow, the density of each fluid
element remains constant.
Momentum Equation:
Energy Equation:
Equation of State:
(Pv= MRT)
II. METHODS OF PREDICTION
Experimental Investigation:
The most reliable information about the physical
process is often given by actual measurement. It
involves full scale equipment, which can be used to
predict how identical copies of the equipment
would perform under the same conditions. They
are often expensive and time consuming. The
resulting information however must be
extrapolated to full scale and general rules for
doing this are often available.
Theoretical Calculations:
It works out on the consequences of
mathematical model rather than those of an actual
physical model. If the methods of classical
mathematics are to be used for solving these
equations then there is a little hope for predicting
many phenomenon of practical interest. It is often
referred as an analytical approach. In this
simplification, assumptions are used in order to
make the problem tractable. These solutions often
contain infinite series, special functions and
transcendental equations for Eigen values so that
the numerical evaluation may be the formidable
task.
Computational Methods:
In the computational approach a limited
number of assumptions were made and high speed
digital computer is used to solve the resulting
governing fluid dynamic equations. The
development of numerical methods and the
availability of large digital computers hold the
promise implicating the mathematical model which
can be worked out on almost any practical
problem.
Advantages:
a) No restriction to linearity
b) Complicated physics can be treated.
c) Time evolution of flow can be obtained.
d) It has potential of providing the information
which cannot be obtained by any other means.
e) Computational can provide investigations which
can be performed with remarkable speed. Designer
can study the implications of hundreds of
differential configurations in minimum time and
choose the optimum design.
f) It gives detailed and complete information. It can
also provide the values of all the relevant variables
such as pressure, velocity, temperature,
concentration, turbulence etc throughout the
domain of interest.
Disadvantages:
i) Truncation errors
ii) Boundary condition problems
iii) Computer costs.
iv) Computer storage and speed.

Applications of CFD:
a) Aerodynamics of aircraft & vehicles: lift
drag
b) Hydrodynamics of ship
c) Metrology: weather prediction
d) Turbo machinery: Flows inside rotating
passage, diffusers.
e) Hydrology and oceanography: flows in
rivers, estuaries and oceans
Problem solving steps: Once you have determined
the important features of the problem we want to
solve, we will follow the basic procedural steps
shown below:
1.Create the model geometry and grid.
2.Start the appropriate solver for 2d or 3d
modeling.
3.Import the grid.
4.Check the grid
5.Scale the grid
6.Select the solver formulation.
7.Choose basic equations to be solved: laminar or
turbulent (or inviscid), chemical species or
reaction, heat transfer models etc. Identify
additional models needed: fans, heat
exchangers, porous media etc.
8.Specify the material properties.
9.Specify the boundary conditions.
10. Adjust the solution control parameters.
11. Initialize the flow field.
12. Calculate the solution.
13. Examine the results.
14. Save the results.
III. MODELING OF THE COMPONENTS
Modeling of Super Sonic Nozzle:
Figure: Nozzle Profile
Figure: Solid model of nozzle profile.
Modeling of Blade Profile:
Figure: Blade Profile.
Figure: Series of Rotor Blade.
Modeling of Nozzle and Turbine Blades as
Partial Admission Case:
As analysis is to be carried out for different axial
gap between rotor blades and nozzle, length of
ordinate is fixed. Rotor blades and nozzle with
different axial gaps are modeled separately as
shown in figures for 3mm, 4mm and 5mm axial
gaps respectively.
Figure: Solid model as a partial admission case of Nozzle
and Rotor blades with 3mm axial gap.

Mesh generation:
Mesh Generation for 3D C-D nozzle:
A 3D Model is created in ANSYS FLUENT 14.5 as
shown in figure. To generate the structured grid as
shown in figure with hexahedral cells ANSYS
FLUENT 14.5 is used. This mesh is in good
agreement with turbulence model and good results
are obtained otherwise if they aren‟t in good
agreement then they won‟t match with the
experimental ones.
Figure: Meshed View of 3 D Nozzle Profile.
Mesh Generation for 2D C-D nozzle:
A 2D nozzle profile is modeled in ANSYS FLUENT
14.5. To generate the structured grid as shown in
figure with quadrilateral elements ANSYS FLUENT
14.5 is used. This mesh is in good agreement with
turbulence model and good results are obtained
otherwise if they aren‟t in good agreement then
they won‟t match with the experimental one.
Figure: Meshed view of 2 D nozzle profile.
Figure: Meshed view as a partial admission case of Nozzle
ANALYSIS OF C-D RECTANGULAR NOZZLE:
Solver: Material selection and operating
condition defining:
The solver is defined first. Solver is taken as
Coupled based and formulation as implicit, space
as 2D and time as steady. Velocity formulation as
absolute and gradient options as Green Gauss Cell
based are taken. Energy equation is taken into
consideration. The viscous medium is also taken.
Then analysis is carried out using K-epsilon
turbulence model. The selection of material is
done. Material selected as gas. The properties of
gas taken as follows:
Density of ideal gas is considered
Cp (Specific heat capacity) = 2034.6J/Kg.K
Thermal Conductivity = 0.0706 W/m-K
Viscosity = 6.07 e-5 (Kg/m-s)
Molecular weight = 23.05 (Kg/Kg-Mol)
The analysis is carried out under operating
condition of Zero
Pascal. Gravity is not taken into consideration.
BOUNDARY CONDITIONS:
Nozzle Inlet:
Pressure inlet is taken as inlet for nozzle. The
value ofpressure is 8101325 Pascal. Initial gauge
pressure is taken as 7898681 Pascal. Temperature
is taken as 1583K.
Nozzle Outlet:
The nozzle outlet is set as pressure outlet with a
value of 13e5.
Controls set up:
The solution controls are set as listed below. The
under relaxation factor was set as given.
Turbulence Kinetic Energy 0.8
Turbulence Dissipation rate 0.8
Turbulence Viscocity 1
Discretization Equation is selected as given:
Flow (Second order up wind)
Turbulence Kinetic Energy (1st order upwind)
Turbulence dissipations rate (1st order upwind)
Initialization:
Solution initialization is done. Initial values of
velocity are taken as 186.3 m/s in y direction.
Temperature is taken as 1583K Residual

monitoring is done and convergence criteria are set
up. The convergence criteria of various parameters
are listed below.
Continuity - 0.001
X Velocity - 0.001
Y Velocity - 0.001
Energy - 0.001
The number of iterations is set up and iterations
starts. The iteration continues till the convergence
is reached and convergence history as shown in
figure.
ANALYSIS OF NOZZLE AND TURBINE ROTOR
BLADES AS A CASE OF PARTIAL ADMISSION:
The analysis is carried in fluent software by
importing the meshed file saved in ANSYS FLUENT
14.5. The steps that are followed are given below
which all the conditions and the boundaries value
for the problem statement are included, for varied
axial gaps of nozzle and turbine rotor blades as a
case of partial admission.
Figure: Convergence history of C-D Nozzle.
Checking of Mesh and Scaling:
The fluent solver is opened where 2DDP is
selected and then the importing of the meshed file
is done. The meshed file then undergoes a checking
where number of grids are found. After this grid
check is done followed by smoothing and swapping
of grid is done.
Following this step, scaling is done. Scale is
scaled to mm. Grid created then converted from m
to mm. After this step, defining of various
parameters is done.
Solver: Material Selection and operating
condition defining:
The solver is defined first. Solver is taken as
Segregated based and formulation as implicit,
space as 2D and time as steady.
Velocity formulation as absolute and gradient
options as Green Gauss Cell based are taken.
Energy equation is taken into consideration. The
viscous medium is also taken. Then analysis is
carried out using K-epsilon turbulence model. The
selection of material is done. Material selected as
gas.
The properties of gas taken as follows:
Density of ideal gas is considered
Cp (Specific heat capacity) = 2034.6J/Kg.K
Thermal Conductivity = 0.0706 W/m-K
Viscosity = 6.07 e-5 (Kg/m-s)
Molecular weight = 23.05 (Kg/Kg-Mol)
The analysis is carried out under operating
condition of Zero Pascal. Gravity is not taken into
consideration.
BOUNDARY CONDITIONS:
Nozzle Inlet:
Pressure inlet is taken as inlet for nozzle. The value
of pressure is 8101325 Pascal. Initial gauge
pressure is taken as 7898681Pascal. Temperature
is taken as 1583K.
Outlet blades of rotor:
The outlet is set as pressure outlet with a value of
101325 Pascal
Controls set up:
The solution controls are set as listed below: The
under relaxation factor was set as given.
Pressure - 0.3
Density - 1
Body forces - 1
Momentum - 0.7
Pressure velocity coupling was taken as SIMPLE.
Discretization Equation is selected as given:
Pressure - standard
Density - 1st order upwind
Momentum - 1st order upwind
Turbulence Kinetic Energy (1st order upwind)
Turbulence dissipations rate (1st order upwind)
Energy - 1st order upwind
Initialization:
Solution initialization is done. Initial values of
velocity are taken as 186.3 m/s in y direction.
Temperature is taken as 1583K Residual
monitorization is done and convergence criteria are
set up. The convergence criteria of various
parameters are listed below.
Continuity - 0.001
X Velocity - 0.001
Y Velocity - 0.001
Energy - 0.001
The number of iterations is set up and iterations
starts.
RESULTS AND DISCUSSION
Theoretical calculation of c-d nozzle:
Properties of Gases:
Gamma of gases = 1.27
Thermal conductivity [K] = 0.0706 w/m-k

Molecular Weight = 23.05 Kg/Kg-Mol
Specific heat of gases [CP] = 2034.6 KJ/Kg-K
Density treated as Ideal gas
Gas constant (R) = 432
Viscosity (Kg/m-s) = 6.07 e -5
Boundary (Inlet/Exit) Conditions
NPR (Nozzle Pressure Ratio) : 6.16
Given Inlet pressure : 80 bar (8x106 pa)
Exit pressure : 12.987 bar
Inlet temperature : 1583K
Geometrical Parameters
Given Throat area : 2.85xe-5 m2
Given inlet Mach number : 0.2
Mach number definition : M = V/C
C = acoustic velocity
RESULTS FOR NOZZLE:
Figure: Convergence history of C-D nozzle
It had been observed that convergence history for
C-D nozzle has good argument and flow is
maintained smooth as shown in figure.
Figure: Co-efficient of drag for C-D nozzle
It had been observed that the co-efficient of drag.
Initially it fluctuates at inlet and later on its
remains stable while the flow continues, the
co-efficient of drag value maintains the limitations
and the results are good in agreement as shown is
figure.
Figure: Co-efficient of lift for C-D nozzle

It had been observed that the co-efficient of drag.
Initially it fluctuates at inlet and later on its
remains stable while the flow continues, the
co-efficient of drag value maintains the limitations
and the results are good in agreement as shown is
figure. As extent of results in normal direction is
nothing but 3D flow through the nozzle for given
input condition with velocity as 183 m/s and
maximum output was observed as a 1423 m/s,
such that Mach Number is increased from 0.2 to
1.54 in which nozzle is acting as supersonic nozzle
and contours of mach number as shown in figure.
The velocity contours of nozzle is plotted in figure,
the pressure contours of nozzle is plotted in figure
and temperature contours of nozzle is plotted in
figure. The velocity, temperature, Mach number,
pressure variation along the nozzle is compared
with theoretical calculation and with experimental
too. These three results are good in agreement with
each other.
Figure: Mach number contours of nozzle.
Figure: Velocity contours of nozzle.
Figure: Pressure contours of nozzle
Figure: Temperature contours of nozzle.
RESULTS FOR INTERACTION OF NOZZLE AND
ROTOR BLADES:
Flow passage from exit of nozzle and entry of
turbine rotor blades is allowed under static
condition of rotor (blades). As a case of partial
admission of axial impulse turbine the flow will
suddenly impact to the blades during the of nozzle
flow at rotor blades so many factors can be
considered to improve the performance of turbine.
In this study the tangential velocity is selected as a
parameter for better performance of turbine speed.
Velocity distribution is shown in figures for 3mm,
4mm and 5mm axial gap respectively. Velocity
vector contours are shown in figures 3mm, 4mm
and 5mm axial gap respectively. By observing the
tangential velocity contours for 3mm, 4mm, 5mm
gap axial clearances of nozzle and turbine rotor
blades as shown in figures. The maximum average
tangential velocity will act a 3mm axial clearence,
3mm gap of axial clearance will be the better one
and also assumed that co-efficient of drag and lift
of nozzle-rotor interaction, all are in good
agreement as shown in figures.
Figure: Velocity contours with 3mm axial gap between
Nozzle and Rotor blades.

Figure: Tangential velocity contours with 3mm axial gap
between Nozzle and Rotor blades.
Figure: Tangential velocity contours with 4mm axial gap
Figure:: Tangential velocity contours with 5mm axial gap
Figure: Co-efficient of drag with 3mm axial gap between
Figure: Co-efficient of lift with 3mm axial gap between
Figure: Co-efficient of lift with 4mm axial gap between
Nozzle and Rotor blades
Figure: Co-efficient of drag with 4mm axial gap between
Nozzle and Rotor blades
IV. LIMITATIONS AND SCOPE
 CD Nozzle flow is carried out for given
pressure ratio and input output conditions
as specified.
 The above nozzle used for this purpose is a
special type manufactured by N.S.T.L for
their special applications.
 During the study of nozzle and turbine the
flow of gases is assumed to be at its full
length and turbine blades are assumed
under static condition.
V. CONCLUSION & FUTURE SCOPE
For the present study, the following conclusions
are drawn:
 CFD results of convergent divergent nozzle
were in good agreement with the
experimental values and theoretical values
and the nozzle is acting as a super sonic
nozzle.
 CFD predictions for convergent – divergent
nozzle and turbine rotor find good
agreement with the experimental results of
N.S.T.L
Future work should focus on employing a finer
resolution grid than the one employed in present
study different turbulence models can be tested in
the simulations.
 Dynamic mesh can employ for further
simulations.
 Nozzle can be studied for different N.P.R to
meet the other range of knots.
 Further research can be done under
dynamic condition of rotor.
REFERENCES
[1] Bong-Gunshin, Young-Jae Kwak, Kui-soon Kim and
Jin-Han Kim, “A study on Nozzle – Rotor interaction
losses for geometric conditions of a partial admission
impulse turbine”. Proceedings of Asian joint
conference on propulsion and power 2006 Beijing,
China, April 20-23, 2006.

[2] A.A.Khan and T.R.Shem bharkar, “Viscous flow
analysis in a Convergent – Divergent
nozzle”. Proceedings of the international conference
on Aerospace science and Technology, Banglore,
India, June 26-28, 2008.
[3] H.K.Versteeg and W.Malala Sekhara, “An
introduction to Computational Fluid Dynamics”,
British library cataloguing pub, 4th edition, 1996.
[4] David C.Wil Cox, “Turbulence modeling for CFD”
Second Edition 1998.
[5] S. Majumdar and B.N.Rajani, “Grid generation for
Abritrary 3-D configuration using a Differential
Algebraic Hybrid Method, CTFD Division, NAL,
Bangalore, April 1995.
[6] Layton, W. Sahin and Volker.J, “A problem solving
Approach using Les for a backward facing – step”
2002.
[7] M.M.Atha vale and H.Q. Yang, “coupled field thermal
structural simulations in Micro Valves and Micro
channels” CFD research corporation.
[8] Lars Davidson, “An introduction to turbulence
Models”, Department of thermo and fluid dynamics,
Chalmers university of technology, Goteborg,
Sweden, November, 2003.
[9] Mas Mohas Rai, “Three-Dimensional Navier – Stokes
Simulations of turbine Rotor-Stator interaction,
NASA Ames Research Center, Moffett Field,
California Vol.5, No.3, May – June 1989.
[10]Kazuhiro Nakahashi, “Navier – Stokes Computations
of two and three dimensional cascade flow fields”,
Vol.5, No.3, May – June 1989.
[11]Philip C.E. Jorojenson and Rodrick V.China, “
Explicit Range–Kutta Method for unsteady Rotor
Stator Interaction”, Vol.6, No.6, June 1989.
[12]Md.Akhtaruzzaman Sarker and David G.Rhodes,
“3D free surface model of laboratory channel with
Rectangular Broad – crested weir”, proceedings
28th IAHR Congress, Graz, Austria, 1999.
[13]Leonid Moroz, Yuri Gororuschenko, Petr Pagur,
“Axial turbine stanges design: ID/2D/3D
simulation, Experimal, Optimization”, Proceedings
of GT 2005 ASME Turbo Expo 2005: Nerada USA.
[14]Nicholas J. Georgia dis, Teryn W.Dal Bellow, chorles
J.Tretry, and Albert L.Johns, “Aerodynamic Design
and analysis of High performance nozzles for Mach 4
Accebrator Vehicles”, 44th AIAA Aero Space Sciences
meeting and exhibit, 9-12 January 2006, Nerada.
[15]Eric Gamble and Dan Haid, “Improving off Design
nozzle performance using fluid injection” AIAA-2004
– 1206 USA.
[16]K.Sukati and Y.Shinizu, “Effect of the increase in the
entrance convergent section length of the gun nozzle
on the high velocity oxygen fuel and cold spray
process”, 24th October 2000 ASM international.
[17]E.V.Volpe, L.C. decastro sartos “Boundary and
internal conditions for adjoins fluid – flow problems”.
29 June 2007 Springer science Business Media B.V.
2008.
[18]B.Jodoin, “Cold Spray nozzle Mach number
limitation‟, February 14 2001, ASM international.

Computational Estimation of Flow through the C-D Supersonic Nozzle and Impulse Turbine Using CFD

More Related Content

What's hot (20)

Similar to Computational Estimation of Flow through the C-D Supersonic Nozzle and Impulse Turbine Using CFD (20)

Recently uploaded (20)

Computational Estimation of Flow through the C-D Supersonic Nozzle and Impulse Turbine Using CFD