A CFD Study of Diffuser Performance of Gas Dynamic Laser

System
S D Ravi1, M A Sriram2, P S Kulkarni3 and N K S Rajan4
1,2,3,4

Department of Aerospace Engineering, Indian Institute of Science, Bangalore

Abstract: Based on the an earlier CFD analysis of the performance of the

gas-dynamically controlled laser cavity it was found that there is a possibility
of optimizing the geometry of the diffuser that can bring about reductions in
both size and cost of the system by examining the critical dimensional
requirements of the diffuser for two cases i.e. case 1 is 1kg/sec and case 2 is
15kg/sec of mass flow rate. Consequently, an extensive CFD analysis has
been carried out for a range of diffuser configurations by simulating the
supersonic flow through the arrangement including the laser cavity driven by
a bank of converging diverging nozzles and the diffuser. The numerical
investigations with 3D-RANS code are carried out to capture the flow patterns
through diffusers past the cavity that has multiple supersonic jet interactions
with shocks leading to complex flow pattern. Varying length of the diffuser
plates is made to be the basic parameter of the study. The analysis reveals
that the pressure recovery pattern during the flow through the diffuser from
the simulation, being critical for the performance of the laser device shows its
dependence on the diffuser length is weaker beyond a critical lower limit and
this evaluation of this limit would provide a design guideline for a more
efficient system configuration.
The observation based on the parametric study shows that the
pressure recovery transients in the near vicinity of the cavity is not affected
for the reduction in the length of the diffuser plates up to its 10% of the initial
size, indicating the design in the first configuration that was tested
experimentally has a large factor of margin. The flow stability in the laser
cavity is found to be unaffected since a strong and stable shock is located at
the leading edge of the diffuser plates while the downstream shock and flow
patterns are changed, as one would expect. Results of the study for the
different lengths of diffusers in the range of 10% to its full length are
presented, keeping the experimentally tested configuration used in the earlier
study [1] as the reference length. The conclusions drawn from the analysis is
found to be of significance since it provides new design considerations based
on the understanding of the intricacies of the flow, allowing for a hardware
optimization that can lead to substantial size reduction of the device with no
loss of performance.
Key words: Nozzles, Gas Diffusers, Numerical Grid Generation, CFD,
Turbulence
Email addresses: [email protected] (S D Ravi), [email protected] (M A Sriram)
[email protected], (P S Kulkarni), [email protected] (N K S Rajan)
1,4
Corresponding author

Proceedings of ICEAE 2009

Performance of Gas Dynamic Laser System

INTRODUCTION
Computational studies have been carried out for a
two design configuration systems involving a parametric
study of Gas Dynamics Laser (GDL) using a 3D-RANS
code. Parametric study is done based on the configurations
selected for the analysis from practical system which
involves aerodynamically controlled laser cavity which is
further used for generation of high powered laser beams.
The parametric study in this arrangement, in order to
produce an active lasing medium, rapid and deep cooling
of mixture of gases from a combustion chamber (CO2-N2H2O) is achieved by allowing the gases to expand in a set
of convergent - divergent nozzles, allowing the supersonic
jets to drive the cavity used as the zone for the laser
activity. This cavity would have a distribution of CO2 in
non-equilibrium condition with many quantum levels,
providing a favorable condition for the lasing activity. The
bank of nozzles followed by the cavity and further by a
diffuser section for stable operation, makes the main parts
of the aerodynamically controlled cavity which is
considered in the present analysis and a 3D view of the
configuration has been shown in Fig.1.
The simulation of flow in a Gas Dynamics Laser
system is done using an industrially standard CFD tool
3D-RANS code. The process of the modeling is taken up
in steps of increasing complexity to ensure the numerical
stability and computational consistency. The different
stages have provided clarity on different parameters that
include the optimal grid distribution and the optimal initial
flow profiles to obtain a stable and consistent solution.
The details of the analysis are explained below.
An approach with multiple geometrical blocks
that are matching to the geometry of the unit considered at
different locations, depending on its complexity, is used
with their matching surfaces blended so as to generate a
good computational mesh. The structured mesh is
preferred since the geometry is more complex and the flow
computations need to capture supersonic, transonic and
subsonic zones over the same computational domain.
Optimization of number of computational nodes is done
by allowing the cluster of grids in critical zones, at places
expected to have larger gradients in the flow parameters
and in places where the geometry is more complex. It is
observed that nearly 3.6 to 5.6 lakh computational nodes
are required for 10% of initial length to full length of the
case-1 design configuration and 6.28 to 11.46 lakh
computational nodes are required for 10% of initial length
to full length of the case-2 design configuration. Fig. 2
gives view of a sample of the computational mesh
generated near in cavity at the upstream of the diffuser.
The computational flow Analysis made during
the preliminary stage of the study includes flow analysis
through a single nozzle followed by flow analysis through
a set of four nozzles and thereafter analysis of flow
through of a set of nozzles along with a set of two diffuser
vanes. In its final stage the computations were taken up to
capture the flow through multiple nozzles and the
diffusers. The method of analysis, the results obtained at
different stages been presented and discussed in the
following sections of the paper.

Proceedings of ICEAE 2009

THE CONFIGURATION AND THE APPROACH

The basic configuration of the system considered
for the analysis, scheme of computation, code validation
with experimental data and other basic aspects of the study
have been brought out in detail in an earlier work [1], and
this work is an extension of this reported work. However,
for clarity, the arrangement of the system configuration is
shown in Fig. 1. In the present analysis, all the other
hardware elements except the diffuser plates are kept
unchanged, in order to have a focused parametric study.
Further for simplicity of analysis, the downstream element
of the diffuser the acoustic control device is taken off
from computation domain.
A 3D-RANS code having upwinding implicit
scheme and k approach for turbulence is used for the
numerical solution. The Reynolds-Averaged NavierStokes Equations are solved for steady, compressible
viscous flow. The governing equations used are the
conventional standard sets that include:
U j

Continuity equation:

x j

=0

Momentum equation:

( Ui )+ x UiU j = xP x ij + uiuj
t
j
i
j

Energy equation:

h +
U j h =
Q j + u j h
t
x j
x j

( )

The boundary conditions and initial conditions

used include (a) no slip, impermeable and adiabatic walls;
(b) inlet pressure and temperature of the fluid entering the
system to be an assigned boundary value (typical outlet of
a rocket combustion chamber with storable liquid
propellants) and (c) the outlet from the diffuser set to
match the ambient conditions. Diminishing residual
criteria of the variables is used for the convergence with a
limit of RMS residuals falling below 10-4.
RESULTS AND DISCUSSION
Results from the study given in the Figs. 3 and 4
shows the Mach plot for single and set of four nozzle. It
can be seen that the details of a choked flow is obtained
with the transonic transients captured well, a first exercise
carried out for validating the approach. Figs. 5 (a), (b) and
(c) belongs to a case-1 (with 1 kg/s flow rate) and Figs. 6
(a), (b) and (c) belongs to a case-2 (with 15 kg/s flow rate)
provide the results of the studies made for the established
design configurations with the diffuser, specified to have
built for the experimental tests and based on empirical
designs. The different plots show the distributions of
important aerodynamic parameters Mach number,
Pressure and velocity (as streamline plot) of the system. It
can be observed that the reflected shocks within the
diffuser plates are adequately strong enough that the
pressure recovery occurs dominantly beyond the trailing
edge of the diffuser plates, ensuring a safe operating
pressure in the aerodynamic lasing cavity, eliminating the
chances of the back pressure affecting the flow in the
cavity. The predate pressure in the aerodynamic cavity is

S D Ravi, M A Sriram, P S Kulkarni, N K S Rajan

found to be close experimentally reported values of the

hardware built elsewhere.
Having observed that multiple shock reflections
are found within the diffusers, it was inferred that the plate
lengths could be brought down with no loss in the
performance as regards to the aerodynamic cavity is
concerned.
A more detailed study was taken up to
investigate the possibility of the extent of reduction of the
diffuser plate lengths that could be a significant
contribution to reduce the size and weights of the
integrated system.
The results shown in Figs. 5 (d) to (i) and Figs. 6
(d) to (i) show the behaviors of the case-1 and case-2 of
the corresponding aerodynamic characteristics with sets of
selective diffusers with reduced plate lengths of 80% ,
60% and 30%. One can observe that only the downstream
flow structure is affected as compared to the initial
configuration and flow in the laser cavity is not affected.
This and above data sets are run for the identical inlet
conditions (28atm chamber pressure, upstream of the
nozzle bank) of the reference design set. From the stabile
flow in the cavity zone in all the cases, it could be clearly
made out that the length of the diffuser plates can be
significantly reduced. Though an alternate arguments
could be made that the built system can handle much
larger range for flow rates since the diffuser are over
designed since the requirements of the cavity conditions
demand a fairly narrow range of working pressures for a
favourable laser activity, the flow rates permissible in the
system is limited to correspondingly narrow range. In this
point of view, one can infer that the device could be
optimized by reducing the diffuser plate lengths as can be
seen by the results.
To make a sensitivity analysis of the diffuser
performance based on its dependence on the inlet pressure,
a set of analysis was made taking the case of a 10% plate
length being the most critical among other cases. A
variation in the inlet pressure of about 7.5% was chosen
for this study. It was found that the higher side of the
pressure did not change the flow pattern categorically,
while the lower pressure provided a significantly unstable
flow structure with the solution strongly oscillating. Figs.
5 (j), (k) and Figs. 6 (j), (k) show the both case studies
having stable flow structure that belong to the standard
pressure.
Results shown in Figs. 5 (l), (m) and
Figs. 6 (l), (m) belong to a case with reduced inlet
pressure. One can see that the flow structure is not fully
formed and the solution is found to be strongly unstable
for both the two cases. It was also found that in this
condition of reduced pressure, a steady state flow solution
could not be reached. This behavior is visibly shown up
clearly in the two plots shown in Figs. 7 and 8 that show
the history of RMS residuals of the computational
parameters with an inference that with the reduced
pressure the solution has been quite oscillating and is
instable. It was found that further reduction in the inlet
pressure did not proceed numerically, leading to an
inference that with the transients of the flow instability
being strong enough not allowing for a steady state

solution. These aspects and observation further lead to

infer the existence of the operating limit of the inlet
pressure in the device. The analysis makes it clear that for
a specified geometry that the working range for the device
is associated with the length of the diffuser plate.
A consolidated plot of pressure recovery and
Mach numbers over the diffuser plates for both the case-1
and case-2 are provided in Figs. 9 to 12 and it shows that
for a specified working pressure, it is possible to arrive at
an optimally lower length of diffuser plate than what was
used in the empirical designs based on the trends of the
pressure recovery transients, in the neardownstream of
the cavity that are not disturbed by reduction in the plate
lengths up to about 10% of its original length. This leads
to a consideration that the design of the device can have a
reduction in the length of the diffuser and the sizes of its
associated hardware significantly, that could lead to a
major advantage in the usability and cost of the device.
The outcome of the analysis also signifies the extent of
analysis that CFD tools have given with an insight of the
flow behavior and to provide good guidelines for the
design.
CONCLUSION
The CFD Analysis using ICEMCFD and
ANSYSCFX for the complex geometry considered for an
aerodynamically driven laser cavity and capturing of the
flow details have shown a distinct insight of the physical
behavior of the configuration studied. The observations
based on the parametric study show that the pressure
recovery transients in the near vicinity of the cavity, the
most functionally critical part of the device, is not affected
with a reduction in the length of the diffuser plates up to
10% of the empirical design size. A validation of the
approach is made by checking the cavity pressure close to
experimental values and other flow parameters agreeing
broadly on the their behavioral patterns. The analysis
makes it clear that the configuration based on the
empirical design has a large factor of margin that can be
reduced significantly and that there exists a limit in the
operating range for the device. The conclusions drawn
from the analysis is expected to be significant since it
provides new design considerations, allowing for a
hardware optimization with substantial size reduction in
the device with no loss of performance.
REFERENCE
1. Complex Flow Analysis through a Multiple Nozzle
Driven Laser Cavity M.A.Sriram, N. K. S. Rajan
and P. S. Kulkarni Symposium on Applied
Aerodynamics and Design of Aerospace Vehicle
(SAROD-2007)
November
22-23,
2007,
Thiruvananthapuram, India.
2. Gas Dynamic Lasers; an Introduction by John F
Anderson.
3. David A. Russell Dept. of Aeronautics and
Astronautics, Univ. of Washington, First of the highpower machines, the gasdynamic laser still has a

Proceedings of ICEAE 2009

Performance of Gas Dynamic Laser System

dazzling performance capability; but future

development will most likely come in hard-won
incremental steps,
4. Compressible flow through a converging-diverging
Nozzle by S. J. Phillipson.

Fig. 3, Mach Contour (single nozzle)

Fig. 1, A general view of the initial configuration

Fig. 2, Computational Mesh Model

Fig. 5(a), Mach Contours (case1, full length)

Fig. 5(b), Pressure Contours (case1, full length)

Fig. 5(c), Streamlined Plot (case1, full length)

Proceedings of ICEAE 2009

Fig. 4, Mach Contour (set of four nozzle)

Fig. 6(a), Mach Contours (case2, full length)

Fig. 6(b), Pressure Contours (case2, full length)

Fig. 6(c), Streamlined Plot (case2, full length)

S D Ravi, M A Sriram, P S Kulkarni, N K S Rajan

Fig. 5(d), Mach Contours (case1, 80% Plate length)

Fig. 5(e), Pressure Contours (case1, 80% Plate length)

Fig. 5(f), Mach Contours (case1, 60% Plate length)

Fig. 5(g), Pressure Contours (case1, 60% Plate length)

Fig. 5(h), Mach Contours (case1, 30% Plate length)

Fig. 5(i), Pressure Contours (case1, 30% Plate length)

Fig. 6(d), Mach Contours (case2, 80% Plate length)

Fig. 6(e), Pressure Contours (case2, 80% Plate length)

Fig. 6(f), Mach Contours (case2, 60% Plate length)

Fig. 6(g), Pressure Contours (case2, 60% Plate length)

Fig. 6(h), Mach Contours (case2, 30% Plate length)

Fig. 6(i), Pressure Contours (case2, 30% Plate length)

Proceedings of ICEAE 2009

Performance of Gas Dynamic Laser System

Fig. 6(j), Mach Contours (case2, 10% Plate length)

Fig. 5(k), Pressure Contours (case1, 10% Plate length)

Fig. 6(k), Pressure Contours (case2, 10% Plate length)

Fig. 5(l), Mach Contours (case1, with 7.5% lower pr)

Fig. 6(l), Mach Contours (case2, with 7.5% lower pr)

Fig. 5(j), Mach Contours (case1, 10% Plate length)

Fig. 5(m), Pressure Contours (case1, with 7.5% lower pr)

Fig. 6(m), Pressure Contours (case2, with 7.5% lower pr)

Fig. 7, RMS Residual History (std pressure)

Proceedings of ICEAE 2009

Fig. 8, RMS Residual History (Lower pressure)

S D Ravi, M A Sriram, P S Kulkarni, N K S Rajan

Fig. 9, Consolidated plot of Pressure Recovery over the

diffuser plates of different length (Case1 study)

Fig. 12, Mach number over the diffuser plates of different

length (Case 2 study)

S D Ravi has post graduate education at

Visvesvaraya Technological University and
has been working as a project assistant at
CGPL, Dept of Aerospace Engineering
Indian Institute of Science. His current
interests are R&D, CFD, combustion, Gasification, and
energy.

Fig. 10, Consolidated plot of Pressure Recovery over the

diffuser plates of different length (Case 2 study)

M A Sriram is doing post graduate

education at KTH, Royal institute, Sweden.
and had been working as a project assistant at
CGPL, Dept of Aerospace Engineering
Indian Institute of Science. His current
interests are R&D, CFD, propulsion and combustion.

P S Kulkarni has Ph.D. in Faculty of

Engineering (Aerospace engineering), Indian
Institute of Science and has been working as
a Principal Research Scientist at Dept of
Aerospace Engineering Indian Institute of
Science. His current interests are Advanced Computational
Fluid Dynamics and propulsion.

Fig. 11, Mach number over the diffuser plates of different

length (Case 1 study)

N K S Rajan has Ph.D. in Faculty of

Engineering (Aerospace engineering), Indian
Institute of Science and has been working as
a Principal Research Scientist at Dept of
Aerospace Engineering, Indian Institute of
Science. His current interests are Heat Transfer, Fluid
Dynamics, combustion, Gasification, propulsion, energy,
Development of Prototype, Development of algorithm,
Computer Networking and Instrumentation.

Proceedings of ICEAE 2009