Numerical Inspection of Vortex Generator With help of mid, cross Plane Temperature and Vector on Gas Turbine Blade

film Cooling

Effectiveness

Nirmal Halder

Department of Aerospace Engineering,

Sandip University,

Nashik, Maharashtra-42 20 12, India

ABSTRACT

Present numerical investigation proposes to mitigate the effects of Counter rotating vortex pair by employing vortex

generator. The common-flow-down arrangement of delta winglet pair is preferred for this investigation. The investigat ion

of suitable turbulence model has been studied. Numerical simulation has been carried out to investigate the effect of

placement of vortex generator on the characteristics of film cooling effectiveness. Vortex generator located at before film

hole or at upstream, at after film cooling hole or at downstream and both before, after Film cooling hole. Result indicates

that Vortex generator, located at up-stream of circular film cooling hole will yield better effectiveness. Reynolds number

based on free stream velocity and film cooling-hole dimension is kept at 17000. Moreover consequence, in variation of

density ratio (DR), secondary flow velocity or jet velocity (u j) and T.K.E on cooling effectiveness has been investigated.

Result shows that increment in u j and T.K.E responsible for decrement in cooling effectiveness. The results are compared

with baseline case. It has been observed that numerical investigation implementing FLUENT commercial code adopting

the k-omega SST turbulence model perform better.

1. INTRODUCTION

The intake temperature of a gas turbine determines its power output. Film cooling is utilized when input temperatures goes

above the melting point of gas turbine components. Gas turbine components are protected from high intake temperatures by

film cooling [1]. The development of the counter rotating vortex pair (CRVP) [2] is caused by the interaction of cold and hot

fluids. CRVP directs hot air to the turbine blade wall while also directing cool fluid out of the blade towards heated environments.

The result is, Counter rotating vortex pair CRVP reduces the efficiency of film cooling effectiveness, η (𝜂 =

(𝑇𝑎𝑤 − 𝑇𝑐𝑓 )⁄(𝑇𝑗 − 𝑇𝑐𝑓 )). Here the adiabatic wall temperature is Taw, while the main and secondary fluid temperatures are Tcf,

Tj respectively. The primary fluid bends the secondary flow. CRVP is formed as a result of this phenomena [3]. The longitudinal

vortex (LV) is formed by the angle of attack [4]. The pressure differential between the upstream and downstream sides of the
vortex generator (VG) produces LV. The vortex generator has a triangular and rectangular form that we can detect. According

to Jacobi and Shah [5], the winglet type VG arrangement is the most effective. By installing an upstream ramp with ramp angles

of 8.5, 15 and 24 degrees, Chen et al. [6] utilized an infrared imaging method to increase the film cooling performance. The

blowing ratios (M) were kept at 0.3, 0.4, 0.6 and 1.4 respectively. A greater ramp's angle along with a larger blowing ratio was

shown to enhance cooling efficiency. With an upstream ramp, Barigozzi et al. [7] used an infrared imaging method on cylindrical

and fan-shaped film cooling holes. It was discovered that when the blowing ratio was low, the efficiency of conventional circular

holes increased. An increase in adiabatic efficiency was found when Na and Shih [8] provided a backward facing upstream

ramp. Particle image velocimetry (PIV) and pressure sensitive paint (PSP) were used by Zhou and Hu [9] to improve cooling

efficiency by utilizing an upstream and downstream ramp in the shape of a Barchan dune. This innovative ramp design enhances

the efficiency of film cooling. Using a steady state pressure sensitive paint, Rallabandi et al. [10] studied the influence of an

upstream step for simple, complex, cylindrical cooling holes and cooling holes of fan-shape. The height of three steps and the

blowing ratio of four were maintained. In the compound angled cylindrical hole, there was a most effective film cooling, whereas

the basic fan-shaped hole offered the least efficient film cooling. Wang et al. [11] used a rectangular hole to do a numerical

inquiry with an upstream step. They tried with three different blowing ratios and five different step heights. Fakher et al. [12]

investigated the ramp, analytically and physically. The 0.5, 1.0 and 1.5 blowing ratios were retained. The findings indicated that

utilizing a double ramp was more successful than using a single ramp. Shinn and Vanka [13] used a downstream micro ramp

with a 1.5 blowing ratio and an 8000 Reynolds number to improve cooling efficiency. The use of a micro ramp design has been

shown to increase the efficiency of film cooling. It was discovered that when step height increases, the efficiency of film cooling

decreases. In a low-speed wind tunnel, many vortex generators were tested [14]. The findings of this investigation show that

the use of a vortex generator improves the efficiency of film cooling. Computational investigation was utilized by Yao et al. [15]

whereas Dhungel et al. [16], Kapadia et al. [17], Sargison et al. [18], and Sinha et al. [19] used experimental research. In a low-

speed wind tunnel, many vortex generators are located at the bottom surface of race vehicles [20]. Using a vortex generator,

Angele et al. [21] investigated turbulent boundary layer separation control. Experimental and numerical research were

conducted to examine the effects of vortex generators on the onset of flow instabilities, the trajectories, and characteristi cs of

the generated coherent counter-rotating vortices [22]. The efficiency of film cooling is improved by vortex generator [23]. The

current analysis is carried out using the exact geometry used by Halder et al. [24]. By generating secondary vortices, a vortex

generator improves film cooling [25]. Several experimental and numerical investigations are performed in film cooling [26 -30].

1.1 NOVELTY FOR ACADEMY AS WELL AS FOR INDUSTRY

Above mentioned literature survey indicates that small investigation 0f film cooling is achieved implementing vortex generator
 3
in common flow down (CFD) configuration. Very few investigation has shown effect of jet inlet velocity, Density ratio (DR),

turbulent kinetic energy (T. K. E) on film cooling efficiency. Common flow down configuration can be defined as lower dis tance

in span wise direction among VG at the leading edge location of VG compared to trailing edge of VG in same direction between

VG. When reverse happens then configuration is called common flow up (CFU) which is drawn for various shape of VG as

mentioned in figure 1. Very few study is accomplished with VG is located at different upstream and downstream position. Very

few study has been carried out for the density ratio on the enhancement of effectiveness for CFD configuration of VG with the

assist of centerline, span line effectiveness and vector distribution respectively. Less study presents interaction and deformation

zone with DWP CFD configuration of Turbulent kinetic energy and jet inlet velocity with valuable contribution of Mid plane

temperature and CRVP distribution. Deformation zone of primary and secondary flow presents interaction zone of cross and jet

flow. Less communication between primary and secondary flow presents inferior mixing resulting in low CRVP effect. Low CRVP

effect signifies higher effectiveness. The adverse effect of tab, slot, step or trench configuration in jet arrangement is lack of

strength for installation and performance in film cooling hole. Very few studies demonstrate about the mitigation of CRVP

implementing vortex generator. Current study deals with the mitigation of CRVP. CRVP is responsible for low effectiveness.

Present investigation mitigates CRVP by creating secondary vortices originated from the vortex generator. Secondary vortices

rotational tendency is opposite to CRVP. Current investigation has been performed to evaluate vortex generator (VG) in the

configuration of Delta winglet pair (DWP) & common flow down (CFD), for gas turbine blade film cooling, an external cooling

arrangement.

Figure 1-(i) presents Vortical structure in film cooling geometry [31]. Figure 1-(ii) observed Cross flow with film cooling hole and

generation of kidney vortices [32]. Figure 1-(iii) noticed Vortex structures due to interaction of cross flow with rectangular hole

and vortex generator. Figure 1-(iv) examined Schematic of (a) CFU and (b) CFD configuration of DWP (Delta winglet pair) vortex

generator.

Figure 1-(ii): Cross flow with film cooling hole and generation of
Figure 1-(i): Vortical structure in film cooling geometry [30].
kidney vortices [31].
 Figure 1-(iii): Vortex structures due to interaction of cross flow with rectangular hole and vortex generator.

(a) (b)

Figure 1-(iv): Schematic representing geometry of (a) CFU (common flow up) and (b) CFD (common flow down) configuration of DWP

(Delta winglet pair) vortex generator.

2. PROBLEM FORMULATION

In earlier investigation of same author, at IJHT (publishd) Consequence of, distance among film cooling hole and VG, increase

in VG and Reynolds number (Re), length of VG, height of VG, turbulent intensity (TI) has been talked about with Vector, cross

and mid-plane temperature, Centerline & span-line effectiveness, bottom wall effectiveness distribution respectively.

Also previously at Journal of power and energy (not publishd, just submitted) by identical author impact of, distance among the

hole for film cooling and VG, implementing multiple VG, length of VG has been observed. Title of the paper which is submitted

but not published in ‘Journal of power and energy’: Influence of Vortex Generator on Gas Turbine Blade.

Effect of distance between film cooling hole and vortex generator

Here ‘Effect of distance between film cooling hole and vortex generator’ is discussed with Centerline & span-line effectiveness,

with vector & streamline and CRVP distribution.

Effect of increase in the vortex generator

The above quantity is described with Bottom wall effectiveness and streamline distribution.
 5
Effect of vortex generator’s length

The above parameter is described with vector and streamline distribution.

However these effects in ‘Journal of power and energy’ are not included in present study.

In previous study of same author, at Proc. International Conference on Mechanical, Automotive and Mechatronics Engineering

(ICMAME 2023), 29-30 April 2023, Dubai, UAE. Title of the paper is ‘Impact of vortex generator on gas turbine blade cooling

application. Here Effect of, “distance between film cooling hole and VG”, “increase in VG and density ratio (DR)”, “length of VG

and location”, “height of VG and jet inlet velocity (Uj)”, “Reynolds number (Re) and blowing ratio (M)”, “turbulent intensity (TI)

and turbulent kinetic energy (T. K. E)” are explained.

In above conference, Effect of location is discussed with bottom wall effectiveness distribution. But in present study the earlier

mentioned quantity (location) is described with mid and cross plane non-dimensional temperature, Vector, CRVP, Centreline

and span effectiveness distribution for DWP located at (a) after, (b) before and after, (c) before film hole and (d) baseline .

In above conference, Effect of density ratio (DR) is discussed with vector distribution. In current study D.R is discussed with

CRVP, centreline and span-line, bottom wall effectiveness, mid and cross plane non-dimensional temperature distribution for

different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d) baseline at X/D = 10.

In above conference, Effect of jet inlet velocity (uj) is discussed with CRVP distribution. In present study uj is discussed with

Vector, centreline and span-line, bottom wall effectiveness, mid and cross plane non-dimensional temperature distribution for

different jet inlet velocity, u j = (a) 0.25 u cf, (b) 0.50 u cf, (c) 0.75 u cf, (d) 1.0 u cf and (e) baseline at X/D = 10.

For effect of Turbulent kinetic energy (T.K.E), in this conference, T.K.E is explained with mid plane non-dimensional temperature.

In current investigation T.K.E is discussed with Vector, CRVP, centreline and span-line, bottom wall effectiveness, cross plane

non-dimensional temperature distribution for different Turbulent Kinetic Energy, TKE= (a) Low, (b) High, and (c) baseline at X/D

= 10.

2.1 DISSTRIBUTION OF GRID DATA AND COMPUTATIONAL DOMAIN

Dimensions are expressed with assist of film-hole diameter, D in Figure 1-(a-d) [24]. DWP or triangular shaped VG has been

located at various downstream and upstream position (Table 1). The definition of angle of attack for vortex generator (VG) is

the angle between the VG direction vector and the flow direction at the corresponding location of the top of the VG leading

edge in the baseline case, as illustrated in Fig 1-(iv). The angle of attack, φ is 35 degrees as indicated by figure 1-(iv): Schematic

representing of (a) CFD (common flow down) configuration of DWP (Delta winglet pair) vortex generator (or figure 1-(a-d)). For
 current investigation CFD (Common flow down) arrangement of DWP has been chosen. Grid arrangement has been shown in

Figure 1-(h) [24]. Finer grids are used in the DWP region and film cooling hole region. However 1-(g) represents face numbering

and flow direction. Boundary Conditions will be demonstrated in "2.4 BOUNDARY CONDITION" Section and Table 2.

(a) (b)

(c) (d)

(g) (h)

Fig. 1: (a), (b), (c), (d) shows schematic diagram of numerical domain along with bottom wall, (g) face numbering and direction of

flow, (h) mesh distribution.

Table 1:All arrangement with the location of Delta Winglet Pair Vortex Generator

Table 1. All arrangement and position of DWP VG.

Figure Configuration

Figure 1-(a) (Baseline) Without DWP or without Vortex generator (VG)

Figure 1-(b) 1 DWP before hole (1 pair VG at upstream of film cooling hole)

Figure 1-(c) 1 DWP after hole (1 pair VG at downstream of film cooling hole)

Figure 1-(d) 1 DWP before and 1 DWP after hole (1 pair VG at upstream and 1 pair VG at downstream)
 7

2.2 USED TURBULENCE MODEL

Several turbulence models have been investigated (k- standard, k- RNG, k- realizable, k- standard, and k- SST). The

bottom wall effectiveness distribution and the mid plane temperature distribution are shown in Figures 2 -(i) and 2-(ii)

respectively. All other turbulence models are outperformed by the k-omega SST model in terms of effectiveness distribution.

The model's main benefit is that it accurately forecasts flow including an unfavorable pressure gradient, an airfoil and tran sonic

shock waves. Because of the development of primary vortices (CRVP) of the lowest strength, the k- SST model displays lower

jet lift-off than the other turbulence models, as seen in Figure 2-(ii). Distribution of CRVP is represented in figure 2-(iii). Centerline

as well as span line effectiveness are shown in figure 2-(iv). For the k- SST model, the most CRVP destruction occurred due to

the opposing rotation of secondary vortices against CRVP.

Several turbulence models (k- standard, k- RNG, k- realizable, k- standard, k- SST and RSM) have been studied to ascertain

the best eddy viscosity based model [23] by comparing the computational results with those of experimental investigation of

Ajersch et al. [35]. Figure 2-(v) reveals the wall-normal variation of normalised streamwise (u), wall normal (v), spanwise (w)

and turbulence kinetic energy (k) at different X/D and Z/D locations. The jet velocity, u j has been used for the normalization.

Among various turbulence models, k- SST model shows less discrepancy with computational and experimental results of

Ajersch et al. [35].

(a) (b)

(c) (d)
(e)

Figure 2-(i): Effectiveness distribution at bottom wall of K-ε standard, K-ε realizable, K-ε RNG, K-ω standard and K-ω Shear Stress

Transport turbulence models.

(a) K-epsilon standard (b) K-epsilon realizable

(c) K-epsilon RNG (d) K-omega standard

(e) K-omega SST

Figure 2-(ii): Non-dimensional temperature distribution at mid plane of K-ε standard, K-ε realizable, K-ε RNG, K-ω standard and K-ω

Shear Stress Transport turbulence models.

 9

(a) (b) (c)

(d) (e)

Figure 2-(iii) : Distribution of CRVP at cross plane for different turbulence model: (a) K-epsilon standard, (b) K-epsilon relizable,

(c) K-epsilon RNG, (d) K-omega standard, (e) K-omega SST.

(a) (b)

Figure 2-(iv): Distribution of Centreline (2-iv-a) and span line effectiveness (2-iv-b) for K-ε standard, K-ε realizable, K-ε RNG, K-ω

standard and K-ω Shear Stress Transport turbulence models.

(a) (b)
 (c) (d)

Figure 2-(v): Comparison of present simulation using different turbulence models with previous experimental

and simulation results: (a) Stream wise velocity (U/u j) profile for X/D = 1 and Z/D = 0, (b) Vertical velocity (V/u j)

profile for X/D = 1 and Z/D = -1, (c) Cross-stream velocity (W/u j) profile for X/D = 1 and Z/D = -0.5 and (d)

Turbulence intensity (√k/u j) for X/D = 5 and Z/D = 0.

2.3 GOVERNING EQUATION

A three-dimensional simulation is performed to solve governing equation. Primary assumption is fluids are Newtonian fluid and

physical properties are invariant. Physical properties which are invariant are dynamic viscosity (μ), kinematic viscosity (ν),

density (ρ), specific weight (w), specific gravity (s). For present study Reynolds Averaged Navier-Stokes (RANS) and energy

equations are used. RANS can be divided into steady and unsteady RANS. Steady RANS (SRANS) is fou nd to be unsuitable for

predicting the flow field. Unsteady RANS (URANS) predicts well the flow involving adverse pressure gradient, airfoil and

transonic shock waves. However, for obtaining URANS solution initially we have to take the assistance of SRANS. Equations for

continuity, momentum and energy in standard (dimensional) form are defined as

ui
0
xi
(1)

ui (ui u j )  p 1  2 ui  ij
   
t x j xi Re x j 2 x j
(2)


 
  u j

1  2  q j

t x j Re Pr x j 2 x j
(3)

Where  ij  uiuj and q j  uj 

Here for turbulence modelling we have used k- SST turbulence model.
 11
Transport equations can be defined as

𝜕 𝜕 𝜕 𝜕𝑘
(𝜌𝑘 ) + (𝜌𝑘𝑢 𝑖 ) = ( 𝛤𝑘 ) + 𝐺𝑘 − 𝑌𝑘 (4)
𝜕t 𝜕𝑥𝑖 𝜕𝑥𝑗 𝜕𝑥𝑗

𝜕 𝜕 𝜕 𝜕𝜔
(𝜌𝜔 ) + (𝜌𝜔 𝑢 𝑖 ) = ( 𝛤𝜔 ) + 𝐺𝜔 − 𝑌𝜔 + 𝐷𝜔 (5)
𝜕t 𝜕𝑥𝑖 𝜕𝑥𝑗 𝜕𝑥𝑗

Non-dimensional scale of length, velocity, time, and temperature is D, uj, D/uj and 𝜃 = (𝑇 − 𝑇𝑐𝑓 )/(𝑇𝑗 − 𝑇𝑐𝑓 ) respectively. Here

̅̅̅̅̅̅̅
𝑇𝑗 and 𝑇𝑐𝑓 defines the cross-flow and coolant jet temperatures, respectively. 𝜏𝑖𝑗 = −𝑢 ′ 𝑢 ′ is turbulent stress (Reynolds stress
𝑖 𝑗

̅̅̅̅̅̅
tensor) and 𝑞𝑗 = −𝑢 ′ 𝜃 ′ is turbulent heat flux. In this inquiry, Pradntl number (Pr ) is assumed to be 0.85.
𝑗 t

𝐺̃𝑘 indicates the production of turbulence kinetic energy. 𝐺𝜔 delineates about the generation of 𝜔. The effective diffusivity of 𝑘

& 𝜔 is specified as 𝛤𝑘 & 𝛤𝜔 . 𝑌𝑘 and 𝑌𝜔 show dissipation due to turbulence. Cross-diffusion term is represented by 𝐷𝜔 . ANSYS-

FLUENT [33] offers additional details on the 𝑘- 𝜔 SST model. The thermal field and flow with time-average are provided with

the assist of unsteady energy equation and RANS in Current investigation. Re, Pr represents Reynolds & Pradntl number.

First Navier-stoke equations (continuity, momentum and energy equations) are specified in dimensional form followed by

dimensionless form. Finally with the help of Reynolds decomposition time average forms are obtained. After that the boundary

layer simplified time averaged equations are furnished. Here we noticed that there are 5 unknowns but we have 3 equations.

So to close the system we need 2 additional equations. These 2 equations are can be taken from Turbulence modelling.

2.4 CONDITION OF BOUNDARY

Table 2 and Figure 1-(g) [24] present the boundary condition. Primary velocity (uα or ucf) of 104 m/s and 2.54 mm size of the film

cooling hole diameter, D is utilized for a constant Reynolds number (u αD/ ν) of 17000. Here blowing ratio (M = ρ juj / ρcfucf) is

kept as 1. Cross-flow inlet temperature kept as 300 K and film-hole outlet temperature has been considered as 150 K. At the

inlet, the freestream temperature is 300 K, and the coolant temperature is 250 K [36]. Cross flow inlet temperature is specif ied

as 300 K, and the plenum coolant temperature is set at 187.5 K [37]. The main flow has a free stream temperature of 298 K and

the coolant has a temperature of 188 K [8].

Outflow conditions have been used for the outlet. Symmetry and no-slip adiabatic wall boundary condition is provided for top

and bottom wall surface respectively. For cross and jet flow domain the application of periodic boundary condition is done in

span wise direction. For other faces and coolant hole an adiabatic wall boundary condition has been used. k- SST model is

used in commercial computational fluid dynamics software (ANSYS FLUENT) to solve governing equation.

Table 2: Boundary condition

 Boundaries Boundary conditions
Velocity Temperature Turbulent Intensity
1-2-3-4, 9-10-12-11 Velocity inlet (104 m/s of cross flow or 300 K for cross flow 1.75% for cross flow or
primary flow velocity, 52 m/s of jet velocity or or primary flow, primary flow,
secondary flow for M=0.5). 150 K for jet flow or 5.5% for jet flow or
secondary flow. secondary flow
8-7-2-1 (Bottom wall of Wall (No slip) Adiabatic wall
primary domain) (denotes no heat
transfer across the
wall)
9-13-14-11, 10-12-16-15 Wall (No slip) Adiabatic wall
(denotes no heat
transfer across the
wall)
Vortex generator, pipe Wall (No slip) Adiabatic wall
(denotes no heat
transfer across the
wall)
5-6-7-8 (outlet face or
right face of primary Outflow (denotes flow is fully developed at outlet)
domain)
1-4-5-8, 2-3-6-7 and 10-
15-13-9, 12-16-14-11
(forward, backward of
Periodic (in span-wise direction)
primary domain and
forward, backward of
secondary domain)
5-6-3-4 (Top wall of
Symmetry
primary domain)
1-2-3-4, 5-6-7-8, 1-4-5-8, 2-3-6-7, 5-6-3-4 and 8-7-2-1 denotes primary domain
10-12-16-15, 13-14-11-9, 10-15-13-9 and 12-16-14-11 denotes secondary domain

2.5 GRID INDEPENDENCE AND CONVERGENCE ANALYSIS ALONG WITH GRID DISTRIBUTION

Grid independence was carried out as mentioned in Figure 3. Fine mesh is kept near the wall so that it can solve the laminar

sub layer. In the DWP and film cooling hole region, comparatively finer grids are used. Fine grids are used near the wall region.

Away from the wall coarse grid is used. Unstructured mesh has been used for the current experiment. In order to resolve the


laminar sub layer, the near wall mesh is maintained at an extremely fine resolution ( Y  1 ). Both structured and unstructured

mesh can be used with FLUENT. Utilizing a collocated grid is FLUENT. Both pressure and velocity are retained at the cell's center

in a collocated grid. FLUENT, on the other hand, computes the face values of pressure from the cell values using interpolation

techniques. On a wide range of grids, including O and C type, conformal block, structured grids, multiblock structured grid, non-

conformal and unstructured triangular, tetrahedral, quadrilateral, hexahedral, and hexagonal grids, FLUENT can solve

problems.

Uniform grid convergence is reported by Roache et al. [34] originated from Richardson Extrapolation. Three type of meshes are

considered for this investigation namely coarse (3325674 numbers of cells), intermediate (4352732) and fine (5461548). Grid
 13
independence and convergence study has been performed with earlier mentioned meshes. Little disparity is noticed between

coarse and fine grid results. It shows that fine grid approaches grid independence. Hence for subsequent simulation fi ne grid

has been considered.

Grid Independence study express that the calculation result is independent of Grid size (coarse, intermediate and fine grid).

Structured and un-structured grid distribution are shown in Figure 3-(iii). Unstructured mesh kept near the film cooling hole and

for the rest, structured mesh used.

(i) (ii)

(iii)

Figure 3: (i) Grid independence study, (ii) Grid distribution of transverse plane at hole location, (iii) Structured and un structured

grid distribution.

2.6 SIMULATION PROCEDURE

The governing equations are resolved using FLUENT (version 14), a commercial computational fluid dynamics program. An
unstructured mesh was produced using ICEMCFD. For the near-wall treatment, a low Reynolds number model (low RE model)

was used. Pressure and velocity are connected using the SIMPLE algorithm. The discretization of momentum, energy, turbulent

dissipation rate, and turbulent kinetic energy is done using the second -order upwind technique. For continuity and momentum

equations, convergence criteria is maintained at order 10 -6 , whereas for the energy equation, it is set at order 10 -8 . The turbulent

kinetic energy dissipation convergence criteria is also maintained at order 10 -6 . By measuring the temperature and velocity

fluctuations as a function of iteration count and mass balance, the convergence was confirmed.

2.6.1 Method of solution

At each time step (∆t), 20 iterations are carried out to achieve convergence for continuity, momentum and temperature field.

All the discretized equations are solved using pressure based (segregated) solver. A transient solution with unsteady RANS ha s

been carried out using a fine mesh with an appropriate time step (∆t). ANSYS FLUENT utilizes an implicit scheme. Hence the

time step (∆t) is not limited by the stability limits. However, to model transient phenomena properly, there is a preference to

select the time step. The choice of ∆t should be at least one order of magnitude smaller than the shortest time constant in the

system being modeled. Simulations are performed with time step size, (∆t) = (∆x) min /(U)max = 1 x 10 -6 seconds (which corresponds

to a non-dimensional time step size ∆t x V/D = 4.54 x 10 -4 ) at a Reynolds number of 2369. Where (∆x) min represents minimum

grid spacing and (U) max provides maximum velocity in computational domain. With this time step, FLUENT converges the

solution at each time step. Then, it is steadily ramped up to a constant value 1 x 10 -5 seconds. The simulations are carried out

long enough duration to enable the statistics of the flow field to be determined. Initially, the steady -state RANS simulations

(SRANS) are carried out adopting the k- SST model, until the solution is converged. After setting the solver parameters

mentioned above, the transient method is started with converged SRANS and continued until the statistically steady -state is

achieved. The duration of the transient run is calculated by flow-through time (Tflow-through = L/V, where L is the characteristic

length of solution domain and V is a mean flow velocity). After the statistically steady state is reached, the data sampling for

time statistics option is enabled in the iterate panel (Run calculation) of the FLUENT menu to collect the data for time statistics.

The sampling interval is set as one so that the data sampling for time statistics can be executed in every time step. Therefo re,

the collected statistics can be post-processed at each sampling interval for mean and RMS values. The simulations are carried

out for several flow-through time (L/V) to obtain the statically stable data. The choice of simulation run is verified by comparing

the time-averaged velocities taken during the different flow through time after the initial transient run. Then, the post-

processing is undertaken and its results are plotted. However in ANSYS FLUENT, under “solution method” if we select first ord er

implicit for transient formulation then, adaptive time stepping method begins to perform. Adaptive method comes under “Run

calculation”. In this method, no need to mention time step, FLUENT itself decide required time step. Here maximum CFL (couran t

friedrichs levy) is defined as (uΔt/Δx). Where u, Δt and Δx represents maximum velocity (m/s), time step size (s), minimum grid
 15
size (m) respectively. Maximum CFL is obtained as 1.

The simulations are performed on Intel Xeon E5-2670V 2.5GHz 2 CPU-20-coreIvy Bridge on HP-Proliant-SL-230s-Gen8 servers

with 128GB of RAM. For film cooling cases, approximately 1500 -2200 CPU hours are required for solving the flow and thermal

fields and for collecting the time statistics.

2.7 VALIDATION STUDY

To validate the current investigation, the configuration without vortex generator known as baseline was compared to the

experimental inquiry. The baseline (present study) and experimental research conducted by Kapadia et al. [17] were found to

be quite similar. Figure 4 depicts the effectiveness distribution along the centerline. For the purposes of this investigation, the

averaged value in the span-wise direction was used.

Figure 4: Validation study for baseline (centreline effectiveness (averaged))

3. RESULT AND DISCUSSIONS

A gas turbine blade's ability to effectively cool a film using a vortex generator is the main goal of the current investigation. The

CRVP mitigation mechanism has been discussed. The effectiveness of bottom wall film cooling, mid plane temperature, cross

plane temperature and cross plane vector distribution have all been mentioned as results from the current study. Here, the

maximum CFL (courant friedrichs levy) is determined by (uΔt/Δx). Where u represents the maximum speed (m/s), Δt denotes

the time step size (s) and Δx describes the minimum grid size (m). Maximum CFL is expected to be 1 with Δt maintained at 10 -5

seconds. For stability criteria CFL number should be less than or equal to 1. That is why CFL number is too less.

3.1 EFFECT OF LOCATION

From Figure 5-(iii) of CRVP distribution it has been perceived that core of CRVP gradually moves upward and span -wise direction

starting from figure 5-(iii): (a) to 5-(iii): (c). More influence of CRVP has been observed with baseline (figure 5-(iii): (d)). Higher

strength of CRVP is noticed with DWP placed after film cooling hole (figure 5-(iii): (c)). ‘DWP placed before film cooling hole’

decreases the influence of CRVP by having the reverse motion to CRVP resulting better effectiveness. ‘DWP placed before hole’

create down wash flow resulting lower mixing between cross and jet flow. Figure 5-(i) represents mid plane non-dimensional

temperature distribution. Higher temperature distribution is depicted in baseline. This may be attributed to the creation of up

wash flow and lateral entry of heated cross-flow, greater mixing of primary and secondary flow. Fig. 5-(ii) discuss about vector

distribution for placement of DWP. In Figure 5-(iv), greater jet lift-off is detected for baseline. Highest effectiveness distribution

is associated with configuration of DWP before film cooling hole while lowest is related with baseline. With VG, the RMS (root

mean square) value decreases. When VG is located before film hole then it indicates reduction in amplitude of fluctuation

compared to baseline. The overall amplitude of fluctuation is higher for ‘DWP placed after film cooling hole’ compared to that

of ‘DWP placed before and after film cooling hole’ and ‘DWP placed before film cooling hole’. Also in baseline appearance of

DWP is not inspected, as a consequence down wash phenomena is not evolved. Hence lowest centerline and span line

effectiveness is observed with baseline (see figure 5-(v)).

Only value has been calculated for RMS (root mean square). As we are receiving the value which serve our purpose so we have

not plotted any figure related with RMS. It (RMS) indicates the amplitude of fluctuation. Lowest value of RMS has been

observed, with placement of DWP at before film cooling hole, lower value of jet inlet velocity, lower T.K.E (Turbulent kinetic

energy).

Here to discuss the effect of density ratio, jet flow velocity, turbulent kinetic energy we are implementing geometry which h ave

DWP placed before the film cooling hole as mentioned in figure 1-(b) in section 2.1 named Computational domain and grid

distribution. Flow conditions for baseline remain identical with section “2.4 Boundary condition”. Baseline represent only basic

geometry of film cooling without vortex generator as mentioned in figure 1-(a) in section 2.1 named Computational domain

and grid distribution.

(a) (b)
 17

(c) (d)

Figure 5-(i): Non-dimensional temperature distribution in the middle plane (Z = 0) for Delta Winglet Pair VG located at (a) after, (b)

before and after, (c) before film hole and (d) baseline

(a) (b) (c)

(d)
Figure 5-(ii): Vector distribution for placement of DWP at (a) before, (b) before and after, (c) after film hole and (d) baseline, at

X/D = 10

(a) (b) (c)

(d)

Figure 5-(iii): CRVP distribution for placement of DWP at (a) before, (b) before and after, (c) after film hole and (d)
 configuration without vortex generator at X = 10 D

(a) (b) (c)

(d)

Figure 5-(iv): Temperature distribution that is non-dimensional for placement of Delta Winglet Pair VG at (a) before, (b) before

and after, (c) after film hole and (d) configuration without vortex generator at X = 10 D

(a) (b) (a)

Figure 5-(v): Centreline and span effectiveness for placement of DWP at (a) before, (b) before and after, (c) after film hole at

X/D = 10

3.2 DENSITY RATIO'S IMPACT

Impact of density ratio is discussed with mid and cross plane non-dimensional temperature and vector allocation. Effect of

density ratio is also discussed with bottom wall effectiveness distribution. When DR= (a) 0.5, (b) 1.0, (c) 2.0 then Blowing ratio,

M = (a) 3.0, (b) 2.0, (c) 1.0 and (a) Uj = 6.0 UCf, (b) Uj = 2.0 UCf, (c) Uj = 0.5 UCf respectively. Here Uj = jet flow velocity, UCf = cross

flow velocity. Geometry used here is ‘vortex generator placed before film cooling hole’ as mentioned in figure 1-(b) in section

2.1 named “Computational domain and grid distribution”.

 19

3.2.1 CRVP DISTRIBUTION

Figure 6-(i) represents CRVP distribution for different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d) baseline at X/D = 10

3.2.2 CENTRELINE AND SPAN-LINE EFFECTIVENESS

Figure 6-(ii) represents higher centerline and span line effectiveness distribution for higher density ratio (DR). This may be

attributed to CFD configuration. CFD configuration creates down wash flow leads to lower jet lift-off, less entry of hated primary

flow in lateral direction. CRVP creates higher jet lift-off, permits more lateral entry of heated primary flow beneath the cold jet.

Greater density ratio mitigates detrimental effect of CRVP resulting higher effectiveness (Figure 6-(i)).

(a) (b)

(c) (d)

Figure 6-(i): CRVP distribution for different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d) baseline at X/D = 10

(a) (b) (a)

Figure 6-(ii): (a) centreline and (b) span line film cooling effectiveness at X = 10 D for different DR

3.2.2 BOTTOM WALL EFFECTIVENESS DISTRIBUTION

From bottom wall effectiveness distribution in Figure 6-(iii), with increment in density ratio results increase in bottom wall

effectiveness. Increment in blowing ratio results reduction in strength of primary vortices (CRVP). When density ratio reduce s

then enhance in jet lift off and decrease in effectiveness is observed. For increase in density ratio results less mixing among

cross and jet flow. Less communication among cross and jet flow outcomes reduction in detrimental effect of CRVP resulting

better cooling effectiveness.

(a) (b)

(c) (d)
Figure 6-(iii): Effectiveness distribution of bottom wall for different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d) baseline

3.2.3 TEMPERATURE DISTRIBUTION AT MID PLANE

Figure 6-(iv) exemplifies disintegrated jet lift off with boost in D.R. Increase in density ratio indicates enhancement in cooling

effect at mid plane. Disintegration in jet lift-off assists to retain cold fluid close to the bottom wall surface which enhances

bottom wall effectiveness. It may be attributed to less ventilation of cross flow, less detrimental effect of CRVP. Detrimental

effect of Counter Rotating Vortex Pair is, it directs the cross flow towards turbine blade, increasing the blade wall temperature

resulting lower film cooling effectiveness. Also CRVP is responsible for jet lift-off.

(a) (b)

(c) (d)

Figure 6-(iv): Non-dimensional temperature distribution at middle plane (Z = 0) for various density ratio, DR= (a) 2.0, (b) 1.0, (c)

0.5 and (d) baseline

 21
3.2.4 DISTRIBUTION OF TEMPERATURE AT CROSS-SECTIONAL PLANE

Figure 6-(v) presents less jet lift off with boost in density ratio. Increment in density ratio shows enhancement in cooling effect

at cross-sectional plane. Increase in density ratio assists to retain cold fluid close to the bottom wall surface which enhances

bottom wall effectiveness. This may be attributed to less entry of heated primary flow beneath the cold jet. CRVP improves cold

jet lift-off from bottom wall towards hot environment, increasing the turbine blade wall temperature resulting lower film

cooling effectiveness.

(a) (b) (c)

(d)

Figure 6-(v): Distribution of Non-dimensional temperature for various density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d)

configuration without vortex generator at X = 10 D

3.3 IMPACT OF JET INLET VELOCITY

Impact of jet inlet velocity (u j) is discussed with mid and cross plane non-dimensional temperature and vector distribution. Effect

of uj is also discussed with bottom wall effectiveness distribution. When Uj= (a) 0.25 UCf, (b) 0.50 UCf, (c) 0.75 UCf, (d)1.0 UCf then

Blowing ratio, M = (a) 0.5, (b) 1.0, (c) 1.5, (d) 2.0 and DR = (a) 1.0, (b) 2.0, (c) 3.0, (d) 4.0 respectively. Here UCf = 1.0 while

discussing Uj (Uj = jet inlet velocity, UCf = cross flow velocity). Geometry used here is ‘vortex generator placed before film cooling

hole’ as mentioned in figure 1-(b) in section 2.1 named “Computational domain and grid distribution”.

3.3.1 VECTOR and CRVP DISTRIBUTION

Figure 7-(i) explicates enhancement of u j accountable for less in jet lift off from bottom wall surface. From this figure we can

detect that intensification in strength of primary vortices has been taken place with strengthen in u j. Strengthen in CRVP are

primary responsible for devastation of cooling effectiveness.

(a) (b) (c)

(d) (e)

Figure 7-(i): Vector distribution for different jet inlet velocity, u j = (a) 0.25 ucf, (b) 0.50 ucf, (c) 0.75 ucf, (d) 1.0 ucf and (e) baseline

at X/D = 10

3.3.2 CENTRELINE AND SPAN-LINE EFFECTIVENESS

Figure 7-(ii) represents higher centerline and span line effectiveness distribution for lower uj. This may be attributed to CFD

configuration. CFD configuration creates down wash flow leads to lower jet lift-off, less entry of hated primary flow in lateral

direction. CRVP creates higher jet lift-off, permits more lateral entry of heated primary flow beneath the cold jet. Greater u j

enhances detrimental effect of CRVP resulting lower effectiveness.

(a-i) (b-i)

Figure 7-(ii): (a-i) centreline effectiveness, (b-i) span effectiveness for different jet inlet velocity, uj = (a) 0.25 ucf (0.236), (b)

0.50 ucf (0.471), (c) 0.75 ucf (0.707), (d) 1.0 ucf (0.942) at X/D = 10
 23

3.3.3 BOTTOM WALL EFFECTIVENESS DISTRIBUTION

From bottom wall effectiveness distribution in Figure 7-(iii), with increment in u j results decrease in bottom wall effectiveness.

Increment in uj results enlargement in strength of primary vortices (CRVP). When u j reduces then decrease in jet lift off and

enhance in effectiveness is observed. For increase in u j results more mixing among cross and jet flow. More communication

among cross and jet flow calculates enhancement in detrimental effect of CRVP resulting lower cooling effectiveness.

(a) (b)

(c) (d)

(e)

Figure 7-(iii): Effectiveness distribution of bottom wall for different jet inlet velocity, uj = (a) 0.25 u cf, (b) 0.50 u cf, (c) 0.75 u cf, (d)

1.0 u cf and (e) baseline at X/D = 10

3.3.4 TEMPERATURE DISTRIBUTION AT MID PLANE

Figure 7-(iv) exemplifies disintegrated jet lift off with boost in u j. Increase in u j indicates decrease in cooling effect at mid plane.

Disintegration in jet lift-off assists to retain cold fluid close to the bottom wall surface which enhances bottom wall effectiveness.

The reason is that less ventilation of cross flow results less detrimental effect of CRVP. Detrimental effect of Counter Rotating

Vortex Pair is, it controls the cross flow to the blade wall, increasing blade temperature resulting lower film cooling

effectiveness. Also CRVP is responsible for jet lift-off.

(a) (b)

(c) (d)

(e)

Figure 7-(iv): Non-dimensional temperature distribution in the middle plane for many jet inlet velocity, uj = (a) 0.25 ucf, (b) 0.50

ucf, (c) 0.75 ucf, (d) 1.0 ucf and (e) configuration without vortex generator at X = 10 D

3.3.5 DISTRIBUTION OF TEMPERATURE AT CROSS-SECTIONAL PLANE

Figure 7-(v) presents less jet lift off with decrease in u j. Increment in u j shows decrement in cooling effect at cross-sectional

plane. Decrease in u j assists to retain cold fluid close to the bottom wall surface which enhances bottom wall effectiveness. This

might be explained by less entry of heated primary flow beneath the cold jet. CRVP improves cold jet lift-off from bottom wall

towards hot environment, increasing the turbine blade wall temperature resulting lower film cooling effectiveness. Primary

vorticity (CRVP) decreases the cooling effectiveness. Strength of CRVP enhances with increase in u j.

(a) (b) (c)

 25

(d) (e)

Figure 7-(v): Temperature distribution that is non-dimensional for different jet inlet velocity, uj = (a) 0.25 ucf, (b) 0.50 ucf, (c) 0.75

ucf, (d) 1.0 ucf and (e) configuration which does not consist vortex generator at X = 10 D

M = (D.R) X (V.R)

From the above equation we can easily achieved that Density ratio, (D.R) is a fun ction of Blowing ratio, (M) and (V.R), Velocity

ratio or (D.R) = f((M),(V.R)). Also we can describe that Velocity ratio, (V.R) is a function of Blowing ratio, (M) and (D.R) , Density

ratio or (V.R) = f((M),(D.R)). For (V.R) = f((M),(D.R)) Density ratio can be treated as Temperature ratio, T.R or (V.R) = f((M),(T.R))

or Jet inlet velocity = f((M),(T.R)).

Moreover with the help of Temperature distribution at cross-sectional, mid plane and Bottom wall effectiveness distribution,

Centerline and span-line effectiveness, vector and CRVP distribution both parameters (Density ratio and Jet inlet velocity)

contribute to the strengthen of surface cooling effectiveness. It may be attributed to less jet lift off, keeping of cold flu id close

to the bottom wall surface, less entry of heated primary flow beneath the cold jet, less communication between primary and

secondary flow, with decrease in jet inlet velocity (u j), boost in density ratio (D.R) and CFD (common flow down) configuration.

3.4 TURBULENT KINETIC ENERGY'S IMPACT

Influence of Turbulent kinetic energy (T.K.E) is discussed with mid, cross plane non-dimensional temperature, vector

respectively. Effect of T.K.E is also discussed with bottom wall effectiveness distribution.

3.4.1 VECTOR AND CRVP DISTRIBUTION

From vector and CRVP distribution (Figure 8-(i) and 8-(ii)) it is observed that with the boost in T.K.E, strength of primary vortices

(CRVP) will be enhanced. This ultimately decreases the cooling effectiveness. As a result greater bottom wall effectiveness is

obtained for lowest T.K.E. When T.K.E enhances then decrease in η is noticed due to greater mixing between primary and

secondary flow. Also when T.K.E enhances then increase in blowing ratio is observed. Higher T.K.E indicates higher velocity o f

cold jet flow. As a result lower cooling effectiveness is perceived. Figure 8-(i) and 8-(ii) explicates enhancement of T.K.E

accountable for higher jet lift off from bottom wall surface. Figure 8-(i): (b) shows more intensification in strength of CRVP than
figure 8-(i): (a) in span wise direction means more lateral entry of heated primary flow for figure 8-(i): (b) than 8-(i): (a). Higher

jet lift off has been taken place with figure 8-(i): (c). Also origin of CRVP moves upward gradually for figure 8-(i). In other ward

we can say higher jet lift off is seen for figure 8-(i): (c).

(a) (b) (c)

Figure 8-(i): Vector distribution for different Turbulent Kinetic Energy, TKE= (a) Low, (b) High, and (c) baseline at X/D = 10

(a) (b) (c)

Figure 8-(ii): CRVP distribution for different Turbulent Kinetic Energy, TKE= (a) Low, (b) High, and (c) baseline at X/D = 10

3.4.2 CENTRELINE AND SPAN-LINE EFFECTIVENESS

Figure 8-(iii) represents lower centerline and span line effectiveness distribution for higher T.K.E. This may b e attributed to

higher mixing between cross flow and jet flow which creates higher jet lift-off, more entry of hated primary flow in lateral

direction. For higher T.K.E strength of CRVP increases. CRVP creates higher jet lift-off, permits more lateral entry of heated

primary flow beneath the cold jet.

(a) (b)

Figure 8-(iii): (a) centreline, (b) span-line film effectiveness at X = 10 D for different TKE
 27

3.4.3 BOTTOM WALL EFFECTIVENESS DISTRIBUTION

From bottom wall effectiveness distribution in Figure 8-(iv), it is observed that with increment in T.K.E results decrease in bottom

wall effectiveness. Increment in T.K.E results enlargement in strength of primary vortices (CRVP). When T.K.E reduces then

mitigation in jet lift off and increase in effectiveness is observed. For increase in T.K.E results greater cross-jet flow interaction.

Greater communication among cross and jet flow indicates enhancement in detrimental effect of CRVP.

(a) (b)

(c)

Figure 8-(iv): Effectiveness distribution of bottom wall for different Turbulent Kinetic Energy, TKE = (a) Low, (b) High, and (c)

baseline at X/D = 10

3.4.4 TEMPERATURE DISTRIBUTION AT CROSS-SECTIONAL PLANE

Figure 8-(v) presents less jet lift off with decrease in T.K.E. Decrement in T.K.E assists to retain cold fluid close to the bottom

wall surface which enhances bottom wall effectiveness and increase in cooling effect at cross-sectional plane. This may be

attributed to less entry of heated primary flow entering below the secondary flow and less strength in CRVP. It (CRVP) improves

jet lift-off by retaining hot fluid near to the turbine blade and sending secondary flow towards hot surrounding. As a

consequence the turbine blade wall temperature increases resulting lower film cooling effectiveness.
(a) (b) (c)

Figure 8-(v): Temperature distributions that are non-dimensional for various Turbulent Kinetic Energy, TKE = (a) Low, (b) High,

and (c) baseline with 10 X/D

4. CONCLUSION

A simulation research was conducted to look into the impact of common flow down DWP vortex genrator configuration on film

cooling effectiveness of gas turbine blade. Unsteady Reynolds averaged Navier-Stokes equation, energy equations and K-ω SST

model for turbulent flow have been implemented. Among the various turbulence models, The K-ω SST model shows the best

results in terms of effectiveness distribution. Reynolds number has been fixed at 17000. Time-averaged velocity field,

temperature field and vector distribution results are reported in this study.

Major findings are:

1. When compared to the baseline instance, the delta winglet pair arrangement exhibits higher film cooling

effectiveness

2. Position of VG is examined for both upstream and downstream to explain the impact of DWP. With DWP

situated before the hole, the best effectiveness has been attained, and with DWP located after the hole, the

worst is seen.

3. Due to the interaction of main and secondary flow, only primary vortices (CRVP) are generated in baseline

flow. Due to CRVP's ability to let cross-flow enter beneath the jet, the efficiency of film cooling is decreased.

4. The harmful effects of CRVP are lessened by the vortex generator. The core CRVP structures are weakened by

the addition of DWP.

5. When compared to the baseline situation, the delta winglet pair arrangement exhibits less turbulent mixing

between the jet and the cross-flow.

6. Lower mixing of the cross stream flow with the film cooling jet indicates lower averaged velocity and

associated flow in the near wall region at lower blowing ratio in DWP setup, leading to improved film cooling

effectiveness.

7. When D.R enhances then increment in η is noticed.

8. With the decrease in M, u j and T.K.E cooling effectiveness is increased.

 29

NOMENCLATURE

CRVP Counter rotating vortex pair

𝑀 Blowing ratio

𝜌𝑗 Density of jet

uj Jet exit velocity or jet inlet velocity

𝜌𝑐𝑓 Density of cross flow

ucf Cross flow inlet velocity

 Film cooling effectiveness

Taw Adiabatic wall non-dimensional temperature

𝑇𝑐𝑓 Cross flow inlet non-dimensional temperature

𝑇𝑗 Non-dimensional temperature of jet

VG Vortex generator

TV Transverse vortices

LV Longitudinal vortices

β Distance of DWP from leading edge of hole in down as well as upstream direction.

φ The angle of attack

DR Density ratio

PR Pradntl Number

LVGs Longitudinal vortex generators

WP Winglet pair

VG Vortex generator

DWP Delta Winglet pair

TI Turbulent intansity

DWs Delta wings

D Diameter of jet

 Dynamic viscosity

Re Reynolds number
 X/D Ratio of downstream distance from the centre of the film hole to film diameter

Z/D Ratio of transverse distance from the centre of the film hole to film diameter

U Stream wise velocity

V vertical velocity

W Span wise velocity

CFU Common-flow-up

CFD Common-flow-down

FIGURE LIST

Figure 1-(i): Vortical structure in film cooling geometry [30].

Figure 1-(ii): Cross flow with film cooling hole and generation of kidney vortices [31].

Figure 1-(iii): Vortex structures due to interaction of cross flow with rectangular hole and vortex generator.

Figure 1-(iv): Schematic of (a) CFU and (b) CFD configuration of DWP (Delta winglet pair) vortex generator.

Figure 1-(v): Schematic of different vortex generators in CFU configuration of (a) DWP (Delta winglet pair), (b) RWP (Rectangular

winglet pair) and (c) TRP (Trapezoidal winglet pair) vortex generator.

Fig. 2-(i): Effectiveness distribution at bottom wall of K-ε standard, K-ε realizable, K-ε RNG, K-ω standard and K-ω Shear Stress

Transport turbulence models.

Fig. 2-(ii) non-dimensional temperature distribution at mid plane of K-ε standard, K-ε realizable, K-ε RNG, K-ω standard and K-

ω Shear Stress Transport turbulence models.

Fig. 2-(iii): Distribution of CRVP at cross plane for K-ε standard, K-ε realizable, K-ε RNG, K-ω standard and K-ω Shear Stress

Transport turbulence models.

Fig. 2-(iv): Distribution of Centreline (2-iv-a) and span line effectiveness (2-iv-b) for K-ε standard, K-ε realizable, K-ε RNG, K-ω

standard and K-ω Shear Stress Transport turbulence models.

Fig. 3: Grid independence study

Fig. 4: Validation study for baseline (centerline effectiveness)

Fig. 5-(i): Mid plane non-dimensional temperature distribution for DWP (Non-dimensional temperature distribution in the

middle plane (Z = 0) for Delta Winglet Pair VG) located at (a) after, (b) before and after, (c) before film hole and (d) baseline

Fig. 5-(ii): vector distribution for placement of DWP at (a) before, (b) before and after, (c) after film hole for and (d) baseline at

X/D = 10

Fig. 5-(iii): CRVP distribution for placement of DWP at (a) before, (b) before and after, (c) after film hole and (d) configuration
 31
without vortex generator at X = 10 D

Fig. 5-(iv): Temperature distribution that is non-dimensional for placement of Delta Winglet Pair VG at (a) before, (b) before

and after, (c) after film hole and (d) configuration without vortex generator at X = 10 D

Fig. 5-(v): Centreline and span effectiveness for placement of DWP at (a) before, (b) before and after, (c) after film hole and (d)

baseline at X/D = 10

Fig. 6-(i): CRVP distribution for different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d) baseline at X/D=10

Fig. 6-(ii): (a) centreline and (b) span line film cooling effectiveness at X = 10 D for different DR

Fig. 6-(iii): Effectiveness distribution of bottom wall for different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2 .0 and (d) baseline

Fig. 6-(iv): Non-dimensional temperature distribution at middle plane (Z = 0) for various density ratio, DR= (a) 2.0, (b) 1.0, (c)

0.5 and (d) baseline

Fig. 6-(v): Distribution of Non-dimensional temperature for various density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d) configuration

without vortex generator at X = 10 D

Fig 7-(i): CRVP distribution for different jet inlet velocity, u j= (a) 0.236, (b) 0.471, (c) 0.707, (d) 0.942 and (d) baseline at X/D =

10

Fig 7-(ii): Vector distribution for different jet inlet velocity, u j= (a) 0.236, (b) 0.471, (c) 0.707, (d) 0.942 and (d) baseline at X/D

= 10

Fig. 7-(iii): (a) centreline effectiveness, (b) span effectiveness at X/D = 10 for different jet inlet velocity, u j= (a) 0.236, (b) 0.471,

(c) 0.707, (d) 0.942 and (d) baseline

Fig. 7-(iv): Effectiveness distribution of bottom wall for different jet inlet velocity, u j= (a) 0.236, (b) 0.471, (c) 0.707, (d) 0.942

and (d) baseline

Fig. 7-(v): Temperature distribution that is non-dimensional for different jet inlet velocity, uj = (a) 0.25 ucf, (b) 0.50 ucf, (c) 0.75

ucf, (d) 1.0 ucf and (e) configuration which does not consist vortex generator at X = 10 D

Fig 8-(i): Vector distribution for different Turbulent Kinetic Energy, TKE= (a) Low, (b) High, and (c) baseline at X/D = 10

Fig 8-(ii): CRVP distribution for different Turbulent Kinetic Energy, TKE= (a) Low, (b) High, and (c) baseline at X/D = 10

Fig. 8-(iii): (a) centreline, (b) span-line film effectiveness at X = 10 D for different TKE

Fig. 8-(iv): Effectiveness distribution of bottom wall for different Turbulent Kinetic Energy, TKE = (a) Low, (b) High, and (c)

baseline at X/D = 10

Fig. 8-(v): Temperature distributions that are non-dimensional for various Turbulent Kinetic Energy, TKE = (a) Low, (b) High,

and (c) baseline with 10 X/D

TABLE LIST

Table. 1 All arrangement and position of DWP VG.

Table. 2 Boundary conditions

REFERENCES

[1] Foster, N.W., Lampard, D., (1980). The Flow and Film Cooling Effectiveness Following Injection Through a Row of Holes.

ASME Journal of Eng. Power. 102: 584-588.

[2] Haven, B. A., Yamagata, D. K., Kurosaka, M., Yamawaki, S., and Maya, T., (1997). Anti -Kidney Pair of Vortices in Shaped

Holes and Their Influence on Film Cooling Effectiveness. ASME Paper No: 97-GT-45.

[3] Lemmon, C. A., Kohli, A., and Thole, K. A., (1999). Formation of Counter-Rotating Vortices in Film-Cooling Flows. ASME Paper

No: 99-GT-161.

[4] Fiebig, M., (1995). Embedded vortices in internal flow: heat transfer and pressure loss enhancement. International Journal

of Heat and Fluid Flow 16(5): 376-388.

[5] Jacobi, A., and Shah R., (1995). Heat transfer surface enhancement through the use of longitudinal vortices: a review of

recent progress. Experimental Thermal and Fluid Science. 11(3): 295-309.

[6] Chen, p., chya, k., shih, p., (2011). Effects of upstream ramp on the performance of film cooling. International journal of

thermal sciences 50: 1085-1094.

[7] Barigozzi, G., Franchini, G., and Perdichizzi, A., (2007). The Effect of an Upstream Ramp on Cylindrical and Fan Shaped Film

Cooling Hole Film Cooling—Part II Adiabatic Film Cooling Effectiveness. ASME Paper No: GT2007 -27079.

[8] Na, S., and Shih, T., (2007). Increasing Adiabatic Film Cooling Effectiveness Using an Upstream Ramp. ASME J. Heat Transfer.

129: 464–471.

[9] Zhou, W., and Hu, H. (2016). Improvements of film cooling effectiveness by using Barchan dune shaped ramps. International

Journal of Heat and Mass Transfer 103: 443-456.

[10] Rallabandi, A.P., Grizzle, J., Han, J.C., (2011). Effect of upstream step on flat plate film cooling effectiveness using PSP.

ASME J. Turbomach. 133: 041024 1-8.

[11] Zhang, F., Wang, X., Li, J., (2016). The effects of upstream steps with unevenly spanwise distributed height on recta ngular

hole film cooling performance. International Journal of Heat and Mass Transfer. 102: 1209 -1221.

[12] Yousif, H., Kutaeba, J.M., Khishali, AL., Falah, F., (2013). Film cooling experimental investigation for ramped -conical holes

geometry. International Journal of Scientific & Engineering Research. 4: 736-745.

 33
[13] Shinn, A.F., Vanka, S.P., (2011). Numerical Simulation of a Film-Cooling Flow with a Micro-ramp Vortex Generator. 49th

AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition 4 - 7 January 2011, Orlando,

Florida: 1-12, DOI: 10.2514/6.2011-767.

[14] Wijdevan, T.V.D., and Katz, J., (2014). Automative application of vortex generators in ground effect. ASME J. Fuid

engineering. 136: 1–8, DOI: 10.1115/1.4025917.

[15] Jun Yao, Yufeng Yao, (2011). Computational study of hole shape effect on film cooling performance. Proc. IMechE Part A:

Journal of Power and Energy 225: 1-14.

[16] Alok Dhungel, Yiping Lu, Wynn Phillips, Srinath V. Ekkad, James Heidmann (2002). Film cooling from a row of holes

supplemented with anti-vortex holes. Journal of Turbomachinary. 131(2): 021007 1 -10.

[17] Kapadia, S., Roy S., and Heidmann, J (2003). Detached eddy simulation of turbine blade cooling. In 36th AIAA Thermophysi cs

Conference, AIAA-2003-3632, Orlando, Florida: 23–26.

[18] Sargison, J. E. (2001). Development of a novel film cooling hole geometry. PhD Thesis, 2001 (University of Oxford).

[19] Sinha, A. K., Bogard, D. G., and Crawford, M. E. (1991). Film-cooling effectiveness downstream of a single row of holes with

variable density ratio. J. Turbomach. 113: 442–449.

[20] Katz, J., and Morey, F., (2008). Aerodynamics of large scale vortex generator in ground effect. ASME J. Fuid engineering .

130: 1–6.

[21] Angele, K.P., and Grewe, F., (2007). Instantaneous Behaviour of streamwise vortices for turbulent boundary layer

separation control. ASME J. Fuid engineering. 129: 226–235.

[22] Park, J., Pagan-Vazquez, A., Alvarado, J.L., Chamorro, P.L., Lux, M.S., Marsh P.C., (2017). Characterization of Tab-induced

Counter-Rotating vortex pair for mixing application. ASME J. Fuid engineering. 139: 1 –12.

[23] Halder N, Saha A, Panigrahi P, (2020). Enhancement in Film Cooling Effectiveness Using Delta Winglet Pair. Journal of

thermal science and engineering application 13: 1-17.

[24] Halder, N. and Panigrahi, P.K., (2021). Cooling performance of vortex generator. Proceedings of the Institution of

Mechanical Engineers, Part A: Journal of Power and Energy 235: 1619 -1638.

[25] Halder, Nirmal, Arun K. Saha, and P. K. Panigrahi. (2017), Influence of Delta Wing Vortex Generator on Counter Rotating

Vortex Pair in Film Cooling Application of Gas Turbine Blade. Fluid Mechanics and Fluid Power–Contemporary Research.

Springer, New Delhi: 95-103.

[26] Li, Q. and Liu, C., 2011. Implicit LES for supersonic microramp vortex generator: new discoveries and new

mechanisms. Modelling and Simulation in Engineering, 2011, pp.1-15.

[27] Chang, Y.T., Lin, H.C., Huang, C.J., Tsai, G.L. and Jiang, J.F., 2014. Numerical study of th ermal and flow characteristics of
plate-fin heat sink with longitudinal vortex generator installed on the ground. Mathematical Problems in Engineering, 2014.

[28] Gawande, V.B., Dhoble, A.S. and Zodpe, D.B., 2014. CFD analysis to study effect of circular v ortex generator placed in inlet

section to investigate heat transfer aspects of solar air heater. The Scientific World Journal, 2014.

[29] Tao, S.B., Tang, A.P., Xin, D.B., Liu, K.T. and Zhang, H.F., 2016. Vortex -induced vibration suppression of a circular cylinder

with vortex generators. Shock and Vibration, 2016.

[30] Yang, F. and Taslim, M.E., 2022. Experimental and numerical studies of the film cooling effectiveness downstream of a

curved diffusion film cooling hole. International Journal of Rotating Machinery, 2022.

[31] Fric, T., and Roshko, A., 1994, “Vortical Structure in the Wake of a Transverse Jet,” J. Fluid Mech., 279, pp. 1 –47.

[32] Haven, B.A. and Kurosaka, M., 1996, Improved Jet Coverage through Vortex Cancellation, AIAA Journal Vol.34, No. 1 1, pp.

2443-2444.

[33] ANSYS FLUENT User's Guide.

[34] Roache, P.J. (1994). Perspective: a method for uniform reporting of grid refinement studies. ASME J. Fluid engineering 1 16

(3): 405–413.

[35] Ajersch, P. Zhou, J-M. Ketler, S. Salcudean, M. Gartshore, L.S. Multiple jets in a crossflow: detailed measurements and

numerical simulations. J. Turbomach, Vol 119(2): pp.330-342, 1997.

[36] Sarkar, S. and Ranakoti, G., 2017. Effect of vortex generators on film cooling effectiveness. Journal of

Turbomachinery, 139(6), p.061009.

[37] Baheri, S., Tabrizi, S.A. and Jubran, B.A., 2008. Film cooling effectiveness from trenched shaped and compound

holes. Heat and mass transfer, 44, pp.989-998.

Comments:

Thanks for the author’s detailed revision. I still tend to reject this paper. There still many

problems need to be solved. The most important thing is that the paper is not innovative

enough and the draft is hard to read. As mentioned by the author, many similar works have

been done by the author, especially in “Problem Formulation” part, at the journal of power

and energy, impact of location, distance between film cooling hole and VG has been published.

What's the difference between current study and that paper?

Answer: Title of dubai paper: Impact of vortex generator on gas turbine blade cooling application.

Effect of location:

Dubai: Effect of location is discussed with bottom wall effectiveness distribution.

TSEA: Figure 5-(i): Mid and cross plane non-dimensional temperature, Vector, CRVP, Centreline andspan
effectiveness distribution for DWP located at (a) after, (b) before and after, (c) before film hole and (d)
baseline

Effect of density ratio:

Dubai: Effect of density ratio (DR) is discussed with vector distribution.

TSEA: Figure 6-(i): CRVP, centreline and span-line, bottom wall effectiveness, mid and cross plane non-
dimensional temperature distribution for different density ratio, DR= (a) 0.5, (b) 1.0, (c) 2.0 and (d)
baseline at X/D = 10

Effect of jet inlet velocity:

Dubai: Effect of jet inlet velocity (uj) is discussed with CRVP distribution.

Figure 14. CRVP distribution for different jet inlet velocity, uj = (a) 0.236, (b) 0.471, (c) 0.707, (d) 0.942 and
(e) baseline at X/D = 10

TSEA:

Figure 7-(i): Vector, centreline and span-line, bottom wall effectiveness, mid and cross plane non-
dimensional temperature distribution for different jet inlet velocity, u j = (a) 0.25 ucf, (b) 0.50 ucf, (c) 0.75
ucf, (d) 1.0 ucf and (e) baseline at X/D = 10

Effect of Turbulent Kinetic Energy:

Dubai: Effect of Turbulent kinetic energy (T.K.E) is discussed with mid plane non-dimensional temperature.

TSEA:

Figure 8-(i): Vector, CRVP, centreline and span-line, bottom wall effectiveness, cross plane non-
dimensional temperature distribution for different Turbulent Kinetic Energy, TKE= (a) Low, (b) High, and
(c) baseline at X/D = 10

For the above mentioned justification texts are highlighted with turquoise color in section ‘2. PROBLEM
FORMULATION’.

Title of the paper which is submitted but not published in ‘Journal of power and energy’: Influence of
Vortex Generator on Gas Turbine Blade.

Effect of distance between film cooling hole and vortex generator

Here ‘Effect of distance between film cooling hole and vortex generator’ is discussed with Centerline &
span-line effectiveness, with vector & streamline and CRVP distribution.

Effect of increase in the vortex generator

The above quantity is described with Bottom wall effectiveness and streamline distribution.
Effect of vortex generator’s length

The above parameter is described with vector and streamline distribution.

However these effects in ‘Journal of power and energy’ are not included in present study and texts are
highlighted with turquoise color in section ‘2. PROBLEM FORMULATION’.

Please clearly demonstrate the contribution of present paper and avoid copyright problem.

Answer: To avoid plagarism texts are colored with red color in almost every section.

Another big problem is that the title did help readers to distinguish the main research content of this
paper.

Answer: Title is modified.

The detailed problems are listed as below:

1. In the section of “2.2 USED TURBULENCE MODEL”, how do you judge that k-w SST

model outperforms other turbulence models? Do you have experimental data to validate it?
It's better to put "2.2 USED TURBULENCE MODEL" part after introduction of BCs and

Governing Equation parts when compared simulation results.

Answer:

Halder et al. [23] compared numerical investigation involving k-w SST turbulence model withexperimental
as well as computational study of Ajersch et al. [35], highlighted with turquoise color in section ‘2.2 USED
TURBULENCE MODEL’.

2. It's not convenient for readers to read when change the sequential order of different

turbulent models in Figure 2-(ii).

Answer: The sequential order of different turbulent models is modified in Figure 2-(ii), highlighted with
turquoise color in section‘2.2 USED TURBULENCE MODEL’.

3. What physical properties are invariant in this paper? At what temperature are these physical

properties taken? Why do you choose this temperature since the temperature differe nce

between main flow and jet cooling flow can reach 150K？

Answer: Primary assumption is physical properties are invariant in this paper. Physical properties which
are invariant are dynamic viscosity (μ), kinematic viscosity (ν), density (ρ), specific weight (w), specific
gravity (s).

Physical properties are taken based on 300 k, 150 k for cross flow, jet flow respectively.

At the inlet, the freestream temperature is 300 K, and the coolant temperature is 250 K [36]. Cross flow
inlet temperature is specified as 300 K, and the plenum coolant temperature is set at 187.5 K [37]. The
main flow has a free stream temperature of 298 K and the coolant has a temperature of 188 K [8].

4. Is current simulation steady or unsteady? If it is unsteady simulation, what is the time step?

Did you do time step independence study?

Answer: Present simulation is unsteady with a time step ≥ 1 x 10-5 seconds.

At each time step (∆t), 20 iterations are carried out to achieve convergence for continuity, momentum and
temperature field. All the discretized equations are solved using pressure based (segregated) solver. A
transient solution with unsteady RANS has been carried out using a fine mesh with an appropriate time step
(∆t). ANSYS FLUENT utilizes an implicit scheme. Hence the time step (∆t) is not limited by the stability
limits. However, to model transient phenomena properly, there is a preference to select the time step. The
choice of ∆t should be at least one order of magnitude smaller than the shortest time constant in the system
being modeled. Simulations are performed with time step size, (∆t) = (∆x)min /(U)max = 1 x 10-6 seconds
(which corresponds to a non-dimensional time step size ∆t x V/D = 4.54 x 10-4 ) at a Reynolds number of
2369. Where (∆x)min represents minimum grid spacing and (U)max provides maximum velocity in
computational domain. With this time step, FLUENT converges the solution at each time step. Then, it is
steadily ramped up to a constant value 1 x 10-5 seconds. The simulations are carried out long enough duration
to enable the statistics of the flow field to be determined. Initially, the steady-state RANS simulations
(SRANS) are carried out adopting the k-ɷ SST model, until the solution is converged. After setting the
solver parameters mentioned above, the transient method is started with converged SRANS and continued
until the statistically steady-state is achieved. The duration of the transient run is calculated by flow-
through time (Tflow-through = L/V, where L is the characteristic length of solution domain and V is a mean
flow velocity). After the statistically steady state is reached, the data sampling for time statistics option is
enabled in the iterate panel (Run calculation) of the FLUENT menu to collect the data for time statistics.
The sampling interval is set as one so that the data sampling for time statistics can be executed in every
time step. Therefore, the collected statistics can be post-processed at each sampling interval for mean and
RMS values. The simulations are carried out for several flow-through time (L/V) to obtain the statically
stable data. The choice of simulation run is verified by comparing the time-averaged velocities taken during
the different flow through time after the initial transient run. Then, the post-processing is undertaken and
its results are plotted. However in ANSYS FLUENT, under “solution method” if we select first order
implicit for transient formulation then, adaptive time stepping method begins to perform. Adaptive method
comes under “Run calculation”. In this method, no need to mention time step, FLUENT itself decide
required time step. Here maximum CFL (courant friedrichs levy) is defined as (uΔt/Δx). Where u, Δt and
Δx represents maximum velocity (m/s), time step size (s), minimum grid size (m) respectively. Maximum
CFL is obtained as 1.

The simulations are performed on Intel Xeon E5-2670V 2.5GHz 2 CPU-20-coreIvy Bridge on HP-Proliant-
SL-230s-Gen8 servers with 128GB of RAM. For film cooling cases, approximately 1500-2200 CPU hours
are required for solving the flow and thermal fields and for collecting the time statistics.
Which is kept as ‘2.6.1 Method of solution’. Highlighted in turquoise color.

5. Did the structure in Figure 1-(v) have been investigated in present study? If no, there is no

necessity to put these structures in "1.1 NOVELTY FOR ACADEMY AS WELL AS FOR

INDUSTRY "part. It has no relation with current study.

Answer: Figure 1-(v) is removed.

6. What is the geometry of DWP since DWP will affect the main flow and cooling flow mixing

procedure significantly? What is the definition of angle of attack？Is it the value of 35

degree marked in Figure 1 (a-d)?

Answer: Geometry of DWP is presented in figure 1-(iv). Definition of angle of attack and it’s value is
provided. It is kept in section 2.1 COMPUTATIONAL DOMAIN AND GRID DISTRIBUTION highlighted in
turquoise color.
7. In “3. RESULT AND DISCUSSIONS” part, when discussing the effect of density ratio, the

velocity ratio and blowing ratio are also changed; when discussing the effect of jet inlet

velocity, the temperature ratio and blowing ratio are also changed. It’s hard to tell which

parameter contributes to the strengthen or weaken of surface cooling effectiveness.

Answer:

M = (D.R) X (V.R)

From the above equation we can easily achieved that Density ratio, (D.R) is a function of Blowing ratio,
(M) and (V.R), Velocity ratio or (D.R) = f((M),(V.R)). Also we can describe that Velocity ratio, (V.R) is a
function of Blowing ratio, (M) and (D.R), Density ratio or (V.R) = f((M),(D.R)). For (V.R) = f((M),(D.R))
Density ratio can be treated as Temperature ratio, T.R or (V.R) = f((M),(T.R)) or Jet inlet velocity =
f((M),(T.R)).

Moreover with the help of Temperature distribution at cross-sectional, mid plane and Bottom wall
effectiveness distribution, Centerline and span-line effectiveness, vector and CRVP distribution both
parameters (Density ratio and Jet inlet velocity) contribute to the strengthen of surface cooling
effectiveness. It may be attributed to less jet lift off, keeping of cold fluid close to the bottom wall surface,
less entry of heated primary flow beneath the cold jet, less communication between primary and
secondary flow, with decrease in jet inlet velocity (uj), boost in density ratio (D.R) and CFD (commonflow
down) configuration.

Which is kept after ‘3.3.5 TEMPERATURE DISTRIBUTION AT CROSS-SECTIONAL PLANE’. Highlighted in

turquoise color.

8. It's better to tell reader that BCs will be demonstrated in "2.4 BOUNDARY CONDITION"

part in "2.1 COMPUTATIONAL DOMAIN AND GRID DISTRIBUTION" part when

reading the sentence "However 1-(g) and (i, j) represents boundary condition and

geometrical view respectively." while no BCs are found in Fig.1 (g). There is no need to

put another Figure 1(i) to show geometry since it gives no more information.

Answer: It is conveyed to the reader that BCs will be demonstrated in "2.4 BOUNDARY CONDITION"

Section and table 2. Caption of figure 1-(g) is modified. Figure 1-(i) is removed. It is kept in last two
sentence in section 2.1 COMPUTATIONAL DOMAIN AND GRID DISTRIBUTION highlighted in turquoise
color.
9. The statement of “Unstructured mesh has been used for the current experiment.” in Section

2.5 is quite confusing? In Fig.1(h)，the author showed a structure grid. But in "2.5 Grid

Independence and Convergence Analysis" part and "2.6 Simulation Procedure" part, the

author said that unstructured grid is used. Please show where structured and unstructured

grids are used for current study.

Answer: Structured and un-structured grid distribution are shown in Figure 3-(iii). Unstructuredmeshkept
near the film cooling hole. For the rest, structured mesh used. It is kept in section 2.5 GRID INDEPENDENCE
AND CONVERGENCE ANALYSIS ALONG WITH GRID DISTRIBUTION highlighted in turquoise color.

10. In Figure 5-(v): Centreline and span effectiveness for placement of DWP at (a) before, (b)

before and after, (c) after film hole and (d) baseline at X/D = 10, there is no (c) and (d).

Answer: Caption (Title) is modified for figure 5-(v) as mentioned in section ‘3.1 EFFECT OF LOCATION’
highlighted in turquoise color.

11. In "3.2 EFFECT OF DENSITY RATIO" part, in the sentence of "When DR= (a) 0.5, (b)1.0,

(c) 2.0 then Blowing ratio, M = (a) 3.0, (b) 2.0, (c) 1.0 and Uj = (a) 6.0, (b) 2.0, (c) 0.5

respectively. Here UCf = 1.0 while discussing Uj (Uj = jet flow velocity, UCf = cross flow

velocity). " Uj is non-dimensional parameter, but it is also jet flow velocity. It's quite

confusing

Answer: Text is modified like below mentioned sentence

When DR= (a) 0.5, (b) 1.0, (c) 2.0 then Blowing ratio, M = (a) 3.0, (b) 2.0, (c) 1.0 and (a) U j = 6.0 UCf, (b) Uj
= 2.0 UCf, (c) Uj = 0.5 UCf respectively. Here Uj = jet flow velocity, UCf = cross flow velocity. It is kept at
section "3.2 EFFECT OF DENSITY RATIO" highlighted in turquoise color.