GT2014-26781

1

Proceedings of ASME Turbo Expo 2014: Power for Land, Sea and Air

GT2014-26781

June 16-20, 2014, Düsseldorf, Germany

EXPERIMENTAL INVESTIGATION OF TURBINE PHANTOM COOLING ON ENDWALL WITH TRAILING EDGE

DISCHARGE FLOW

Yang Zhang, Xin Yuan*

Key Laboratory for Thermal Science and Power Engineering of Ministry of Education

Tsinghua University

Beijing 100084, P.R. China

*Email: [email protected]

ABSTRACT

The film cooling ejection on High Pressure (Hp) turbine component surface is strongly affected by the complex flow structure in the nozzle guide vane or rotor blade passages. The action of secondary flow in the main passage could dominate the film cooling effectiveness distribution on the component surfaces. The film cooling ejections from endwall and airfoil trailing edge are mixed by the secondary flow. Considering a small part of the coolant ejection from trailing edge discharge flow will move from the airfoil trailing edge pressure side to endwall downstream and then cover some area, the interaction between the coolants injected from endwall and airfoil trailing edge is worth investigating. Though the temperature of coolant discharge flow from trailing edge increases after the mixing process in the internal cooling procedure, the ejections moving from airfoil to endwall still have the potential of second order cooling. This part of the coolant is called “Phantom cooling flow” in the paper. A typical scale-up model of Hp turbine NGV is used in the experiment to investigate the cooling performance of ejection from trailing edge. Instead of the airfoil trailing edge platform itself, the film cooling effectiveness is measured on the downstream part of the endwall. This paper is focused on the trailing edge discharge flow with compound angle effects and the coolant from discharge holes moving from trailing edge to endwall surface. The coolant flow is injected from the straight discharge holes with a compound angle of 15deg and 45deg respectively. The film cooling holes on the endwall are used simultaneously to investigate the combined effects. The blowing ratio and different configurations of compound angle holes are selected to be the changing parameters in the paper. The experiment is completed with the blowing ratio changing from M=0.7 to M=1.3 and the compound angle is introduced to the entire row of trailing edge discharge holes (full span), with inlet Reynolds numbers of Re=3.5×105 and an inlet Mach number of Ma=0.1 INTRODUCTION

Higher performance of future gas turbines requires

efficiency improvements usually achieved by increasing the turbine inlet temperatures. However, turbine inlet temperatures (about 1600 ºC) are generally above the material failure limit of turbine components (about 1300 ºC), driving the need for newer cooling methods that reduce thermal loads on the turbine components. Methods such as film cooling and internal cooling have led to improvements in modern gas turbine performance.

As for the film-cooling research using pressure sensitive

painting (PSP), Zhang and Jaiswal [1] measured film-cooling effectiveness on a turbine vane endwall surface using the PSP technique. Using PSP, it was clear that the film-cooling effectiveness on the blade platform is strongly influenced by the platform’s secondary flow through the passage. Zhang and Moon [2] utilised the back-facing step to simulate the discontinuity of the nozzle inlet to the combustor exit cone. Nitrogen gas was used to simulate the cooling flow as well as a tracer gas to indicate oxygen concentration, such that the film’s effectiveness by the mass transfer analogy could be obtained. An experimental study was performed by Wright et al. [3] to investigate the film-cooling effectiveness measurements by three different steady state techniques: pressure sensitive paint, temperature sensitive paint and infrared thermography. They found that detailed distributions could be obtained in the critical area around the holes, and the true jet separation and re-attachment behaviour is captured with the PSP. Wright et al. [4] employed the PSP technique to measure the film-cooling effectiveness on a turbine blade platform due to three different stator-rotor seals. Three slot configurations placed upstream of the blades were used to model advanced seals between the stator and rotor. PSP was proven to be a valuable tool in obtaining detailed film-cooling effectiveness distributions. Gao et al. [5] studied turbine blade platform film cooling with typical stator-rotor purge flow and discrete-hole film cooling. The shaped holes presented higher film-cooling effectiveness and wider film coverage than the cylindrical holes, particularly at higher blowing ratios. The detailed film-cooling effectiveness distributions on the

2

platform were also obtained using the PSP technique. The results showed that the combined cooling scheme (slot purge-flow cooling combined with discrete-hole film cooling) was able to provide full film coverage on the platform. The measurements were obtained by Charbonnier et al. [6] applying the PSP technique to measure the coolant gas concentration. An engine representative density ratio between the coolant and the external hot gas flow was achieved by the ejection of CO2. The studies of the incidence angle effect on the flow field and heat transfer were also performed by researchers. Gao et al. [7] studied the influence of the incidence angle on the film-cooling effectiveness for a cutback squealer blade tip. Three incidence angles were investigated 0 at the design condition and ±5 at the off-design conditions. Based on the mass transfer analogy, the film-cooling effectiveness is measured with PSP techniques. It was observed that the incidence angle affected the coolant jet direction on the pressure side near tip region and the blade tip. The film-cooling effectiveness distribution was also altered.

In the trailing edge film cooling research field, Murata et al. [8] compared four different cutback geometries at trailing edge: two smooth cutback surfaces with constant-width and converging lands and two roughened cutback surfaces with transverse ribs and spherical dimples. The dimple surface was proven to be a favourable cutback-surface geometry, because it gave enhanced heat transfer without deterioration of the high film cooling effectiveness. Benson et al. [9] used magnetic resonance imaging experiments to provide the three-dimensional mean concentration and three component mean velocity field for typical trailing edge film-cooling cutback geometry. They designed new trailing edge geometries to modify the large scale mean flow structures responsible for surface effectiveness degradation. Martini et al. [10] conducted the experiments covering a broad variety of internal cooling designs with trailing edge discharge. The results clearly demonstrated the strong influence of the internal cooling design and the relatively thick pressure side structure on film cooling performance downstream of the trailing edge ejection slot. Dannhauer [11] used a combination of IR-thermography and thermocouples to investigate two different trailing edge geometries with coolant ejection. The first configuration was pressure side cutback while the other was a row of cylindrical holes. Telisinghe et al [12] studied the aero performance differences between a conventional turbine blade trailing edge and a trailing edge with a sharp cut-back. With regard to the discharge coefficient, the new configuration had a larger discharge coefficient compared with the conventional configuration.

As for blade endwall/platform film-cooling research, Yang et al. [13] used numerical simulation to predict the film-cooling effectiveness and heat transfer coefficient distributions on a rotating blade platform with stator-rotor purge flow and downstream discrete film-hole flows in a 1–1/2 turbine stage. The effect of the turbine work process on the film-cooling effectiveness and the associated heat transfer coefficients has been reported. The research by Kost and Mullaert [14] indicates that both the leakage flow of endwall upstream slots and the film-cooling ejection are strongly

influenced by the endwall pressure distribution. The leakage flow and the film-cooling ejection will move towards the low pressure region where high film-cooling effectiveness is captured. The influence of the pressure distribution could also explain why the suction side is cooled better than the pressure side. Another important factor is the passage vortex moved by the pressure gradient in the cascade. It could lead the coolant to move towards the suction side. Similar results were found in the research report by Papa et al. [15]. They captured the phantom cooling phenomenon on the rotor blade suction side and the coolant was ejected form an upstream slot. The paper indicates that the coolant from the endwall moves towards the suction side and then forms a triangular cooled area.

The effects of rotation on platform film cooling has been

investigated by Suryanarayanan et al. [16], who found that secondary flow from the blade pressure surface to the suction surface was strongly affected by the rotational motion causing the coolant traces from the holes to clearly flow towards the suction side surface. As regards investigations into combustor-turbine leakage flow, Thole’s group made significant contributions. With detailed investigations, the influence of slot shape and position as well as width has been analysed in a series of studies [17–19].

Oke and Simon [20] investigated the film-cooling flow

introduced through two successive rows of slots: a single row of slots and slots that have particular area distributions in the pitchwise direction. Wright et al. [21] used a 30 ° inclined slot upstream of the blades to model the seal between the stator and rotor. Twelve discrete film holes were located on the downstream half of the platform for additional cooling. Rehder and Dannhauer [22] experimentally investigated the influence of turbine leakage flows on the three-dimensional flow field and endwall heat transfer. In the experiment, pressure distribution measurements provided information about the endwall and vane surface pressure field and their variation with leakage flow. Additionally, streamline patterns (local shear stress directions) on the walls were detected by oil flow visualisation. Piggush and Simon [23] investigated the leakage flow and misalignment effects on the endwall heat transfer coefficients within a passage which had one axially contoured and one straight endwall. The paper documented that leakage flows through such gaps within the passage could affect endwall boundary layers and induce additional secondary flows and vortex structures in the passage near the endwall.

Past research has shown that strong secondary flow can result in changes to the local heat transfer on the endwall and platform. Many studies have investigated the effects of the blowing ratio or geometry on the endwall film cooling, indicating the flow field parameter could change the ejection flow trace. Few studies, however, have considered the combined effect of trailing edge discharge flow film cooling and endwall film cooling. To help fill this gap, the current paper discusses the effect of trailing edge discharge flow phantom cooling Nozzle Guide Vane (NGV) endwall. The factor of the blowing ratio and compound angle are considered.

3

EXPERIMENTAL METHODOLOGY

The film-cooling effectiveness is measured using the PSP technique. PSP is a photo luminescent material that, excited by visible light at 450 nm, emits light that could be detected by a high spectral sensitivity Charge-coupled Device (CCD) camera (PCO Sensicam Qe high performance cooled digital 12 bit CCD camera) fitted with a 600 nm band pass filter. The light intensity is inversely proportional to the local partial pressure of oxygen. The image intensity obtained from the PSP by the camera is normalised with a reference image

intensity ( refI ) taken without mainstream flow. Background

noise in the optical setup is eliminated by subtracting the nitrogen/air ejection image intensities with the image intensity obtained without mainstream flow and the light excitation ( blkI ). The recorded light intensity ratio can be

converted to the partial pressure ratio of oxygen with the parameters obtained in calibration, as shown in Equation (1):

2

2

Oref blk airratio

blk O ref

PI If f P

I I P

(1)

2 2

2

O Oair mix air mix

air O air

P PC C

C P

(2)

where I represents the intensity obtained at each pixel

and ratiof P is the parameter indicating the relationship

between the intensity ratio and the pressure ratio.

Figure 1. CALIBRATION CURVE FOR PSP.

The film-cooling effectiveness can be determined by the correlation between the PSP emitting intensity and the oxygen partial pressure. Calibration of the PSP was performed in a vacuum chamber by varying the pressure from 0 atm to 1.0 atm at three different temperatures. A PSP coated test coupon was placed in the vacuum chamber with transparent windows through which the camera could detect the light intensity on the coupon surface. The calibration curve is shown in Figure 1. A temperature difference of less than 0.5K between the

mainstream and the secondary flow should be guaranteed during the tests. To obtain film-cooling effectiveness, both air and nitrogen are used as the coolant. The molecular weight of nitrogen is almost the same as that of air, which makes the density ratio close to 1.0. By comparing the difference in oxygen partial pressure between the air and nitrogen ejection cases, the film-cooling effectiveness can be obtained using Equation (2).

EXPERIMENTAL FACILITY

The test section consists of an inlet duct, a linear turbine cascade, and an exhaust section. The inlet duct had a cross section 338 mm wide and 129 mm high. Not considering the ununiformed effect of the outlet flow field of the combustor, the incidence angle was not selected to be the variable in the experiment. The predominant vortex in the combustor made the velocity direction in the outlet section difficult to predict. The position of the stagnation point is strongly affected by the indefinite inlet flow angle, which in turn changed the leading edge and gill region film-cooling effectiveness distribution. During the test, the cascade inlet air velocity was maintained at 35 m/s for all the inlet flow conditions, corresponding to a Mach number of 0.1. A two times scale model of the GE-E3 guide vanes with a blade span of 129 mm and an axial chord length of 79 mm was used. [24]

Figure 2. THE TEST SECTION WITH MOVABLE TRAILING EDGE MODULE

AND THE ASSEMBLY DRAWING OF THE TEST SECTION

To obtain easier assemble method, the trailing edge module was movable and could be changed with the main blade part untouched. The trailing edge module had its own coolant supply plenum which could make the trailing edge discharge flow blowing ratio controlled independently. During the test the trailing edge module was kept at the same relative position with the Charge-coupled Device (CCD) so that the film cooling effectiveness distribution could be compared accurately. For coolant air supply, compressed air was delivered to a plenum located below the wind tunnel test section before being injected into the mainstream, as shown in the schematic diagrams in Figure 2. Four vanes and three

4

passages are included in the cascade. Two of the vanes have surface film cooling and only one passage has endwall film cooling.

Figure 3. SCHEMATIC OF CASCADE TEST RIG

Figure 4. THE TEST CASCADES WITH PSP ON ENDWALL SURFACE

Figure 5. DETAILS OF TRAILING EDGE DISCHARGE HOLES

Figure 6. THETRAILING EDGE DISCHARGE HOLES ON THE PRESSURE SIDE

AND THE INNER PIN FIN STRUCTURE

Past studies in the open literature have shown that the passage cross flow sweeps the film coolant from endwall to mid-span region due to the vortex in the passage. To reflect this phenomenon more clearly, all of the film-cooling holes were positioned in straight lines. Studies on the flat plates show that coolant from compound angle holes covers a wider area due to jet deflection. Four rows of radial cylindrical film-cooling holes were arranged on the gill region to form full covered coolant film. Figures 3–8 show the test cascades configurations and the geometric parameters of the blade.

Four rows of compound angle laidback fan-shaped holes were arranged on the endwall to form a full covered coolant film. Figure 7 shows the hole configurations and the blade’s geometric parameters. The first row was located upstream of the leading edge plane. The following three rows were evenly positioned inside the vane channel, with the last one located at 65% of the axial chord, downstream of the leading edge plane. The four rows of fan-shaped holes were inclined 30 ° to the platform surface and held at an angle of 0, 30, 45 and 60 ° to axial direction respectively. The laidback fan-shaped holes were featured with a lateral expansion of 10 ° from the hole-axis and forward expansion of 10 ° into the endwall surface, as shown in Figure 6. The diameter in the metering part (cylindrical part) of the shaped holes was 1 mm, and the expansion starts at 3D. Four coolant cavities were used for the four rows of holes respectively, as shown in Figure 7. (The extra coolant plenum chamber was designed to simulate the purge flow which was used as leakage flow supply in this experiment). The coolant supplied to each cavity was independently controlled by a rotameter dedicated to that cavity.

Figure 7. DETAILS OF THE FAN-SHAPED ENDWALL FILM-COOLING HOLES

One row of trailing edge discharge hole was arranged on the trailing edge surface at axial location of 77 mm (TE, 20

5

holes) as shown in figure 7. The trailing edge discharge holes were located on the pressure side trailing edge region so that the film-cooling effectiveness of the airfoil surface was difficult to access due to camera position limitation when the endwall film-cooling effectiveness was investigated. The trailing edge discharge flow was injected from the coolant plenum with inner pin fin structure, which consisted of five rows of pin fins and two connected cavities with a width of 4mm respectively. The spanwise height of the trailing edge discharge hole was 4mm and the axial length was 9 mm. The discharge holes were connected to the coolant plenum by cylindrical channel with a diameter of 2mm and a length of 15mm. The discharge holes were located along the spanwise direction; therefore, the distance between the last discharge hole and enwall surface should be the same (3mm). Due to the strong secondary flow in the trailing edge wake, it was difficult to control the local blowing ratios for every hole with one common coolant plenum chamber. One coolant cavity was used for the trailing edge discharge flow, as shown in Figure 8. (The coolant plenum had two inlet tubes as shown in Figure 8. However, only the bottom one was used in the test, while the top one was blocked in this investigation). The coolant supplied to the cavity was controlled by a rotameter. As shown in Figure 8, the trailing edge discharge holes were inclined 15° and 45° to the endwall surface respectively to study the effect of compound angle on phantom cooling.

Figure 8. TRAILING EDGE DISCHARGE HOLE CONFIGURATION (WITH

INNER STUCTURE OF COOLANT SUPPLY CHANNEL)

The uncertainties of the dimensionless temperature and

the film-cooling effectiveness are estimated as 3% at a typical value of 0.5 based on a 95% confidence interval. When the value is approaching zero, the uncertainty rises. For instance, the uncertainty is approximately 20% at the value of 0.05. This uncertainty is the cumulative result of uncertainties in calibration, 4%, and image capture, 1%. The absolute uncertainty for effectiveness varied from 0.01 to 0.02 units. Thus, relative uncertainties for very low effectiveness magnitudes can be very high, 100% at an effectiveness magnitude of 0.01.

Table 1 Discrete film hole location and orientation Hole Name

PositionX/Cax

Number D (mm)

Radial/ Compound

Angle to Surface

ROW1 -0.19 27 1/Fan 90 30

ROW2 0.02 13 1/Fan 60 30

ROW3 0.32 11 1/Fan 45 30

ROW4 0.59 11 1/Fan 30 30

Table 2 Experimental conditions considered in the test

Cases TE Film Cooling Endwall Film Cooling M

Air (L/min)

N2

(L/min) Air

(L/min) N2

(L/min)

Film Cooling With 15deg TE Ejection 1 52 54 72 75 0.7 2 75 78 102 106 1.0 3 97 100 133 138 1.3

Film Cooling With 45deg TE Ejection 4 52 54 72 75 0.7 5 75 78 102 106 1.0 6 97 100 133 138 1.3

Table 3 Geometric and flow conditions Scaling factor 2.20

Scaled up chord length 135.50 mm

Scaled up axial chord length 79.00 mm

Pitch/chord 0.80

Span/chord 0.95

Reynolds number at inlet 3.5×105

Inlet and exit angles 0 & 72 °

Inlet/Outlet Mach number 0.1 & 0.3

Inlet mainstream velocity 35 m/s

Mainstream flow temperature 305.5 K

Ejection flow temperature 305.0 K

Figure 9. SPATIALLY RESOLVED MAP OF NON-DIMENSIONAL TOTAL

PRESSURE LOSS WITHOUT FILM COOLING

Pitch

Sp

an

0 0.5 1 1.5 2

0

0.25

0.5

0.75

1 0.965

0.97

0.975

0.98

0.985

0.99

0.995

1

6

The pressure coefficient at outlet was measured by the five-hole probe and the airfoil surface pressure distribution was measured by the Kulite sensor. The aerodynamic results show the periodicity of the cascades and structure of secondary flow. The simple case without any film cooling ejection was measured as the baseline condition. The test range covers two pitches and the near endwall area where non-dimensional span height is between 0.75 and 1.0, as shown in the black rectangular, Figure 9. The test time is 5 seconds at every point for the five-hole probe, and the measurement frequency is 100HZ. The airfoil surface pressure was probed at 23 points on PS and SS respectively.

Figure 10. NON-DIMENSIONAL PRESSURE COEFFICIENT DISTRIBUTIONS

ON AIRFOILS

The figure shows the pressure coefficient distribution at

the outlet plain with a distance of 0.5 axial chord length from trailing edge. The high pressure loss area could be captured in the map which demonstrates the position of wake and corner vortex core. The low pressure coefficient area along the spanwise direction shows the position and strength of the trailing edge wake. The shape of wakes and vortex show that the cascades have a reasonable vane to vane periodicity quality. The secondary flow can be captured in the map obviously According to the airfoil surface pressure coefficient distribution shown in Figure 10, Vane-to-vane comparisons of the experimental measurement points demonstrate that a good level of periodicity, too. RESULTS AND DISCUSSION

Though the cascade is 2-d linear, the relative ejection

direction of the coolant is different at the different positions on the endwall. The strong secondary flow causes the ejection direction to be different relative to the endwall main flow direction. The interaction between the endwall film-cooling coolant and the secondary flow, especially the wakes, means that the endwall near the trailing edge hardly cooled, while the different flow direction in the main passage avoids this harmful interaction. According to the contours, without the trailing edge ejection the pressure gradient will strongly keep the coolant outside the inner flow of the wakes, leaving an apparent uncooled area near the trailing edge, especially near the discharge holes. Five coolant cavities are used for the trailing edge discharge holes and four rows of fan-shaped

endwall holes respectively. The coolant supplied to each cavity is controlled by a shared rotameter. During the test, the optical window and the CCD camera are fixed to the same relative position so that the condition with different trailing edge film cooling could be compared precisely. In this study, three different blowing ratios are chosen for the typical operational condition, low, medium and high cooling requirements. The blowing ratio of the coolant is varied, so the film-cooling effectiveness can be measured over a range of blowing ratios varying from M=0.7 to M=1.3 based on the mainstream flow inlet velocity.

The film-cooling effectiveness distributions and laterally averaged values at different incidence angles are shown in Figures 11–16, of which three typical blowing ratios are chosen: M=0.7, 1.0 and 1.3. The same trend could be found in the contours, so that the area coverage of coolant film is larger at higher blowing ratios. Figures 11–13 show the film-cooling effectiveness distribution on the endwall surface with 15deg and 45deg trailing edge film cooling, while the blowing ratio is controlled at M=0.7, M=1.0 and M=1.3 respectively. With the blowing ratio increasing, the area protected by the coolant is increasing. Though the coolant could cover the main part of the endwall surface, the unprotected area near the trailing edge is still apparent (shown with the red curve). This phenomenon represents the strong pressure gradient in the turbine cascades, dominating the moving direction of the coolant traces. The momentum of the coolant ejection is not strong enough to take the cool air into the high pressure area near the corner region (axial chord position between 1 and 1.1). A similar case could be observed near the leading edge where the coolant could only inject, apparently from the cooling holes near but not at the leading edge. The PS and SS leg of the horse shoe vortex could prevent the coolant attaching to the airfoil, creating a low film-cooling effectiveness area near the leading edge. All of the cooling holes unused on the pressure side were internally blocked, which caused the slight effect of the holes outlet geometry on the flow field being avoided in the experiment.

Figure 11 FILM-COOLING EFFECTIVENESS DISTRIBUTION ON ENDWALL (THE BLOWING RATIO IS 0.7, WITH 15DEG AND 45DEG TRAILING EDGE EJECTION)

0 0.2 0.4 0.6 0.8

0.96

0.97

0.98

0.99

1

X/Cax

Pre

ssu

re C

oe

ffici

en

t

VaneA PS

VaneB PS

VaneA SS

VaneB SS

trailing15 M=0.7 i= 0deg

Z/Z P

X/C ax

0 0.2 0.4 0.6 0.8 1

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 0 0.2 0.4 0.6


Z/Z P

X/C ax

0 0.2 0.4 0.6 0.8 1

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 0 0.2 0.4 0.6

7



The left subplot in Figures 11–13 shows the film-cooling effectiveness distributions on the endwall with 15deg trailing edge discharge hole film cooling when the blowing ratio on the endwall is controlled to be M=0.7, M=1.0 and M=1.3 respectively. The right subplot in Figures 11–13 shows the film-cooling effectiveness distributions on the endwall with 45deg trailing edge discharge hole film cooling. When the blowing ratio is M=0.7, the cooled area is slightly larger in the red curves of the contour for the 45deg case, while the cooled area is restricted to the Trailing Edge (TE) corner region (red lines). At higher blowing ratios, near the TE corner region, the cooled area is relatively larger. When the blowing ratio is M=0.7, an apparent unprotected area can be found at the edge of TE corner region for 15deg case, while this area is covered by the TE ejection coolant at the blowing ratio of M=1.0. The right subplot in Figure 13 shows the film-cooling effectiveness distributions in the corner region with 45deg trailing edge discharge hole film cooling when the blowing ratio is controlled to be M=1.3. Similar to the medium blowing ratio case, the high film-cooling effectiveness area near TE is obviously larger than the 15deg

case. Although valuable insight can be obtained from the

distribution maps (Figs. 11–13), the spanwise averaged plots (Figs. 14–16) offer additional insight and provide clear comparisons for large amounts of data. The effectiveness is averaged from the SS to the PS (Figs. 11–13) of the passage in the axial chord direction. The data outside the airfoil were deleted from the averaged results. The peaks in the plot correspond to the film-cooling holes’ location. Figures 14–16 indicate that, with the 45deg TE film cooling ejection, the end wall film-cooling effectiveness increases in the downstream area. Locally, the largest film-cooling effectiveness difference near the trailing edge appears at Cax=1.02. The average for 45deg case is significantly higher because the coolant injected from the trailing edge covers the endwall sufficiently, especially near the corner region where the local pressure is relatively high in the wake. The 45deg trailing edge ejection effect is clearly seen near the trailing edge part (axial chord position between 1 and 1.05) of the endwall.

Figure 14 LATERALLY AVERAGED FILM-COOLING EFFECTIVENESS ON THE ENDWALL (THE BLOWING RATIO IS 0.7, WITH 15DEG AND 45DEG TRAILING EDGE EJECTION)



Z/Z P

X/C ax

0 0.2 0.4 0.6 0.8 1

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 0 0.2 0.4 0.6


Z/Z P

X/C ax

0 0.2 0.4 0.6 0.8 1

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 0 0.2 0.4 0.6


Z/Z P

X/C ax

0 0.2 0.4 0.6 0.8 1

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 0 0.2 0.4 0.6


Z/Z P

X/C ax

0 0.2 0.4 0.6 0.8 1

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1 0 0.2 0.4 0.6

-0.2 0 0.2 0.4 0.6 0.8 1 1.20

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

e

ndw

all

X/C ax

i= 0deg M=0.7 trailing15i= 0deg M=0.7 trailing45

-0.2 0 0.2 0.4 0.6 0.8 1 1.20

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

e

ndw

all

X/C ax


8


Figure 17 FILM-COOLING EFFECTIVENESS DISTRIBUTION NEAR THE TRAILING EDGE EJECTION POSITION (THE BLOWING RATIO IS 0.7, WITH 15DEG AND 45DEG TRAILING EDGE EJECTION)



Figure 20 LOCAL FILM-COOLING EFFECTIVENESS DISTRIBUTION ON THE ENDWALL ALONG LINE 1 (THE BLOWING RATIO IS 0.7, WITH 15DEG AND 45DEG TRAILING EDGE EJECTION, LINE 1 IS IN FIGURES 17-19)

Figure 21 LOCAL FILM-COOLING EFFECTIVENESS DISTRIBUTION ON THE ENDWALL ALONG LINE 1 (THE BLOWING RATIO IS 1.3, WITH 15DEG AND 45DEG TRAILING EDGE EJECTION, LINE 1 IS IN FIGURES 17-19, THE MEANING OF TRAILING, FAR, LEFT AND RIGHT IS IN FIGURE 15)

With the trailing edge film cooling, the momentum of

the coolant is high enough to cover the endwall surface though the cooled area is limited to a small region near the trailing edge. A higher blowing ratio leads to more coolant being injected from the trailing edge cooling holes near the corner region such that the cooled peak area becomes wider on the endwall (represented by red curves in Figs. 17–19, 15deg and 45deg TE discharge holes are located near the endwall). As the coolant leaves the cooling holes, the trace of the ejection flow is led by the wake vortex developing downstream the trailing edge PS and SS. The vortex is strong at the junction region, which causes the boundary of coolant to move along the wake central line and towards the outlet. The film-cooling effectiveness distributions indicate that the cooling performance of the TE discharge holes is enough to cool the endwall surface covered by wakes. With a high blowing ratio the ejection could cover the area near the trailing edge, while the near TE region is still exposed to the hot environment when the blowing ratio is low. Increasing the blowing ratio could obviously improve the cooling effectiveness, so the performance near the trailing edge corner

-0.2 0 0.2 0.4 0.6 0.8 1 1.20

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

e

ndw

all

X/Cax



Z/Z P

X/C ax

1

2

3

0.9 1 1.1

-0.2

-0.1

00 0.2 0.4 0.6


Z/Z P

X/C ax

1

2

3

0.9 1 1.1

-0.2

-0.1

00 0.2 0.4 0.6


Z/Z P

X/C ax

1

2

3

0.9 1 1.1

-0.2

-0.1

00 0.2 0.4 0.6


Z/Z P

X/C ax

1

2

3

0.9 1 1.1

-0.2

-0.1

00 0.2 0.4 0.6


Z/Z P

X/C ax

1

2

3

0.9 1 1.1

-0.2

-0.1

00 0.2 0.4 0.6


Z/Z P

X/C ax

1

2

3

0.9 1 1.1

-0.2

-0.1

00 0.2 0.4 0.6

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

TR

AIL

ING

FA

R

L

ine

1

Z/Length


0 0.05 0.10

0.1

0.2

Trailing0.9 0.95 10

0.1

0.2

0.3

Far

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

TR

AIL

ING

FA

R

L

ine

1

Z/Length


0 0.05 0.10

0.1

0.2

Trailing0.9 0.95 10

0.1

0.2

0.3

Far

9

region is satisfied even for 15deg case.

Figures 11–13 and Figures 14–16 indicate the difference in film-cooling effectiveness distribution along the wake central line and along the vertical direction versus central line. When the blowing ratio is M=0.7, as shown in Figures 11 and 14, the main difference with 15eg and 45deg TE discharge film cooling is that, at a low blowing ratio, the ejection area of 15deg case is small. The coolant could hardly inject from the cooling holes on the trailing edge (pitch between -0.05 and -0.06, axial chord between 0.98 and 1.0, one discharge hole is located near the endwall surface with a distance of 3mm). This phenomenon shows that the low blowing ratio ejection could not overcome the high pressure factor in the wake, that the wake vortex and secondary flow weaken the trailing edge film cooling, even for the 45deg case. This condition is obviously changed at higher blowing ratios as shown in Figures 13 and 16. When the blowing ratio is M=1.3, the coolant could inject from the trailing edge discharge holes near the endwall for the 45deg case, while the film-cooling effectiveness is high not only near the trailing edge but also in the downstream area, especially in the wake trace. Figure 13 shows the trend that the behaviour of the ejection flow could apparently influence the downstream effectiveness distribution at high blowing ratios. The coolant from the trailing edge cooling holes will move along the wake vortex and then arrive at the downstream part which causes the film-cooling effectiveness at the outlet to be higher, especially in the wake region, as shown in Figure 11.

The phenomenon captured in this experiment has a close relationship with the secondary flow field in the turbine cascade. Previous literature could provide some important support material. The research by Rehder and Dannhauer [18] indicates that the coolant flow has apparent influence on the three-dimensional flow field of the turbine passage. The flow visualisation experiment shows that the moving trace of the passage vortex is from the pressure side to the suction side. The passage vortex, as well as the pressure gradient in the cascade could simultaneously force the coolant on the endwall to move onto the airfoil suction side. Similar results were found in the research report by Papa et al. [10]. They captured the phantom cooling phenomenon on the rotor blade suction side and the coolant was ejected from an upstream slot. The paper indicates that the coolant from the endwall would move towards the suction side and then form a triangular cooled area. Though the passage vortex and the pressure gradient in the rotor passage are stronger than that of the NGV, the mechanism of suction side over-cooling is similar. The comparable results provide a reasonable explanation of the over cooling phenomenon near the suction side in this experiment.

Figures 20 and 21 compare the local film-cooling

effectiveness distribution at streamwise location 1 with different blowing ratios. The position of the computing area is indicated by the TRAILING to FAR white line along the wake central line direction in Figures 11–13. With the 45deg trailing edge ejection, the local film-cooling effectiveness apparently improves near the trailing edge, as shown in

Figures 20 and 21 where the curve representing the 45deg ejection condition is apparently higher near the TE. Meanwhile, the film-cooling effectiveness in the wake and near the outlet is hardly changed. The well protected region is limited to the TE corner region. After cooling the TE comer region, the coolant strongly interacts with the secondary flows such as the passage vortex and wakes. The main flow eliminates the momentum of the trailing edge film cooling quickly, which makes the film-cooling effectiveness of the TE to be the same. On the other hand, the main flow further mixes the coolant and the hot gas on the endwall, which leads the ejection flow to lift off the endwall surface and then move to the main flow. These two factors hardly cause the film-cooling effectiveness to change at the far side of the trailing edge which is close to the outlet.



Figures 22 and 23 compare the local film-cooling effectiveness distribution at vertical location 2 with different blowing ratios. As the blowing increases, the film-cooling

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

LE

FT

RIG

HT

L

ine

2

Z/Length


0 0.05 0.10

0.1

0.2

Left0.9 0.95 10

0.1

0.2

0.3

Right

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

LE

FT

RIG

HT

L

ine

2

Z/Length


0 0.05 0.10

0.1

0.2

Left0.9 0.95 10

0.1

0.2

0.3

Right

10

effectiveness apparently improves near the trailing edge. Meanwhile, the higher effectiveness area approaches the two sides of the wake central line, LEFT and RIGHT respectively. The well protected region is at the central line and the near region by the central line (non-dimensional length Z/Length is between 0.3 and 0.7). In the TE corner region of the passage, the coolant strongly interacts with the secondary flows such as the passage vortex and wakes. The main flow pushes the coolant towards the mid-passage region, which causes the protected area to be larger on the left side. However, the main flow still mixes the coolant and the hot gas in the passage, which leads the ejection flow to lift off the endwall surface, which hardly causes the film-cooling effectiveness to change at the off central line region where the endwall surface is not covered by the TE ejection.



Figures 24 and 25 show the local film-cooling

effectiveness distribution at vertical location 3 where the

coolant is moved to the downstream part of the endwall near outlet, with the blowing ratio controlled at M=0.7 and M=1.3. When the blowing ratio is M=0.7 (Fig.22), no apparent high effectiveness area could be found along the vertical line (non-dimensional length Z/Length is between 0.3 and 0.7), while the influence of the trailing edge film cooling could not be probed in this area, the downstream part at the wake central line. This indicates that the effects of the trailing edge film cooling are not apparent far away from the TE corner region of endwall surface when the blowing ratio is relatively low. Figure 25 compares the local film-cooling effectiveness distribution in the downstream area when the blowing ratio is M=1.3. The figure shows that the increase in the blowing ratio improves the local film-cooling effectiveness of the TE corner region where the ejection is fully dominated by the wakes. The lower film-cooling effectiveness near the LEFT and RIGHT side indicates that the coolant ejection is influenced by the main wake flow field, especially the wake vortex which causes strong mixing flow in the wake trace. In this area, the main flow is dominated by the trailing edge wakes. Lower effectiveness means stronger influence of the vortex, which shows that the distance from the TE could change the influence of the trailing edge film cooling on the endwall. The film-cooling effectiveness curve representing the case of 45deg TE film cooling is almost the same as the curves representing the 15deg case in the mid-pitch region (non-dimensional length Z/Length is between 0.3 and 0.7) as shown in Figure 25. At the downstream part dominated by the TE wake, the influence of trailing edge ejection is apparently weakened by the secondary flow. The higher momentum of the coolant ejection flow could not effectively overcome the mixing trend of the wake vortex and then form a low film-cooling effectiveness area in the wake.

Figure 26 SPATIALLY RESOLVED MAP OF NON-DIMENSIONAL TOTAL PRESSURE LOSS WITH TRAILING EDGE DISCHARGE (45DEG), M=0.7


0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

LE

FT

RIG

HT

L

ine

3

Z/Length


0 0.05 0.10

0.1

0.2

Left0.9 0.95 10

0.1

0.2

0.3

Right

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

LE

FT

RIG

HT

L

ine

3

Z/Length


0 0.05 0.10

0.1

0.2

Left0.9 0.95 10

0.1

0.2

0.3

Right

Pitch

Sp

an

0 0.5 1

0.75

0.875

10.97

0.98

0.99

1

Pitch

Sp

an

0 0.5 1

0.75

0.875

10.97

0.98

0.99

1

11

Figure 28 SPATIALLY RESOLVED MAP OF NON-DIMENSIONAL TOTAL PRESSURE LOSS WITH ENDWALL FILM COOLING, M=0.7 (BASELINE)

Figure 29 COMPARISON OF SPANWISE AVERAGED PRESSURE LOSS ALONG PITCH WITH DIFFERENT TRAILING EDGE DISCHARGE GEOMETRY (M=0.7)

When the trailing edge blowing ratio is controlled to be

M=0.7, the geometry of trailing edge has strong influence on the outlet total pressure loss. The 45deg discharge flow can obviously increase the pressure loss in the trailing edge wakes and corner vortex, as shown in Figures 26-28. The 45deg discharge flow will lead another high total pressure loss region above the corner vortex to form. However, the total pressure loss does not increase apparently when the discharge angle is 15deg. And the width of the wake is less than that of the baseline case without any trailing edge discharge. This result shows the 15deg discharge has better aerodynamic performance than the 45deg design, though its secondary cooling performance on endwall surface is worse. The Figure 29 shows the spanwise averaged total pressure coefficient distribution along pitch direction. Compared with 45deg case, the 15deg design has less total pressure loss in the wake area.



Figure 32 SPATIALLY RESOLVED MAP OF NON-DIMENSIONAL TOTAL PRESSURE LOSS WITH ENDWALL FILM COOLING, M=1.0 (BASELINE)


The high pressure loss area near corner region represents

the strength and position of secondary flow in the main passage. When the trailing edge discharge angle is 45deg, a new vortex forms above the original one at a non-dimensional span of 0.875, as shown in Figure 26, 30 and 34. It is the result of interaction between main passage secondary flow and the discharge flow. The higher total pressure loss has negative influence on the rotating blades downstream. The mechanism behind this phenomenon could be illustrated though formation of shear flows. The big angle between the discharge flow and main flow causes strong shear effects on the turning air, which makes a new vortex possible. Then the new vortex moves up with the lift-off flow on suction side. The final double vortex structure is the result of interaction between 45deg coolant ejection and the secondary flow in upstream passage.

Pitch

Sp

an

0 0.5 1 1.5 2

0.75

0.875

10.97

0.98

0.99

1

0 0.5 1 1.5 2

0.98

0.99

1

Pitch

Mea

n P

ress

ure

Coe

ffic

ient

LE 45deg M=0.7 TE 15deg M=0.7

Pitch

Sp

an

0 0.5 1

0.75

0.875

10.97

0.98

0.99

1

Pitch

Sp

an

0 0.5 1

0.75

0.875

10.97

0.98

0.99

1

Pitch

Sp

an

0 0.5 1 1.5 2

0.75

0.875

10.97

0.98

0.99

1

0 0.5 1 1.5 2

0.98

0.99

1

Pitch

Mea

n P

ress

ure

Coe

ffic

ient


12



Figure 36 SPATIALLY RESOLVED MAP OF NON-DIMENSIONAL TOTAL PRESSURE LOSS WITH ENDWALL FILM COOLING, M=1.3 （BASELINE）


As the blowing ratio increases, the advantage of 15deg

design is still apparent. The high total pressure loss regions of 45deg case exist in the wake and corner vortex range. And the spanwise averaged pressure coefficient distributions in Figures 33 and 37 show the worse aerodynamic performance of 45deg design. Come to the conclusion, the 15deg railing edge discharge design has smaller aerodynamic loss and the position of the corner vortex core does not change. However, the 45deg design causes extra aerodynamic loss and the high total pressure loss region interacts with the corner vortex.

Though the secondary cooling function of 45deg case is better, its disadvantage of aerodynamic loss cannot be ignored. CONCLUSIONS

In general, the trailing edge discharge flow affects the

coolant distribution on the downstream part of endwall surface. The results show that with an increasing blowing ratio, the film-cooling effectiveness increases on the endwall surface, especially near the TE corner region. The film-cooling effectiveness difference is weakened with the axial chord increase, indicating that the trailing edge phantom film-cooling ejection mixes with the main flow strongly at the passage outlet, thus forming a low influence region in the area further downstream. With increasing blowing ratios, the improvement is also captured at the part further downstream near the trailing edge region. The influence of the blowing ratio is apparent for the trailing edge phantom coiling on the endwall surface.

As the blowing ratio varies from M=0.7 to M=1.3, the

influence of the trailing edge discharge flow on the endwall film cooling increases near the TE corner region. Simultaneously, the area of influence will be limited at the trailing edge region. In conclusion: 1）the film-cooling effectiveness increases on the endwall surface near the trailing edge region, with the highest parameter at the impingement point: 2) a one-peak cooled region develops along the mean line downstream the trailing edge as the blowing ratio increases; 3）the advantage of the trailing edge discharge flow phantom cooling was apparent when the compound angle was 45deg, while the coolant film was obviously weakened along the axial chord at smaller compound angle 15deg. The influence of the trailing edge discharge flow phantom cooling could only be detected at the downstream area of the trailing edge, while it has little influence on the main passage endwall film cooling. NOMENCLATURE C =concentration of gas / actual chord length of scaled up

blade profile

D =film-hole diameter, mm

i =incidence angle

I =light intensity

L =length of film hole, mm

M =blowing ratio, ρcVc/ρ∞V∞

Ma =Mach number

PS =pressure side

P =partial pressure

PSP =pressure sensitive paint Re =Reynolds number

SS =suction side

TE =Trailing Edge

V =velocity, m/s

X , Z =Cartesian coordinate system =film cooling effectiveness

Subscripts

Pitch

Sp

an

0 0.5 1

0.75

0.875

10.97

0.98

0.99

1

Pitch

Sp

an

0 0.5 1

0.75

0.875

10.97

0.98

0.99

1

Pitch

Sp

an

0 0.5 1 1.5 2

0.75

0.875

10.97

0.98

0.99

1

0 0.5 1 1.5 2

0.98

0.99

1

Pitch

Mea

n P

ress

ure

Coe

ffic

ient


13

aw =adiabatic air =air condition ax =axial chord blk =back ground value c =coolant fluid in =inlet mix =mixture condition O2 =pure oxygen ratio =partial pressure of oxygen ref =reference value sp =span wise =free stream condition REFERENCES [1] Zhang, L., Jaiswal, R.S., 2001. “Turbine Nozzle Endwall

Film Cooling Study Using Pressure-Sensitive Paint”.

ASME Journal of Turbomachinery, 123, pp.730–738. [2] Zhang, L., Moon, H.K., 2003. “Turbine Nozzle Endwall

Inlet Film Cooling: The Effect of a Back-Facing Step”. In ASME Turbo Expo 2003, collated with the 2003 International Joint Power Generation Conference, Atlanta, ASME Paper No.GT2003–38319.

[3] Wright, L.M., Gao, Z., Varvel, T.A., and Han, J.C., 2005. “Assessment of Steady State PSP, TSP, and IR Measurement Techniques for Flat Plate Film Cooling”. In ASME 2005 Summer Heat Transfer Conference, ASME Paper No.HT2005–72363.

[4] Wright, L.M., Blake, S., Han, J.C., 2006. “Effectiveness Distributions on Turbine Blade Cascade Platforms through Simulated Stator-Rotor Seals”. In 9th AIAA/ASME Joint Thermophysics and Heat Transfer Conference, San Francisco, AIAA Paper No.2006–3402.

[5] Gao, Z., Narzary, D., Han, J.C., 2009. “Turbine Blade Platform Film Cooling with Typical Stator-Rotor Purge Flow and Discrete-Hole Film Cooling”. Journal of Turbomachinery, 131, pp.041004/1–11.

[6] Charbonnier, D., Ott, P., Jonsson, M., Cottier, F., Köbke, Th., 2009. “Experimental and Numerical Study of the Thermal Performance of a Film Cooled Turbine Platform”. In ASME Turbo Expo 2009: Power for Land, Sea, and Air, Orlando, ASME Paper No.GT2009-60306.

[7] Gao, Z., Narzary, D., Mhetras, S., Han, J.C., 2009. “Effect of Inlet Flow Angle on Gas Turbine Blade Tip Film Cooling”. Journal of Turbomachinery, 131, pp.031005/1–12.

[8] Murata, A., Nishida, S., Saito, H., Iwamoto, K., Okita, Y., Nakamata, C., 2011. “Effects of Surface Geometry on Film Cooling Performance at Airfoil Trailing Edge”. In ASME Turbo Expo 2011, Vancouver, British Columbia, Canada, ASME Paper No. GT2011-45355.

[9] Benson, M., Yapa, S., Elkins, C., Eaton, J., 2012. “Experimental-Based Redesigns for Trailing Edge Film Cooling of Gas Turbine Blades”. In ASME Turbo Expo 2012, Copenhagen, Denmark, ASME Paper No. GT2012-68067.

[10] Martini, P., Schulz, A., Bauer, H. 2006. “Film Cooling Effectiveness and Heat Transfer on the Trailing Edge Cutback of Gas Turbine Airfoil with Various Internal

Cooling Designs”. Journal of Turbomachinery, 128, 196–205.

[11] Dannhauer, A., 2009. “Investigation of Trailing Edge Cooling Concepts in a High Pressure Turbine Cascade – Analysis of the Adiabatic Film Cooling Effectiveness”. In ASME Turbo Expo 2009, Orlando, Florida, USA, ASME Paper No. GT2009-59343.

[12] Telisinghe, J., Ireland, P., Jones, T., Barrett, D., Son, C., 2006. “Comparative Study Between A Cut-Back and Conventional Trailing Edge Film Cooling System”. In ASME Turbo Expo 2006, Barcelona, Spain, ASME Paper No. GT2006-91207.

[13] Yang, H., Gao, Z., Chen, H.C., Han, J.C., Schobeiri, M.T., 2009. “Prediction of Film Cooling and Heat Transfer on a Rotating Blade Platform with Stator-Rotor Purge and Discrete Film-Hole Flows in a 1–1/2 Turbine Stage”. Journal of Turbomachinery, Transactions of the ASME, 131, pp. 041003/1–12.

[14] Kost F., Mullaert, A., 2006. “Migration of Film-Coolant from Slot and Hole Ejection at a Turbine Vane Endwall”. ASME Turbo Expo 2006: Power for Land, Sea, and Air (GT2006), Barcelona, Spain, ASME Paper No. GT2006-90355.

[15] Papa, M., Srinivasan, V., Goldstein, R.J, 2010, “Film Cooling Effect of Rotor-stator Purge Flow on Endwall Heat/Mass Transfer”. ASME Turbo Expo 2010: Power for Land, Sea, and Air (GT2010), Glasgow, UK, ASME Paper No.GT2010-23178.

[16] Suryanarayanan, A., Ozturk, B., Schobeiri, M.T., Han, J.C., 2010. “Film-Cooling Effectiveness on a Rotating Turbine Platform Using Pressure Sensitive Paint Technique”. Journal of Turbomachinery, 132, pp.041001/1–13.

[17] Hada, S., Thole, K.A., 2011. “Computational Study of a Midpassage Gap and Upstream Slot on Vane Endwall Film-Cooling”. Journal of Turbomachinery, 133, pp. 011024/1–9.

[18] Knost, D.G., Thole, K.A., 2005. “Adiabatic Effectiveness Measurements of Endwall Film-Cooling for a First-Stage Vane”. Journal of Turbomachinery, 127, pp. 297–305.

[19] Cardwell, N.D., Sundaram, N., Thole, K.A., 2006. “Effect of Midpassage Gap, Endwall Misalignment, and Roughness on Endwall Film-Cooling”. Journal of Turbomachinery, 128, pp. 62–70.

[20] Oke, R.A., Simon, T.W., 2002. “Film Cooling Experiments with Flow Introduced Upstream of a First Stage Nozzle Guide Vane Through Slots of Various Geometries”. ASME Turbo Expo 2002: Power for Land, Sea, and Air (GT2002), Amsterdam, The Netherlands, ASME Paper No. GT2002-30169.

[21] Wright, L.M., Gao, Z., Yang, H, Han, J.C., 2008. “Film Cooling Effectiveness Distribution on a Gas Turbine Blade Platform with Inclined Slot Leakage and Discrete Film Hole Flows”. Journal of Turbomachinery, 130, pp. 071702/1–11.

[22] Rehder, H., Dannhauer, A., 2007. “Experimental Investigation of Turbine Leakage Flows on the Three-Dimensional Flow Field and Endwall Heat Transfer”. Journal of Turbomachinery, 129, pp. 608–618.

14

[23] Piggush, J.D., Simon, T.W., 2007. “Heat Transfer Measurements in a First-Stage Nozzle Cascade Having Endwall Contouring: Misalignment and Leakage Studies”. Journal of Turbomachinery, 129, pp. 782–790.

[24] Timko, L.P., “Energy Efficient Engine High Pressure Turbine Component Test Performance Report”. NASA Report CR-168289

GT2014-26781

Documents

film cooling holes

phantom cooling flow

film cooling ejections

filmcooling research

order cooling

trailing edge increases

airfoil trailing edge

airfoil trailing edge