Overall Effectiveness of a Blade

Endwall With Jet Impingement

Amy Mensch
e-mail: aem277@psu.edu
Karen A. Thole
e-mail: i<thole@psu.edu
Department ot Meciianicai and
Nuclear Engineering,
The Pennsyivania State University,
137 Reber Buiiding,
University Park, PA 16802
and Film Cooling
Ever-incr-easing ther-mal loads on gas turbine components require improved cooling
schemes to extend component life. Engine designers often r~ely on multiple thermal pro-
tection techniques, including intermal cooling and external film cooling. A conjugate heat
transfer model for the endwall of a seven-blade cascade was developed to examine the
impact of both convective cooling and solid conduction through the endwall. Appropriate
parameters wer-e scaled to ensure engine-relevant temperatures were reported. External
film cooling and internal jet impingement cooling were tested separately and together for
their combined effects. Experiments with only film cooling showed high effectiveness
around film-cooling holes due to convective cooling within the holes. Internal impinge-
ment cooling provided more uniform effectiveness than film cooling, and impingement
effectiveness improved markedly with increasing blowing ratio. Combining internal
impingement and external film cooling produced overall effectiveness values as high as
0.4. A simplified, one-dimensional heat transfer analysis was used to develop a prediction
ofthe combined overall effectiveness using results fi-om impingement only and film cool-
ing only cases. The analysis resulted in relatively good predictions, which served to rein-
force the consistency ofthe experimental data. [DOI: 10,1115/1,4025835]

Introduction Relevant Literature

Advances in gas turbine technology for both power generation
and aircraft propulsion are focused on increasing efficiency as
well as maintaining or extending part service life. Even with cur-
rent designs, gas temperatures at the inlet to the turbine often
exceed allowable material temperatures, and internal cooling and
external film cooling are used extensively to cool turbine walls.
Thermal degradation of turbine parts has the potential to cause
major engine problems, giving rise to costly repairs and down-
time. Improved gas turbine efficiency will lead to even higher
thermal loads on the turbine. The ability to evaluate cooling
schemes accurately with regard to meaningful performance met-
rics is crucial.
Ultimately, the parameter of interest is the predicted metal
temperature of the airfoils and associated surfaces. The nondi-
mensional, scaled metal temperature commonly used is the
overall effectiveness, (f>- Many studies report a distinctly dif-
ferent quantity known as the adiabatic effectiveness, ri, repre-
senting the scaled adiabatic wall temperature, which does not
account for internal cooling or heat conduction. Typically, <> /
is determined through numerical simulation of the solid con-
duction using imposed boundary conditions of the parameters:
ri and the internal and external heat transfer coefficients (A,
and h^). Alternatively, (f) can be measured directly provided
the proper experimental scaling, and the metal temperature
can be predicted directly.
As will be discussed in the following sections, more
recent studies have reported conjugate heat transfer effects
and provided the scaled metal temperature. None of these
papers, however, have reported conjugate heat transfer results
for an endwall surface. Turbine endwall heat transfer is
known to be greatly influenced by secondary flow effects;
thus, measurements of conjugate heat transfer for a film-
cooled endwall with internal impingement cooling are of sig-
nificant value.
Contributed by the Heat Transfer Committee of ASME for publication in the
JotJRNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received August
29, 2013; final manuscript received October 22, 2013; published online November
14, 2013. Editor; David Wisler.
Conjugate heat transfer models to determine the overall effec-
tiveness {(j)) have been applied to various geometries, including
flat plates, leading edge models, and vane models. The conjugate
effects of conduction and film cooling for a flat plate were exam-
ined in a computational study by Bohn et al, [1], Bohn et al, com-
pared adiabatic and overall effectiveness for a fiat plate with one
row of 30 deg film-cooling holes. While ri varied laterally across
the plate, (j) was relatively uniform. In a similar study, Na et al,
[2] reported that laterally averaged 0 was sigtiificantly higher than
laterally averaged rj.
Sweeney and Rhodes [3] experimentally simulated a conjugate
surface by matching the engine Biot number, Bi, for a configura-
tion including internal impingement jets and film cooling.
Between the impingement plate and the wall were heat transfer
enhancement features in a Lamilloy® snowflake design. Their
results showed that impingement cooling dominated over film
cooling in the influence on overall effectiveness. The overall
effectiveness distribution was smooth when impingement jets
were tightly spaced, but nonuniformities were observed when the
jet spacing became large, A recent study by Panda and Prasad [4]
involved experimental and computational measurements of over-
all effectiveness for a film-cooled plate with and without internal
impingement. The authors considered conduction through the wall
but did not report what parameter of their model was matched to
an engine. Their results showed that the increase in overall effec-
tiveness from adding internal impingement became more substan-
tial as blowing ratio increased.
Leading edge and vane studies involving both film cooling and
internal impingement cooling also provide insight into conducting
wall configurations, A series of experimental and computational
studies incorporating impingement cooling and film cooling were
performed by the same research group on a leading edge model
[5-10], The leading edge study by Albert et al, [5] showed that it
was essential to match Bi as well as the ratio of intemal-to-exter-
nal heat transfer coefficients to accurately measure overall effec-
tiveness, Maikell et al, [6] measured (f> in the stagnation region
with impingement and round film-cooling holes. They observed
continuous streaks of high 0 around the rows of film-cooling
holes as blowing ratio increased, Dobrowolski et al, [7] performed
conjugate simulations corresponding to the Maikell et al, [6]

Journal of Engineering for Gas Turbines and Power MARCH2014, Vol, 136 / 031901-1
Copyriglit© 2014 by ASME
experiments. The effect of impingement was applied by setting
the intemal wall temperature distribution to the temperature meas-
ured in the experiments. The same leading edge model with
shaped film-cooling holes and impingement was considered by
Mouzon et al. [8] and Ravelli et al. [9]. Their results showed that the
presence of impingement (H/D ~ 5) did not have much effect on <^
because in-hole convection dominated heat transfer in the stagnation
region. Terrell et al. [10] examined the in-hole convection in more
detail. Their experimental and computational study measured the
total heat transfer to the coolant as it passed through shaped holes in
the stagnation region. The heating of the coolant was found to be sig-
nificant and needed to be accounted for in their analyses.
Experimental work of a conducting vane was pioneered by
Hylton et al. [11,12] and Tumer et al. [13] using a C3X vane.
Although engine relevance was limited because matched Bi and
h^lh, were not confirmed, these studies improved understanding
of the thermal fields of a conducting vane and provided experi-
mental data for benchmarking computational work. In a series of
studies [14-18], multiple authors examined different areas of a
C3X vane constructed to match engine Bi. Nathan et al. [14]
measured (p on the stagnation region of the vane, which was
cooled intemally with impingement and extemally with a shower-
head arrangement of cooling holes. On the vane pressure side,
Albert and Bogard [15] found that (j) decreased with increasing
blowing ratio due to jet detachment. Dees et al. [16] and Ledezma
et al. [17] examined (p on the vane suction side experimentally
and computationally, respectively. Intemal cooling was provided
by a u-bend passage inside the vane that fed film-cooling holes.
Impingement cooling was added to the suction side of the vane in
Williams et al. [18]. In addition to measuring </>, the authors tried
to analytically predict <j). They measured r] on an adiabatic vane
and the overall effectiveness with impingement only by blocking
some film-cooling holes in the suction side row. The unblocked
film-cooling holes provided a path for the impingement jets to
exhaust. The prediction performed reasonably well for a range of
momentum fiux ratios.
Although much attention has been paid to conjugate simula-
tions for intemally and extemally cooled vanes, no studies have
been done that consider the endwall surface. Part of the unique-
ness of endwall heat transfer is the presence of secondary passage
fiows based on endwall fiow models, such as that of Langston [19].
The passage vortices that develop along the endwall skew the direc-
tion of endwall fiow and locally increase the extemal heat transfer
coefficient. What distinguishes this work ÍTom previous studies is the
presentation of experimental data for a conjugate endwall with prop-
erly scaled conditions, as discussed in the next section.
Conjugate Endwall Surface
To achieve relevant experimental data for overall effectiveness,
a conjugate surface with appropriate nondimensionless parameters
must be matched between the engine and the experimental model.
Using such a model provides a scaled wall temperature measure-
ment that best represents what would occur in the engine, consid-
ering all of the convection and solid conduction heat transfer. The
experimental endwall model in this study incorporates extemal
film cooling, wall conduction, and intemal impingement jet cool-
ing, thereby including the conjugate effects.
The nondimensional parameters inñuencing the heat transfer
are given in Eq. (1), which is derived from a one-dimensional con-
sideration of the heat transfer from the extemal driving tempera-
ture, expressed as Taw^ to the intemal coolant temperature at the
plenum supplying the impingement cooling, rcintemai [18].
Xr,>l (1)
T'oo — ^cintemal 1 4- B i +
Equation (1) demonstrates the importance of matching Bi and the
ratio of extemal to intemal heat transfer coefficients. A correction
031901-2 / Vol. 136, MARCH 2014
factor for t}, ;(,,, is defined in Eq. (2) to ensure that the normalizing
temperature difference matches that of (/).
^ oo ^ c,exit
Xn = (2)
' 00 ~ -' c,internal
The Xn parameter represents the intemal heating of the coolant
from the supply plenum, at r,..intemai. through impingement and
the film-cooling hole, exiting at T^.exit-
The nondimensional parameters matched for the endwall in this
study relative to those of a typical engine are given in Table I.
The extemal heat transfer coefficient, h^, is enhanced due to pas-
sage secondary flows and has been previously reported for this
cascade by Lynch et al. [20]. Although Lynch et al. [20] measured
/¡•o without film cooling, film-cooling augmentation on the end-
wall is assumed to be minor [21,22]. The intemal heat transfer
coefficient, /Î,, is enhanced with impingement jet cooling. To esti-
mate the average h¡, Nusselt number correlations in the literature
for impingement cooling with and without crossflow [23,24] were
applied. The range of haclh¡ values had some variation with blow-
ing ratio but stayed reasonably close to 1.0. A reasonable Bi range
was achieved by scaling the endwall thickness and using Corian ,
a DuPont material, for the endwall.
Experimental Methods
Measurements of endwall overall effectiveness were obtained
for a linear Pack-B blade cascade using the large-scale, low-speed,
closed-loop wind tunnel depicted in Figs. l(a) and l(b). In this fa-
cility, the fiow was split into three fiow paths. The top and bottom
portions, with dark walls in Fig. l(a), passed through chilled water
heat exchangers, while the middle portion passed through a heater
bank supplying the mainstream flow to the test section. More
details regarding the wind tunnel and now conditioning elements
can be found in Ref. [20]. Mainstream temperatures were meas-
ured by a five-thermocouple rake inserted 0.52Cax upstream of the
blade leading edge at multiple locations across the cascade. The
mainstream temperatures measured were averaged to find T^o.
The maximum variation from the average T^ for any location
was ±0.6 °C.
The secondary flow for the endwall coolant supply was
removed from the top white channel of the wind tunnel and fur-
ther cooled by an auxiliary heat exchanger as shown in Fig. l(b).
The auxiliary heat exchanger circulated a sub-O°C glycol-water
mixture from the auxiliary chiller. To prevent any ice formation
on the heat exchanger fins, a desiccant drier was installed
upstream of the heat exchanger. A laminar flow element (LFE)
measured the total coolant flow rate before the lines split the cool-
ant feed for three separate plenums, which are described in detail
later. The configuration shown in Figs. l(a) and \{b) provided a
mainstream-to-coolant temperature difference of about 40 °C,
resulting in a density ratio (DR) of 1.15. Coolant temperature was
measured by two thermocouples: ~8.7D below the impingement
plate and ~8.7D below the endwall when there was no impinge-
ment plate. The two thermocouples agreed to within ± 3 ° C or
less.
The top view of the test section, shown in Fig. 2, contained a
seven-blade, six-passage linear cascade based on the low-pressure
turbine Pack-B airfoil, a generic geometry that has been used
Table 1 Conjugate endwall parameters
Model Typical engine
Aï. 0.6, 1.0,2.0 1.0-2.0
0.99-1.06 22
Í, cm 1.27 0.20
hJh, 0.4-2.5 1.0
Bi 0.3-0.7 0.27
Transactions of the ASME
 Test
Section

(b)

Chiller

Fig. 1 Depiction of the (a) large-scale, low-speed wind tunnel, with a corner test section housing the Pack-B cascade, and (b)
the coolant loop with auxiliary cooling capability and the inlet flow development section

edge. There was very little variation in Uoo across the cascade, as
the standard deviation over the mean was less than 1%. The test
section inlet boundary layer parameters, also listed in Table 2,
were measured previously by Lynch et al. [20]. Based on the
upstream boundary layer measurements, at the blade inlet plane
Ô/S was 0.071 and the freestream turbulence was 4% [20].
To ensure uniformity and periodicity of the cascade, static pres-
sure taps in the blade midspan were used to measure the pressure
distribution before all experiments. A typical set of pressure coef-
Only ficient, Cp, data is plotted in Fig. 3 as a function of normalized
impingem :nt axial distance for all the blades. The measured Cp agreed well
with the inviscid computational fiuid dynamics (CFD) prediction
[20], confirming flow uniformity for all passages.
The endwall in this study was constructed with a geometric
III configuration and parameters of Bi and h^/h^ relevant to engine
Incoming Film and design. The schematic in Fig. 4(a) shows the generic intemal and
Flow impingement extemal endwall cooling scheme used. Coolant flow is directed
into a stagnant plenum passing around a splash plate. The plenum
feeds an array of 28 staggered holes in an impingement plate,
which feed ten angled film-cooling holes in the endwall. The
diameter, D, is the same for the film and impingement holes.
Film-cooling holes are inclined at an angle of 30 deg relative to
Only film the surface, corresponding to a hole length-to-diameter ratio, L/D,
cooling of 5.8. Figure 4{b) shows the locations of the impingement jets
and film holes. The film-cooling hole inlets are staggered between
the impingement jets. The film-cooling hole exits are oriented to
Fig. 2 Schematic of the Pack-B linear blade cascade with blade
and passage numbering and top view of the conjugate endwall

extensively in the literature [20,25-34]. The operating conditions,

as listed in Table 2, include engine-matched Re and geometric
specifications. The inlet mainstream velocity, U^, was measured CrliFred. ^v^
P -P - ' Blade f ^ ^
.by inserting a Pitot probe 0.52Cax upstream of each blade leading
c= 1 Blade 2 ^ 4
p • Blade 3 \ ^
' -1 Blade 4 ^
Table 2 Flow conditions and blade geometry
> Blade 5 É«

Boundary layer -2 -
:X \ - '' Blade 6 , j ^
Paraineters 2.85C„ • Blade 7 y
Scale factor 8.6 Upstream of blade 4 [20]
Co, 0.218m '-
p/C^ 0.826
2.50 Ô/S 0.061 —1 Í 1 1 1 1 1 i 1 1 1 1 ! 1 1 1 1 l
Inlet Re 1.22 X 10' S*/S 0.0062 0.2 0.4 0.6 0.8
Exit Re 1.98 X 10^ 0/S 0.0046
x/C
Inlet U^ 10.5 m/s Boundary layer 1.34
Inlet, exit flow angles 35 deg, 60 deg shape factor
Inlet, exit Ma 0.029, 0.047 u'/U 0.060 Fig. 3 Pack-B cascade static pressure distribution at the blade
midspan compared to a CFD prediction [20]

Journal of Engineering for Gas Turbines and Power MARCH2014, Vol. 136 / 031901-3

Makstrean
Flow
XTorian
Fig. 4 Schematic of internal and external cooling scheme from
the side view (a) and the top view {b)
align with endwall streaklines obtained using oil flow visualiza-
tion [20], The area outlined in black is used to calculate the area-
averaged (f> presented in the results.
To study the effects of the internal and extemai cooling features
separately as well as together, the endwall was divided into three
sections of two passages. Each section was fed independently by a
separate plenum and incorporated a different cooling arrange-
ment. The center passages, 3 and 4 (see Fig. 2), contained both in-
ternal impingement cooling and extemai film cooling. Passages 1
and 2 had a static plenum and film cooling and no impingement
plate. Passages 5 and 6 had impingement cooling only, which
imposed crossflow. Instead of exhausting the coolant through
film-cooling holes, the coolant flowed out of the channel in the
pitchwise (y) direction from a slot above passage 6. The dark area
of the endwall in Fig. 2 represents the endwall constructed of
Corian material, and the light area represents the rest of the end-
wall constmcted of medium density fiberboard.
Measurement Methods and Uncertainty
The primary measurements in each experiment are the coolant
flow rate used to calculate blowing ratio, the coolant and main-
stream temperatures, and the endwall temperatures. As mentioned
previously, the total coolant flow rate fed to each plenum is meas-
ured with an LFE. For experiments using film cooling, the flow
rate is adjusted to achieve the desired film-cooling blowing ratio.
Blowing ratios reported in this paper reflect the average blowing
ratio of all ten film-cooling holes, hence the use of M^vg. The local
blowing ratio for each film-cooling hole, Mioc, is calculated by
considering the static exit pressure of each film-cooling hole,
031901-4 / Vol. 136, MARCH 2014
1.0
0.9
0,8
0.7
0.6
Film cooling only, L/D = 5.8
Film cooling + impingement,
0.5
= 5.8,H/D = 2.9
0.4 Film cooling + impingement,
0.3 Burd & Simon [36], L/D = 4.6,
p/D = 3, single row
0.2 - B - Barringer et al. [37], L/D = 4.3, _
p/D = 9, single row
0,1 - B - Baninger et al. [37], L/D = 4.3, _
p/D = 3, single row
O.O'---'
1.0000 1.0005 1.0010 1.0015 1.0020 1.0025 1.0030 1.0035 1.0040
Fig, 5 Discharge coefficient measured as a function of pres-
sure ratio compared to Refs. [36] and [37]
measured using pressure taps installed in passage 6. For the cases
with only impingement cooling, the mass fiow rate of coolant is
matched to the total mass flow rate corresponding to the three
Mavg values from film cooling. For a 95% confidence interval, the
uncertainty in coolant fiow rate is estimated to be ± 3 % using the
sequential perturbation method described in Moffat [35].
For film-cooled experiments, the pressure drop can be used to
calculate the average discharge coefficient, C¿, of the film-cooling
holes or the impingement plate plus film holes. Figure 5 shows the
measured values of Q , plotted in filled symbols, as a function of
the average pressure ratio, where F,ot,c. is the plenum pressure
upstream of the impingement plate and P¡oc.<x is the local static
exit pressure of the hole. The results for film cooling only (filled
squares) agree with previous data for 30 deg inclined holes at low
pressure ratios (open squares) [36,37]. The effective Q for the
impingement plate and the film holes (filled circles) is slightly
lower than film cooling only, as expected for an additional flow
restriction.
Steady-state infrared (IR) thermography was used to measure
surface temperatures on the endwall. To maximize the spatial
measurement resolution and take advantage of the scaled-up ge-
ometry, a FLIR P20 IR camera was used to measure endwall sur-
face temperatures. The ceiling of the test section contained 16
removable viewing ports distributed across five blade passages to
allow direct optical access for the IR camera. At each viewing
location, the IR camera was placed perpendicular to the endwall
surface at a distance of approximately 56 cm to acquire images.
With a camera field of view angle of 25 deg and a camera resolu-
tion of 320 X 240 pixels, the resulting image resolution was 1.3
pixels/mm, which equates to 5.7 pixels/D. Thermocouples were
placed in discrete locations on the endwall surface, arranged so
that at least two thermocouples were captured in each image.
When the thermocouples indicated that steady-state was achieved,
five IR images were acquired at each port location. At each loca-
tion, the images were calibrated for emissivity and reflected tem-
perature by minimizing the difference between the thermocouple
readings and the image temperatures. The emissivity was typically
0.92 because all endwall surfaces were painted with flat black
paint. After calibration, the five images were averaged and
exported to an in-house MATLAB program, which assembled the
averaged images from each location. Once a complete endwall
temperature map was obtained, the data were reduced to tp.
The partial derivative method [35] was used to determine the
uncertainty in (f). The largest source of uncertainty comes from the
calibration of the IR images that provide T„. The bias error of T„
was estimated to be, at most, 0.8 °C from the difference of the IR
Transactions of the ASME
 Table 3 Experimental test matrix tion of the coolant temperature: T,. ¡„let for film cooling only versus
T^ciniemai for the cases with impingement cooling. The locations of
Cooling arrangement Measurement these temperatures are shown in Fig, 6, r^,¡ntemai is the coolant
temperature upstream of the impingement plate, r^,¡niet is coolant
Film cooling 2,0 3,5 N/A temperature at the inlet of the film-cooling hole. For the case with
-' Où T^f .inlet 1,0 0.9
no impingement plate, 7^ ¡„let is approximately equal to the stag-
0.6 0,3
Toc -T..,o
nant coolant temperature,
Impingement 2,0 3,5 0,4-1,0
7-^- ' i-,iniemal 1,0 0,9 0,6-1,4
0,6 0,3 1,0-2,2 Film-Cooling Only Results. The overall effectiveness con-
Film and impingement 2,0 3,5 0,5-1,0 tours for film cooling only, ({>/, are presented in Figs, 7(i2)-7(c),
' cx) ^ijntemai 1,0 0,9 0,7-1,6 The ñlm-cooling holes and plenum boundaries are shown for ref-
0,6 0,3 1.1-2.5 erence. Although there is near-stagnant air below the endwall,
conduction caused (¡)f to be above zero along the pressure side of
the passage, In-hole conduction effects are apparent from the
T„ increased (/y upstream ofthe film-cooling hole exits. Also, there is
evidence of cross passage flow sweeping film coolant to the suc-
T. -~
tion side of the passage in all cases. From the contours, the film-
•^ Endwall cooling jets appear lifted off the surface for Mavg= 1,0 and 2,0,
t while most of the jets remain attached to the endwall for
>
Corian*, k„
Mavg = 0,6, Laterally averaged ct>f for film cooling only is plotted
^c,mlet JJ ^ T^,i^ as a function of axial distance in Fig, 7(<f), The (j)f is almost the
>f
same for M^vg = 0,6 and 1,0, There is a peak around x/Cax = 0,15,
1
Impiiïiement plate coinciding with the first row of film-cooling holes, and a dropoff
after x/Ca_x > 0,55 near the last film-cooling hole. Local increases
are apparent for Mavg = 2,0 at the first row of holes, as well as in
the downstream parts of the passage, after x/C^, > 0,3,
Fig. 6 Coolant and wall temperatures of the conjugate wall
with film and impingement cooling
Impingement Cooling Only Results. For impingement cool-
image temperatures fj-om the thermocouples. The precision error ing only, the (¡>„ contours with impingement hole locations are
of T„ was estimated to be 0,3 °C from the standard deviation of shown in Figs, 8(Ö)-8(C), Figure 8(a) specifies direction of coolant
the five images. Using a confidence interval of 95%, the total exiting the impingement channel under the endwall for the case of
uncertainty in measurement of (f) was estimated to be ±0,02, impingement only. Compared to film cooling at the same fiow
rate, the peak values of 4>g are lower than ij)f, but the distribution
of (¡)g is more uniform than 4>f. The infiuence of the external pas-
sage flow can be observed in higher effectiveness near the blade
Results and Discussion suction side compared to the blade pressure side. As hot external
Overall effectiveness was measured for the three cooling fiow travels across the middle of the passage, the cooler endwall
arrangements at three different blowing ratios: Mavg = 0,6, 1,0, reduces the temperature of the fiow. The cooler passage flow
and 2,0 as listed in Table 3, The momentum flux ratio, /avg> and carries less heat to the downstream suction side of the
the associated /JOO/A, are also shown. As mentioned previously, /¡oo passage, resulting in cooler wall temperatures. Laterally averaged
was measured and reported by Lynch et al, [20] and hi was esti- 4>g is plotted as a function of axial distance in Fig, S{d). The high-
mated from correlations [23,24], In the second column of Table 3, est values of 0^ occur in the middle part of the passage and
the nondimensional temperature that was measured for each cool- decrease after x/C_a_y >p,5, near the last row of impingement jets.
ing arrangement is defined. The definitions in Table 3 are impor- In contrast to (j)j:, (ßg increases with blowing ratio due to
tant to note for the analyses that will be described later in the the corresponding increase in impingement jet flow rate and
paper. The differences in overall effectiveness occur in the nota- internal heat transfer. Average impingement Nusselt numbers

^œ ^ w,f
0 0.05 0.1 0.15 0,2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 T -T
^00 '- cinlet

""•"'Film cooling, M,,g = 0,6

-*-Film cooling, M„g = 1,0
""""Film cooling, M = 2,0 .

Fig. 7 Contours of <^,for blowing ratios: (a) Mavg = 0,6, (b) Afavg = 1.0, (c) /Wa^g = 2.0, with 30 deg inclined holes and ple-
num boundaries overlaid, and (d) pitchwise, laterally averaged (^/plotted as a function of axial distance

Journal of Engineering for Gas Turbines and Power MARCH2014, Vol, 136 / 031901-5
 o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95 1 T„-T. cjntental

- 0.6, Nu = 3.9
= 1.0, Nu = 6.1
= 2.0, Nu = 8.9

(a)

Fig. 8 Contours of <^o for blowing ratios: (a) Mavg = 0.6, (b) Mavg = 1 -0, (c) Aigvg = 2.0, with 90 deg
Impingement holes and plenum boundaries overlaid, and (d) pitchwise, laterally averaged <^o plotted as
a function of axial distance

o 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9 0.95
T - cjjitema]

M „ , = 0.6,1.^ =
DR =1.13

"Film & imping., M^^j^^ 0.6

-Film & imping., Mgyg" ^-^
••Film & imping., M,„j,= 1.0, rep^
-Film & imping., M,„,„= 2.0

1.2 0.0 0.2 0.4 0.6 0.8 1.0

(<•) (b) (c) (d)

Fig. 9 Contours of 4> for blowing ratios: (a) Afavg = 0.6, (d) Mavg = 1-0, (c) Afavg = 2.0, with 30 deg inclined
film holes, 90 deg impingement holes, and plenum boundaries overlaid, and (cQ pitchwise, laterally aver-
aged (j) plotted as a function of axial distance

estimated from literature correlations [24] are listed in the
legend in Fig. 8{d). A similar trend for impingement cooling
only was noted for the vane in Williams et al. [18].
Film and Impingement Cooling Results. Overall effective-
ness for intemal impingement and film cooling, (j), was measured
in passages 3 and 4 in the same experiment. The contours are pre-
sented in Figs. 9(a)-9(c) with the impingement and film-cooling
hole locations. Boundaries just below the blades are shown, which
prevented coolant from crossing from one passage to another in
the channel above the impingement plate. Although there is high
(p throughout much of the passage, (j> varies significantly and is
not uniform. The primary effect of film cooling, conduction within
the film-cooling holes, resulted in high (^ around film holes. The
most noticeable blowing ratio effect is at Mavg = 2, with increased
cooling particularly from convection in the holes.
Figure 9{d) shows laterally averaged cj) from passage 4 plotted
as a function of axial distance. Similar to film cooling only, there
is a peak at x/C„^ = 0.15 around the first row of film holes and a
decrease in cf) after x/Ca^ > 0.55. The influence of impingement
can be observed in the increase of (j) with blowing ratio. Figure
9(d) also shows a repeat experiment performed for film and
impingement at Mavg = 1.0. Good agreement between the two
experiments for Mavg = 1.0 as well as both passages in Figs.
9(a)—9(c) supports the repeatability and reproducibility of the
results.
Comparison of Individual and Combined Cooling Effects.
The data presented in Figs. 7-9 are summarized in Fig. 10, in
which area-averaged overall effectiveness is presented as a func-
tion of Mavg. Area-averaged (p for film and impingement increases
from about 0.3 to 0.4 with an increase in Mavg from 0.6 to 2.0.
Consistent with the laterally avçfaged data, an increase in blowing
ratio improves area-averaged cpg more than (pj-. Figure 10 also
highlights the improved cooling that results from adding
impingement. From this perspective, the increase^ of ^ from ^y is
large^for higher blowing ratios. In other words, cj) increases faster
than (j)j-. This trend was also observed for a flat plate in Panda and
Prasad [4]. Altematively, the effect of adding film cooling to an

031901-6 / Vol. 136, MARCH 2014 Transactions of the ASME

 0.7 u. /
IM^^g = 0.6 ~^"~Film cooling, ^^
"Area Avg. Film Cooling, (j)f 0.6
0.6 " x/C,^ = 0.22 "**"" Impingement, ^^
"AreaAvg. Impingement, <|)„
"•"Film and impingement,(j)
"Area Avg. Film and Impingement, 0.5 -
0.5

0.4
0.4 1
0.3
0.3
0.2
0.2
0.1 \-
0.1
A n 1 . . .

-0.2 0.0 0.2 0.4 0.6 0.8

0.0
0.5 1.0 1.5 2.0 y/p
M.,
Fig. 12 Overall effectiveness of all cooling configurations at
Fig. 10 Area-averaged (p (using area outlined in Fig. 4) plotted Mavg = 0.6 plotted as a function of ¡ríp at x/C^x = 0.22
as a function of blowing ratio for all three cooling
configurations
to the film-cooling jets can be seen in the cases with film cooling.
For impingement cooling, (j)„ has a relatively flat distribution from
impingement-cooled plate indicates a smaller benefit, especially y/p = 0.2 to 0.55, corresponding to the impingement jet locations.
for Mavg = 2.0. For film cooling, there is a peak in (/y and cf> around y/p = 0.25,
The laterally averaged effectiveness at Mavg = 1 -0 for film cool- corresponding to one of the film-cooling holes in the diagonal
ing only, impingement only, and film and impingement are plotted row. The drop after this peak, especially for Mavg = 2.0, is attrib-
together in Fig. 11. Laterally averaged 4) for the combined cooling uted to detachment of the film-cooling jets from the vertical row
scheme is higher than ip^ or <^„ from x/Cax = 0.05 around the inlets of holes. Figures 12-14 also show that (/>„ is closest to (/y at
of the film-cooling holes to x/Cax = 0.35 just before the last row of Mavg = 0.6, but (po becomes closest to (j) at Mavg = 2.0.
impingement holes. However, (j)g is not much lower than (fi,
indicating that the effectiveness is dominated by impingement
cooling. Upstream and downstream, x/Cax- < 0.05 and > 0.45, (j)„
is higher than 4>. This unexpected result for impingement only
Calculation of Combined Effectiveness Based on
occurs because, after impingement, coolant flows away from the Individual Cooling Features
middle of the passage and convectively cools the intemal wall in Noting the apparent influence of film cooling and impingement
the region x/Cax < 0.05 and > 0.45. With combined film and on the combined overall effectiveness, an analysis was performed
impingement, the coolant is instead directed into the film-cooling that uses the measured values of 4>f and (pg to calculate the com-
holes. bined film cooling and impingement effectiveness, (p. Comparing
Just downstream of the first row of ñlm-cooling holes, at axial the results of such an analysis to measured 0 confirms the consis-
distance x/C^x = 0.22, 0 is plotted across the passage as a function tency of the data acquired in the different sections of the cascade.
of normalized distance, y/p, in Figs. 12-14. Peaks and valleys due Applying a one-dimensional analysis and reasonable assumptions
to Üie heat transfer of the combined film and impingement-cooled
0.7
0.7
"Lat. Avg. Film Cooling, ^¡ ^avg ~ I
= 1.0 ""*~Film cooling, (j)f
0.6 " " Lat. Avg. Impingement, ij)^
0.6 = 0.22 -»<—Impingement, (|)„
"Lat. Avg. Film and Impingement, (|)
0.5 - "•""Film and impingement,'
"Lat. Avg. Film and Impingement, ^, repeat
0.5

0.4
0.4

0.3
0.3

0.2
0.2

0.1
0.1

0.0
-0.2 0.0 0.2 0.4 0.6 0.8 1.0 0.0
-0.2 0.0 0.2 0.4 0.6 0.8
y/p

Fig. 11 Pitchwise, laterally averaged (^ plotted as a function of Fig. 13 Overall effectiveness of all cooling configurations at
axial distance for the three cooling configurations at Mavg = 1-0 Mavg = 1-0 plotted as a function of yip at x/C^x = 0.22

Journal of Engineering for Gas Turbines and Power MARCH2014, Vol. 136 / 031901-7
 0.7
2.0 "Film cooling, ^f
0.6 Endwall
0.22 ' Impingement, ^„ t
Corian®, k,.
"Fihn and impingement,(j)
0.5 > •^ T„ j . ^ Exitingv
H ='"
"• coolant
1 1 > II II
0.4 j Irrtoingement ptete i
T
0.3 c.mtemal

Fig. 16 Cooiant and wali temperatures of the conjugate wail

0.2 with impingement cooiing oniy

0.1
' film ^ r,inlet
% • = • (4)
0.0
-0.2 0.0 0.2 0.4 0.6 0.8
y/p From Eq. (4), an expression for nondimensional Tfiin, can be found
in terms of measured quantities, including T^.f, expressed as 4>f.
Fig. 14 Overaii effectiveness of aii cooiing configurations at An expression for the nondimensional T^, is found in a similar
JVavg = 2.0 piotted as a function of yip at xlCgx = 0.22 manner, considering the one-dimensional heat transfer through an
impingement-cooled wall, shown in Fig. 16. The heat flux through
the endwall with impingement cooling can be written as
conducting wall, an equation can be written for (j> in terms of (j)f
and (!>„. First, using the notation presented in Fig. 6, the following
expression for the heat flux, q, can be written: : ^ |

(5)

•'film I W,i
(3) which can be rearranged into a nondimensional r„,, in terms of
measured quantities, including r„, „, expressed as 4>g.
Using Eqs. (3)-(5) and the definitions of 4>f and (¡>o, Eq. (6) can
where Tfün, is used to represent the effective driving temperature be written for the calculated 4> with film and impingement
for external convection and T^.¡ is the internal wall surface tem- cooling.
perature. Equation (3) can be used to solve for the wall tempera-
ture, T„, and the remaining parameters are either known (h^ and
k^) or can be estimated from measurements of 4>f and 0„ (Tnim Bi
(6)
and r„,,). Estimating T^jm from the results of (/»y- assumes that the 1 +B\ ^c.intemal '^
external driving temperature is not strongly affected by the pres-
ence of impingement. Likewise, estimating r„., from the results of The quantity, fj, is the internal coolant warming factor necessary
(/)o assumes that the internal wall temperature is not strongly to correct for the difference in normalizing coolant temperatures
affected by the presence of film cooling. This second assumption for 0 and ^f (see definitions in Table 3). The internal coolant
is supported by the observation that even the external wall temper- wanning factor, Xf^ accounts for the coolant warming through in-
ature, expressed as cp, appeared to be more influenced by impinge- ternal impingement, similar to x»; accounting for the coolant
ment than by fllm cooling, because (po was almost as high as (f). wanning thrçugh in-hole convection in Ref. [18]. The coolant
Gathering the nondimensional Tfn„^ from <pf requires another
one-dimensional equation for the heat transfer through a film-
0.7 -T—r—1—1—1—1—n—1—] ' 1 1 ' 1 ' ' ' ' 1 1 1 1 1 1 ' 1 1
cooled conducting wall. Figure 15 shows that, with film cooling
only, the internal cooling is approximated as strictly conduction ^ ^ M , , g = 1.0, (t>nieas - ^ M , , g = 2AKeas
within the coolant air from the location of the thermocouple at 0.6
~8.7D below the endwall. The plenum is nearly stagnant and nat- ""Mavg • ~M,^g = 2.0,(|)^,
ural convection is not expected due to the stable coolant tempera-
ture distribution. Since the conductivity of air is very low, the heat 0.5
— -^ "~»
gained by the coolant due to the internal heat transfer is neglected
and the thermocouple measurement is estimated as ~r(. ¡niet. Using
0.4
X
Fig. 15, the heat flux through the film-cooled endwall, qf, is
defined in Eq. (4).
0.3 1 ^
0.2 -

Endwall _ _
Corian*, k^v 0.1 - M3vg =

»cjnltt . 1 , . 1 , 1 . 1 1 1 < 1 1 1 1 1 1 . I I I . 1 .
0.0
1 1 1 . 1 I . 1 I

L~8.7D
Coolant -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Fig. 15 Coolant and waii temperatures of the conjugate wail Fig. 17 Comparison of laterally averaged <^caic and ^meas P'ot-
with fiim cooiing oniy ted as function of axiai distance for aii three biowing ratios

031901-8 / Vol. 136, MARCH 2014 Transactions of the ASME

temperature increase due to impingement, between the plenum at more deviations from the analyses due to impingement flow
^c.intemai and the inlet to the film-cooling hole at T^i^j^i, can be differences.
estimated from a total energy balance on the case of impingement The current study demonstrates the conjugate methodology
only. Assuming the heat gained by the coolant comes from the and overall trends for a generic impingement and film-cooled
endwall only, ^(.,¡„1« can be estimated as the temperature of the endwall, regardless of the specific airfoil geometry. These
coolant exiting the impingement channel in Fig, 16, leading to results demonstrate the importance of testing internal and
Eq, (7), external cooling schemes together to capture conjugate effects.
There were differences in impingement-cooling and film-
^ OO ^ c,inlet cooling behavior when the two were combined. Measurements
Z/=; (7) of conjugate heat transfer on an endwall fill a gap in overall
3 ^ c,intemal ( J oo ^ c.inte
effectiveness data for gas turbine endwalls. This study also
provides a valuable data set to benchmark future conjugate
The values of Xf were 0,75 for Mavg = 0,6, 0,82 for Mavg= LO,
simulations.
and 0,88 for Mavg = 2,0,
When Eq, (6) is applied, the laterally averaged (p^nic (across
the area outlined in Fig, 4) are calculated for the combined Acknowledgment
film-cooling and impingement-cooling cases and plotted in The authors would like to acknowledge support from the U,S,
Fig, 17, The calculated values agree very well with those Department of Energy (DOE), National Energy Technology Labo-
measured at Mavg = 0,6, Although agreement for all blowing ratory (NETL) through the University Turbine Systems Research
ratios is relatively good in the upstream parts of the passage,
program. Any opinions, findings, conclusions, or recommenda-
more deviation occurs downstream as blowing ratio increases.
tions expressed herein are solely those of the authors and do not
The overpredictions downstream are probably an effect of
necessarily reflect the views of the DOE, The writers would like
higher measured effectiveness for impingement only than for
to thank Mark Zelesky of Pratt & Whitney, Dr, David Bogard of
film and impingement after A / C „ , > 0 , 4 5 , seen in Fig, 11, This
was attributed to differences in the impingement flow with the University of Texas, and Robin Ames of DOE-NETL for their
and without film cooling present. With impingement only, the continued communication and support regarding this research.
impingement jets are exhausted laterally, as shown in Fig,
8(a), and the resulting channel fiow cools the downstream in- Nomenclature
ternal wall more than when the coolant is directed into the A = area
film-cooling holes. Additional sources of deviation arise from Bi = Biot number {h^jlky,)
the assumptions made in the analyses, including using the Cax = axial chord length
same h^ for cases with and without film cooling and assum- Cd = discharge coefficient
ing the coolant air below the film-cooling only endwall was Cp = specific heat
stagnant. The literature has shown that external heat transfer Cp = pressure coefficient
coefficient can increase with the addition of film cooling, and D = hole diameter
this effect increases with blowing ratio [21,22], An underesti- DR = density ratio (pjpoo)
mated /¡oo would result in a higher i^caic^ consistent with the h = convective heat transfer coefficient
results of the calculation in Fig, 17, H = gap height
Overall, the favorable comparison between the individual cool- / = momentum fiux ratio iPcUc^/PacUao^)
ing features and the overall impingement plus film cooling is well k = thermal conductivity
predicted. These results provide insights as to the contribution of L = length
each for a conjugate endwall. M = blowing ratio (pcUJPooUoo)
m = mass fiow rate
Ma = Mach number
Conclusions Nu = Nusselt number (/¡D/kfluij)
p = pitch length
Overall effectiveness data were presented for an endwall with P = pressure
external film cooling and internal impingement jet cooling. These q = heat flux
two cooling mechanisms were examined separately and together Re = Reynolds number {PoJJ^CaJßoo)
at three different film-cooling blowing ratios. 5 = blade span
The results indicated that, with film cooling alone, higher effec- f = thickness
tiveness Values were measured near the film-cooling holes due to T = temperature
convection within the holes. Increasing blowing ratio increased U = streamwise velocity
the cooling effectiveness near the hole exits but did not signifi- u' = fluctuating velocity
cantly improve effectiveness elsewhere in the passage, given the x,y,z= global coordinates, where x is blade axial direction
jets separated from the surface. Alternatively, impingement cool-
ing showed clear improvement with increasing coolant flow rates.
In addition, the effectiveness distribution for impingement cooling Greek Symbols
was very uniform. Ô = boundary layer thickness
Combining internal and external cooling produced higher ¿* = displacement thickness
overall effectiveness compared to impingement cooling or film r] = adiabatic effectiveness (T^—Tay„)/(T^—Tc,exit)
cooling alone. The passages cooled with impingement and 6 = momentum thickness
film cooling showed high effectiveness near the holes and H = dynamic viscosity
increasing effectiveness with blowing ratio. Effectiveness p = density
results from film cooling alone and impingement cooling (j) = overall effectiveness {Taa-Ty^)/iT^-T,.i„temA)
alone were used to calculate the effectiveness of the com- (¡>f= overall effectiveness with film cooling
bined cooling scheme. The calculations compared reasonably f
well to the measured data using a simple one-dimensional 4>g = overall effectiveness with impingement
heat transfer analysis. The agreement was best for Mavg = 0,6 v-* oc * w.o)'\^ oo ^ c.intemal/
and 1,0, As the blowing ratio increased, the increasing influ- Xf = internal coolant warming factor for cßf
ence of internal impingement on overall effectiveness caused (^ oo ^6,inlet)/(^ oo ^c.intemal)

Journal of Engineering for Gas Turbines and Power MARCH2014, Vol, 136 / 031901-9
 X^ = intemal coolant warming factor for //
\^ oo ' c.e\'it)/\l CO ^ c.intemalJ
Subscripts, Accents
avg = average
aw = adiabatic wall surface
c,exit = coolant at film-cooling hole exit
c,inlet = coolant at film-cooling hole inlet
c,intemal = coolant upstream of impingement plate
cale = calculated
/ = intemal
/ = film cooling only
film = extemal driving temperature location
loc = local
meas = measured
o = impingement cooling only
s = static
tot = total
w = wall
TO = mainstream or extemal
0 = laterally averaged
0 = area averaged
References
tlJ Bohn, D., Ren, J., and Kusterer, K., 2003, "Conjugate Heat Transfer Analysis
for Film Cooling Configurations With Different Hole Geometries," ASME Pa-
per No. GT20O.3-38369.
[2] Na, S., Williams. B., Dennis, R. A., Bryden. K. M., and Shih, T. I.-P., 2007,
"Intemal and Film Cooling of a Flat Plate With Conjugate Heat Transfer."
ASME Paper No. GT2007-27599.
[3] .Sweeney, P. C, and Rhodes, J. F., 2000. "An Infrared Technique for Evaluating
Turbine Airfoil Cooling Designs," ASME J. Turbomach., 122(1), pp. 170-177.
[4] Panda, R. K., and Prasad, B. V. S. S. S., 2012. "Conjugate Heat Transfer From
a Flat Plate With Combined Impingement and Film Cooling," ASME Paper No.
GT2012-68830.
[5] Albert. J. E., Bogard, D. G., and Cunha, F., 2004, "Adiabatic and Overall Effec-
tiveness for a Film Cooled Blade." ASME Paper No. GT2004-53998.
[6] Maikell. J., Bogard, D.. Piggush. J.. and Kohli, A.. 2011. "Experimental Simu-
lation of a Film Cooled Turbine Blade Leading Edge Including Thermal Barrier
Coating Effects," ASME J. Turbomach., 133(1), p. 011014.
[7] Dobrowolski, L. D., Bogard, D. G., Piggush. J.. and Kohli, A., 2009,
"Numerical Simulation of a Simulated Film Cooled Turbine Blade Leading
Edge Including Conjugate Heat Transfer Effects," ASME Paper No.
IMECE2009-11670.
18J Mouzon. B. D., Terrell, E. J.. Albert, J. E., and Bogard, D. G., 2005, "Net Heat
Flux Reduction and Overall Effectiveness for a Turbine Blade Leading Edge,"
ASME Paper No. GT2005-69002.
[9] Ravelli. S., Dobrowolski, L., and Bogard, D. G., 2010, "Evaluating the Effects
of Intemal Impingement Cooling on a Film Cooled Turbine Blade Leading
Edge," ASME Paper No. GT2010-23002.
[10] Terrell, E. J.. Mouzon, B. D., and Bogard, D. G., 2005, "Convective Heat
Transfer Through Film Cooling Holes of a Gas Turbine Blade Leading Edge,"
ASME Paper No. GT2005-69003.
1111 Hylton, L. D.. Miheic, M. S., Tumer, E. R.. Nealy, D. A., and York. R. E.,
1983, "Analytical and Experimental Evaluation of the Heat Transfer Distribu-
tion Over the Surfaces of Turbine Vanes," Paper No. NASA-CR-168015.
[12] Hylton, L. D., Nirmalan, V., Sultanian, B. K., and Kauffman, R. M., 1988, "The
Effects of Leading Edge and Downstream Film Cooling on Turbine Vane Heat
Transfer." Paper No. NASA-CR-182133.
[13] Tumer, E. R., Wilson, M. D., Hylton. L. D.. and Kauffman. R. M., 1985.
"Turbine Vane Extemal Heat Transfer. Volume I : Analytical and Experimental
Evaluation of Surface Heat Transfer Distributions With Leading Edge Shower-
head Film Cooling," Paper No. NASA-CR-174827.
[14] Nathan, M. L., Dyson. T. E., Bogard. D. G.. and Bradshaw, S. D.. 2013.
"Adiabatic and Overall Effectiveness for the Showerhead Film Cooling of a
Turbine Vane." ASME J. Turbomach., 136(3), p. 031005.
031901-10 / Vol. 136, MARCH 2014
[15] Albert, J. E., and Bogard. D. G., 2013, "Measurements of Adiabatic Film and
Overall Cooling Effectiveness on a Turbine Vane Pressure Side With a
Trench," ASME J. Turbomach., 135(5), p. 051007.
[16] Dees, J. E., Bogard, D. G., Ledezma, G. A., and Laskowski, G. M., 2013,
"Overall and Adiabatic Effectiveness Values on a Scaled Up, Simulated Gas
Turbine Vane," ASME J. Turbomach., 135(5), p. 051017.
[17] Ledezma. G. A.. Laskowski. G. M., Dees, J. E., and Bogard. D. G.. 2011,
"Overall and Adiabatic Effectiveness Values on a Scaled Up Simulated Gas
Turbine Vane; Part II—Numerical Simulation," ASME Paper No. GT2011-
46616.
[18] Williams, R. P., Dyson, T. E., Bogard. D. G., and Bradshaw, S. D., 2013,
"Sensitivity of the Overall Effectiveness to Film Cooling and Intemal Cool-
ing on a Turbine Vane Suction Side," ASME J. Turbomach., 136(3), p.
031006.
[19] Langston, L. S., 1980, "Crossflows in a Turbine Cascade Passage," ASME J.
Eng. Power, 102(4), pp. 866-874.
[20] Lynch, S. P., Sundaram. N., Thole, K. A., Kohli, A., and Lehane, C , 2011,
"Heat Transfer for a Turbine Blade With Nonaxisymmetric Endwall Con-
touring," ASME J. Turbomach., 133(1), p. 011019.
[21] Eriksen, V. L., and Goldstein, R. J., 1974, "Heat Transfer and Film Cooling
Following Injection Through Inclined Circular Tubes," ASME J. Heat Transfer,
96(2), pp. 239-245.
[22] Dees, J. E., Bogard, D. G., and Bunker, R. S., 2010, "Heat Transfer Augmenta-
tion Downstream of Rows of Various Dimple Geometries on the Suction Side
of a Gas Turbine Airfoil." ASME J. Turbomach., 132(3). p. 031010.
[23] Hollworth. B. R., and Dagan, L., 1980, "Arrays of Impinging Jets With Spent
Fluid Removal Through Vent Holes on the Target Surface—Part 1: Average
Heat Transfer," ASME J. Eng. Power, 102(4), pp. 994-999.
[24] Florschuetz, L. W., Truman. C. R.. and Metzger, D. E., 1981, "Streamwise
Flow and Heat Transfer Distributions for Jet Array Impingement With Cross-
flow." ASME J. Heat Transfer, 103(2), pp. 337-342.
[25] Praisner. T. J., Allen-Bradley, E.. Grover, E. A., Knezevici, D. C , and Sjo-
lander, S. A., 2007. "Application of Non-Axisymmetric Endwall Contouring to
Conventional and High-Lift Turbine Airfoils," ASME Paper No. GT2007-
27579.
[26] Knezevici, D. C , Sjolander. S. A., Praisner, T. J., Allen-Bradley, E., and
Grover. E. A.. 2010, "Measurements of Secondary Losses in a Turbine Cascade
With the Implementation of Nonaxisymmetric Endwall Contouring," ASME J.
Turbomach., 132(1), p. 011013.
[27] Praisner, T. J., Grover, E. A., Knezevici, D. C, Popovic, I., Sjolander, S. A.,
Clark, J. P., and Sondergaard, R., 2008. "Toward the Expansion of Low-Pres-
sure-Turbine Airfoil Design Space," ASME Paper No. GT2O08-5O898.
[28] Lake, J., King, P., and Rivir, R., 1999, "Reduction of Separation Losses on a
Turbine Blade With Low Reynolds Numbers," AIAA Aerospace Sciences
Meeting. Reno, NV, January 11-14, AIAA Paper No. 99-0242.
[29] Murawski, C. G.. and Vafai, K., 2000, "An Experimental Investigation of the
Effect of Freestream Turbulence on the Wake of a Separated Low-Pressure Tur-
bine Blade at Low Reynolds Numbers," ASME J. Fluids Eng., 122(2), pp.
431^33.
[30] Mahallati, A., McAuliffe, B. R., Sjolander, S. A., and Praisner, T. J., 2007,
"Aerodynamics of a Low-Pressure Turbine Airfoil at Low-Reynolds Numbers:
Pan 1—Steady Flow Measurements." ASME Paper No. GT2007-27347.
[31] Zoric, T., Popovic, I., Sjolander, S. A., Praisner. T., and Grover, E., 2007,
"Comparative Investigation of Three Highly Loaded LP Turbine Airfoils: Part
I—Measured Profile and Secondary Losses at Design Incidence," ASME Paper
No. GT2007-27537.
[32] Popovic, I.. Zhu. J., Dai. W.. Sjolander, S. A., Praisner, T., and Grover, E..
2006. "Aerodynamics of a Family of Three Highly Loaded Low-Pressure Tur-
bine Airfoils: Measured Effects of Reynolds Number and Turbulence Intensity
in Steady Flow," ASME Paper No. GT2006-9127I.
[33] Lynch, S. P., Thole. K. A., Kohli. A., and Lehane, C , 2011, "Computational
Predictions of Heat Transfer and Film-Cooling for a Turbine Blade With Non-
axisymmetric Endwall Contouring." ASME J. Turbomach., 133(4), p. 041003.
[34] Lawson, S. A., Lynch, S. P., and Thole, K. A., 2013, "Simulations of Multi-
pha.se Particle Deposition on a Nonaxisymmetric Contoured Endwall With
Film-Cooling," ASME J. Turbomach., 135(3). p. 031032.
[35] Moffat, R. J., 1988, "Describing the Uncertainties in Experimental Results,"
Exp. Therm. Fluid Sei., 1( 1 ), pp. 3-17.
[36] Burd. S. W.. and Simon, T. W., 1999, "Measurements of Di.scharge Coefficients
in Film Cooling." ASME J. Turbomach., 121(2), pp. 243-248.
[37] Barringer, M. D., Richard, O. T., Walter, J. P.. Stitzel. S. M., and Thole. K. A.,
2002, "Flow Field Simulations of a Gas Turbine Combustor," ASME J. Tur-
bomach., 124(3), pp. 508-516.
Transactions of the ASME
Copyright of Journal of Engineering for Gas Turbines & Power is the property of American
Society of Mechanical Engineers and its content may not be copied or emailed to multiple
sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.