The document describes a computational fluid dynamics (CFD) analysis of heat transfer in a rectangular channel with blockages containing elongated holes. A model of the channel was created in design software and meshed before performing CFD simulations. The simulations analyzed how heat transfer coefficients and Nusselt numbers varied with the pitch ratio of the blockages for different Reynolds numbers. Results showed that heat transfer initially increased with pitch ratio up to a value of 6.0, then decreased with further increases, due to changes in flow reattachment. Blockages with shorter holes enhanced heat transfer more but also increased pressure drops more compared to longer holes. The CFD results were also validated against available experimental data, with good agreement.
2. CFD Analysis and Enhancement of Heat Transfer in Rectangular Channel Using Blockage
with Elongated Hole
http://www.iaeme.com/IJMET/index.asp 41 editor@iaeme.com
1. INTRODUCTION
Gas turbine vanes and blades are exposed to high air temperature. And increasing
turbine rotor inlet temperature directly leads to a rise in thermal efficiency and output
power of gas turbines. The rotor inlet temperature is far higher than the melting point
of the blade and vane material. Hence, cooling technology of turbine blades has
become one of the most important key factors for improvement of gas turbine engine
efficiency since 1970s. According to Takeishi [1992] [4], cooling technology enables
turbine inlet temperature to increase by 25 °C per year but the achievement of super-
alloy development technology limits by 10 °C per year. Thus to study cooling
technology is more effective than to develop higher thermal resistance materials for
gas turbine improvement. Gas turbine blades are cooled by the air directly extracted
from the engine compressor. This extracted air causes a disadvantage of thermal
performance by incurring pressure drop. Hence an optimized cooling technique is
needed considering operating conditions. Internal cooling can be performed by
passing the extracted air from the compressor through several serpentine passages
inside the blades moving out the heat from the blades. For the gas turbine blades rib
turbulated cooling, pin-fin cooling, and impingement cooling are applied in the
blade’s internal coolant passage in order to remove heat from the blade inside. Figure
1(a) presents the commonly used turbine blade internal cooling techniques. Moon and
Lau [2003] [2] measured pressure drop and heat transfer coefficient by the liquid
crystal technique on the rectangular duct with perforated walls. They showed that the
number of walls and the configuration of holes did not affect the heat transfer level.
Their results also showed that the smaller holes could increase heat transfer
coefficient but pressure drop also greatly increased.
Figure 1 (a) External Film Cooling (b) Internal Convective Cooling
Lau et al. [2003] [5] examined the heat transfer and pressure drop on a rectangular
duct with perforated walls equipped with staggered holes. They showed that walls
with circular holes and square holes increased heat transfer but the increase in
pressure drop was much severe. They concluded that the shape and size of holes
should be optimized in order to get better thermal performance. In the early period of
gas turbine engines, only jet impingement cooling method was used for the leading
edge cooling of blade. Jet impingement cooling is to cool the blade by air
3. S. Kandwal and Dr. S. Singh
http://www.iaeme.com/IJMET/index.asp 42 editor@iaeme.com
impingement on the surfaces of serpentine passages. Impinging was the most effective
cooling method for the leading edge making other methods unnecessary. But as gas
turbine technology has been developed, the rotor inlet temperature has been also
increased for turbine engine efficiency. Therefore, internal cooling technologies using
turbulators such as ribs and pin-fins have been developed [8].
1.1. Analytical Solution
The hydraulic diameter of test channel is calculated as:
The average Nusselt numbers for each of the three wall segments of the channel
Between two blockages was calculated as:
= × ℎ
The average Nusselt numbers were normalized using reference Nusselt number for
fully developed turbulent flow in the channel with smooth walls. This reference
Nusselt number was defined as:
Nu0 = 0.023Re0.8
Pr0.4
Where the averaged heat transfer coefficients were defined as:
ℎ =
× ×( )
×( )
Where, Tb = Average bulk mean temperature =
( )
Pressure drop across the blockages is calculated as:
∆P = g × y1 × (ρw ₋ ρa)
1.2. Modeling and Simulation
AcuSolve a computer program, based on finite Element Method (FEM) is one of the
powerful packages of existing commercial software for solving fluid flow and heat
transfer problems.
Figure 2 Design of Physical Model in ALTAIR®
HYPERWORKS (11.0)
4. CFD Analysis and Enhancement of Heat Transfer in Rectangular Channel Using Blockage
with Elongated Hole
http://www.iaeme.com/IJMET/index.asp 43 editor@iaeme.com
The purpose of this study is to visualize the performance of air duct by providing
the artificial roughness in the form of blocks, as well as design of artificial roughness
geometry which will give optimum performance. CATIA V5 was used to create
geometries and ALTAIR HYPERMESH used to generate the unstructured mesh. A
schematic of the geometric model of the channel used in the study is shown in above
Figure.
Air flow inside a three-dimensional (3-D) channel object, with the Assumption
that the heat transfer to the atmosphere is negligible. The fluid properties were set to
be similar to that used in the experiment reported in. These blockages were installed
perpendicular to the direction of the main coolant flow in a wide rectangular channel.
Thus blockages had the same cross section as the rectangular channel
Figure 3 Geometry of channel with blockage of six elongated hole
1.3. Heat Transfer characteristics
Heat transfer enhancement effect occurs when the secondary flows mix with the main
flows. There are two different kinds of secondary flows between two staggeringly
arrayed consecutive blockages. One is formed when the coolant passes through the
elongated holes as in rib turbulators for instance [1]. The other is built when the
coolant impinges onto the solid part between the two consecutive holes of blockages
On the basis of flow structure between two consecutive blockages, five heat transfer
enhancement deciding factors can be speculated. The first factor is the number of
impingement region. For instance, the blockages with four holes have four
impingement regions and that with six holes have six impingement regions. The
second and the third are the total and partial areas of impingement region,
respectively. Here, “partial area” means the region between two consecutive holes.
The fourth and the fifth are the total and partial widths of reattachment region,
respectively. Here, “partial width” means the width of one hole. These five heat
transfer enhancement deciding factors are controlled by the relative pitch ratio (P/e)
between the blockages and hence decides the final heat transfer enhancement effect.
The impingement and reattachment regions are not formed when blockages come
5. http://www.iaeme.com/IJMET/index.asp
closer to each other i.e., at lesser value of
passes to next blockage’s opening without giving any secondary flow
1.4. Results and Discussion
Table 1 Heat transfer co-efficient
Results
P/e Re
4.5
4347
6148
8133
10195
6.0
4347
6148
8133
10195
9.0
4347
6148
8133
10195
12.1
4347
6148
8133
10195
The plot shows Convective heat transfer coefficient variation with pitch distance.
Figure 4
S. Kandwal and Dr. S. Singh
ET/index.asp 44 editor@iaeme.com
closer to each other i.e., at lesser value of P/e, because the coolant in this case d
passes to next blockage’s opening without giving any secondary flow [3
Results and Discussion
efficient Comparison between CFD Analysis and Experimental
Results for Blockages with four elongated holes:
CFD
Value (h)
Exp.
Value (h)
% error
36.96 35.62 3.761931
42.08 40.73 3.31451
55.61 55.86 3.132832
69.23 67.01 3.312938
41.02 39.31 4.350038
45.96 44.61 3.026227
62.14 59.92 3.70494
76.39 74.05 3.160027
36.72 35.12 4.555809
41.33 40.04 3.221778
55.32 53.58 3.24748
69.10 66.46 3.972314
31.18 30.00 3.933333
35.601 34.31 3.762751
46.76 45.20 3.451327
58.66 56.68 3.493296
The plot shows Convective heat transfer coefficient variation with pitch distance.
Figure 4 Heat transfer coefficients − CFD Analysis
editor@iaeme.com
, because the coolant in this case directly
[3, 6, 7].
between CFD Analysis and Experimental
Avg. %
Error
3.38
3.56
3.749
3.66
The plot shows Convective heat transfer coefficient variation with pitch distance.
6. CFD Analysis and Enhancement of Heat Transfer
http://www.iaeme.com/IJMET/index.asp
Figure 5 Heat transfer coefficients
As compared to blockages with four holes, the convective heat transfer
coefficients have superior values for blockages with six holes at all Reynolds number.
Table 2 Heat transfer co-efficient
Results
P/e Re
4.5
4347
6148
8133
10195
6.0
4347
6148
8133
10195
9.0
4347
6148
8133
10195
12.1
4347
6148
8133
10195
The reason is that increased total widths of reattachment region (23.1 cm for six
holes as compared to 22.4 cm in four holes) and
dominant factors of heat transfer enhancement in case low hole aspect ratio (
for blockage with six holes.
CFD Analysis and Enhancement of Heat Transfer in Rectangular Channel Using Blockage
with Elongated Hole
ET/index.asp 45 editor@iaeme.com
Heat transfer coefficients − Experimental Analysis
blockages with four holes, the convective heat transfer
coefficients have superior values for blockages with six holes at all Reynolds number.
efficient Comparison between CFD Analysis and Experimental
Results for Blockages with six elongated holes:
CFD
Value (h)
Exp.
Value (h)
% error
42.1 38.7 8.78553
50.71 47.03 8.074685
64.8 60.35 5.3644
78.96 76.24 6.622951
44.57 41.24 7.824793
53.44 49.65 7.633434
68.79 63.53 6.292184
83.52 78.45 5.13321
38.89 36.92 7.373654
48.82 45.93 8.279553
61.67 58.53 5.36477
76.43 71.43 6.795741
32.52 30.5 7.502623
43.08 40.91 6.462715
55.16 51.61 6.99986
65.82 62.85 4.725537
The reason is that increased total widths of reattachment region (23.1 cm for six
holes as compared to 22.4 cm in four holes) and impingement region area are
dominant factors of heat transfer enhancement in case low hole aspect ratio (
for blockage with six holes.
n Rectangular Channel Using Blockage
editor@iaeme.com
blockages with four holes, the convective heat transfer
coefficients have superior values for blockages with six holes at all Reynolds number.
between CFD Analysis and Experimental
Avg. %
Error
7.871
7.612
5.998
5.882
The reason is that increased total widths of reattachment region (23.1 cm for six
region area are
dominant factors of heat transfer enhancement in case low hole aspect ratio (a/h) or
7. http://www.iaeme.com/IJMET/index.asp
Figure 6
As the hole aspect ratio increases both partial width of reattachment
partial impingement area between two consecutive blockages also increases but the
number of impingement regions decreases from six to four. This causes low value of
h at all Reynolds number for large aspect ratio hole i.e., blockage with four ho
Figure 7 Heat transfer coefficients by Experimental Analysis
As the air flow rate increases, the heat transfer enhancement first decreases and
then again increases. Though average Nusselt number was found to increase with
increase in Reynolds number, the Nusselt number ratio decreases. The increase in
average Nusselt number is due to better turbulence mixing at increased flow rate. The
increase in Reynolds number also causes an increase in the unsteady reverse flow just
behind the downstream blockages which helps in better mixing. However, the Nusselt
number ratio decreases due to the fact that as Reynolds number increases the flow
tends to reattach quickly and so the effect on the heat transfer coefficient is
considerably reduced. These results also confirm that even though average convective
S. Kandwal and Dr. S. Singh
ET/index.asp 46 editor@iaeme.com
Figure 6 Heat transfer coefficients by CFD Analysis
As the hole aspect ratio increases both partial width of reattachment
partial impingement area between two consecutive blockages also increases but the
number of impingement regions decreases from six to four. This causes low value of
at all Reynolds number for large aspect ratio hole i.e., blockage with four ho
Heat transfer coefficients by Experimental Analysis
As the air flow rate increases, the heat transfer enhancement first decreases and
then again increases. Though average Nusselt number was found to increase with
increase in Reynolds number, the Nusselt number ratio decreases. The increase in
t number is due to better turbulence mixing at increased flow rate. The
increase in Reynolds number also causes an increase in the unsteady reverse flow just
behind the downstream blockages which helps in better mixing. However, the Nusselt
creases due to the fact that as Reynolds number increases the flow
tends to reattach quickly and so the effect on the heat transfer coefficient is
considerably reduced. These results also confirm that even though average convective
editor@iaeme.com
As the hole aspect ratio increases both partial width of reattachment region and
partial impingement area between two consecutive blockages also increases but the
number of impingement regions decreases from six to four. This causes low value of
at all Reynolds number for large aspect ratio hole i.e., blockage with four holes.
Heat transfer coefficients by Experimental Analysis
As the air flow rate increases, the heat transfer enhancement first decreases and
then again increases. Though average Nusselt number was found to increase with
increase in Reynolds number, the Nusselt number ratio decreases. The increase in
t number is due to better turbulence mixing at increased flow rate. The
increase in Reynolds number also causes an increase in the unsteady reverse flow just
behind the downstream blockages which helps in better mixing. However, the Nusselt
creases due to the fact that as Reynolds number increases the flow
tends to reattach quickly and so the effect on the heat transfer coefficient is
considerably reduced. These results also confirm that even though average convective
8. CFD Analysis and Enhancement of Heat Transfer
http://www.iaeme.com/IJMET/index.asp
heat transfer coefficient for forced convection increases proportionally to Reynolds
number, the increase rate of reference Nusselt number (
that the average Nusselt number ratios of the blockages with four or six holes in case
of Reynolds number of 8133 and 10195 are superior to the Reynolds number value of
6148.
Figure 8 Nusselt number variation with pitch dista
Figure 9 Nusselt number variation with pitch distance for six
This is due to the enhancement in impinging and turbulent mixing effect of
blockages with increase in flow rate and therefore for higher value of
number, the average Nusselt number ratios again
aspect ratio of the holes, the corresponding Nusselt number
increases.
CFD Analysis and Enhancement of Heat Transfer in Rectangular Channel Using Blockage
with Elongated Hole
ET/index.asp 47 editor@iaeme.com
t for forced convection increases proportionally to Reynolds
number, the increase rate of reference Nusselt number (Nu0) surpasses it. Table shows
that the average Nusselt number ratios of the blockages with four or six holes in case
of Reynolds number of 8133 and 10195 are superior to the Reynolds number value of
Nusselt number variation with pitch distance for four elongated holes
Nusselt number variation with pitch distance for six elongated holes
This is due to the enhancement in impinging and turbulent mixing effect of
blockages with increase in flow rate and therefore for higher value of
number, the average Nusselt number ratios again increases. With an increase of the
aspect ratio of the holes, the corresponding Nusselt number ratios also significantly
n Rectangular Channel Using Blockage
editor@iaeme.com
t for forced convection increases proportionally to Reynolds
) surpasses it. Table shows
that the average Nusselt number ratios of the blockages with four or six holes in case
of Reynolds number of 8133 and 10195 are superior to the Reynolds number value of
elongated holes
elongated holes
This is due to the enhancement in impinging and turbulent mixing effect of
blockages with increase in flow rate and therefore for higher value of Reynolds
. With an increase of the
also significantly
9. http://www.iaeme.com/IJMET/index.asp
Table 3 Pressure Drop Comparison between CFD Analysis and Experime
Figure 10
Re
4.5
4347
6148
8133
10195
6.0
4347
6148
8133
10195
9.0
4347
6148
8133
10195
12.1
.
4347
6148
8133
10195
S. Kandwal and Dr. S. Singh
ET/index.asp 48 editor@iaeme.com
Pressure Drop Comparison between CFD Analysis and Experimental Results
Blockages with four elongated holes
Figure 10 Pressure drop plot − CFD analysis
CFD
ΔP
(N/m2)
Experimental ΔP
(N/m2)
57.87 50.62
109.36 97.98
224.12 199.24
346.53 303.76
54.82 47.68
106.76 94.72
181.81 160.04
269.88 238.43
53.66 46.38
95.17 83.28
161 140.44
219.15 195.97
50.96 44.10
81.17 73.49
133.33 117.58
185.24 163.31
editor@iaeme.com
ntal Results for
% Error
14.32
11.61
12.48
14.08
14.97
12.71
13.60
13.19
15.69
14.27
14.63
11.82
15.55
10.45
13.39
13.42
10. CFD Analysis and Enhancement of Heat Transfer
http://www.iaeme.com/IJMET/index.asp
Figure 11
1.5. Pressure contour plot for four elongated holes
Table 4 Pressure Drop Comparison
P/e Re
4.5
4347
6148
8133
10195
6.0
4347
6148
8133
10195
9.0
4347
6148
8133
10195
12.1.
4347
6148
8133
10195
CFD Analysis and Enhancement of Heat Transfer in Rectangular Channel Using Blockage
with Elongated Hole
ET/index.asp 49 editor@iaeme.com
Figure 11 Pressure drop plot − Experimental analysis
Pressure contour plot for four elongated holes
Figure 12 50.8 mm pitch at Re 4347
Pressure Drop Comparison between CFD Analysis and Experimental Results
Blockages with six elongated holes
CFD ΔP
(N/m2)
Experimental ΔP
(N/m2)
% Error
68.262 58.795 16.10171
135.847 117.581 15.53482
239.889 210.673 13.86794
351.133 310.291 13.16248
61.924 55.526 11.52253
124.781 113.665 9.779616
210.237 202.507 3.817152
270.642 244.96 10.48416
53.984 48.994 10.18492
119.714 109.419 9.408786
198.738 192.708 10.39396
252.985 232.14 8.979495
45.605 42.461 7.404442
108.46 99.62 8.87372
152.691 127.91 19.37378
240.177 214.21 12.12222
n Rectangular Channel Using Blockage
editor@iaeme.com
between CFD Analysis and Experimental Results for
% Error
16.10171
15.53482
13.86794
13.16248
11.52253
9.779616
3.817152
10.48416
10.18492
9.408786
10.39396
8.979495
7.404442
8.87372
19.37378
12.12222
11. http://www.iaeme.com/IJMET/index.asp
Figure 13
Figure 14
Pressure contour plot for six elongated holes
2. CONCLUSIONS
For the CFD analysis performed to study average heat transfer enhancement on
rectangular channel wall, the conclusions are as follows:
S. Kandwal and Dr. S. Singh
ET/index.asp 50 editor@iaeme.com
Figure 13 Pressure drop plot by CFD analysis
Figure 14 Pressure drop plot by Experimental analysis
Pressure contour plot for six elongated holes
Figure 15 50.8 mm pitch at Re 4347
For the CFD analysis performed to study average heat transfer enhancement on
rectangular channel wall, the conclusions are as follows:
editor@iaeme.com
For the CFD analysis performed to study average heat transfer enhancement on
12. CFD Analysis and Enhancement of Heat Transfer in Rectangular Channel Using Blockage
with Elongated Hole
http://www.iaeme.com/IJMET/index.asp 51 editor@iaeme.com
1. The low aspect ratio case (Blockages having six holes) showed the effective heat
transfer enhancement as compared to high aspect ratio case (Blockages having four
holes).
2. For the blockages, as hole aspect ratio increases, the heat transfer on the wall
segments is enhanced due to the increase in the reattached region and partial
impingement area. In this case, the flow reattachment and partial impingement area
are the dominant factors determining the heat transfer on the channel wall.
3. In the CFD analysis we have assumed that the duct wall as prefect adiabatic wall as
wall boundary condition which is not true in real case, no object is perfectly acts as
insulator or adiabatic i.e. heat transfer through the duct walls are assumed to be zero
hence heat transfer to atmosphere is zero. There is complete heat transfer from heated
wall to air is happening hence total heat transfer to air more than that of experimental
results.
4. Comparison of the computationally derived results with that of the experiments
shows a good correlation. Hence the proposed analysis method has demonstrated a
workable alternative to obtain heat transfer enhancement and thermal performance by
manipulating the results from ACU Solve simulation.
In this study, analysis to measure average heat transfer enhancement were
performed for blockages with only elongated holes for the purpose of application of
gas turbine trailing edge or middle portion internal cooling. Based on the results, P/e
ratio of 6.0 and Case 1(blockages with four holes) is the best option based on TP
value, but Case 2 (blockages with four holes) can be accepted if the level of penalty
from pressure drop is acceptable.
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to Dr. Satyendra Singh for his guidance
and assistance in this study work. The reality is that Dr. Satyendra Singh was much
more than an advisor for me. He always helped me in all the technical and non-
technical issues during the production of this work. His encouragement and efforts led
this report to successful completion in a timely fashion.
REFERENCES
[1] Pandey, R. Heat transfer enhancement through blockages with elongated holes in
a rectangular channel. Ph. D thesis, Mechanical Engineering Department, India,
DIT University, 2010.
[2] Moon, S. W. and Lau, S. C. Heat transfer between blockages with holes in a
rectangular channel. ASME J. Heat transfer, 125, 2003, pp. 587–594.
[3] Metzger, D. E., Fan, Z. X. and Shepard, W. B. Pressure loss and heat transfer
through multiple rows of short pin fins, in Heat Transfer, 3, Grigull U. et al., eds.,
Washington, DC: Hemisphere, 1982, pp. 137–142.
[4] Takeishi, K. Heat transfer research in high temperature industrial gas turbines, in
Proc. International Symposium on Heat Transfer in Turbomachinery, Marathon,
Greece, 1992.
[5] Lau, S. C., Cervantes, J., Han, J. C., Rudolph, R. J. and Flannery, K.
Measurements of wall heat(mass) transfer for flow through blockages with round
and square holes in a wide rectangular channel. Int J. Heat mass Transfer, 46,
2003, pp. 3991–4001.
[6] Lee, Y. Heat transfer enhancement for turbulent flow through blockages with
elongated holes in a rectangular channel, Ph. D thesis, Mechanical Engineering
Department, Texas: A & M University.
13. S. Kandwal and Dr. S. Singh
http://www.iaeme.com/IJMET/index.asp 52 editor@iaeme.com
[7] Shin, S. and Kwak, J. S. Effect of hole shape on the heat transfer in a rectangular
duct with perforated blockage walls, 2008.
[8] Taslim, M. E. and Spring, S. D. Effects of turbulator profile and spacing on heat
transfer and friction in a channel. AIAA Journal of Thermodynamics and Heat
Transfer, 8, 1994, pp. 555–562.