Nu) Around Cylinder in Porous Medium Could Be

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

The Numerical Prediction of Air-side Heat Transfer Performance of Single Tube Cross-flow

Heat exchanger Embedded in Porous Medium


Prinya P.
Introduction
The present work numerically examines the cross-flow and air-side heat transfer performance of
a single cylindrical tube embedded directly in a porous medium under forced flow condition.
Several works have been devoted in the same topic [xx], and found that porous medium is the
notable technology for enhance the air-side heat transfer performance of both single tube [xx]
and tube bank [xx] heat exchanger. The heat transfer augmentation produced by the porous
medium is attributed to a combination effects, including thinning of both hydrodynamic and
thermal boundary layer around tube surface, increasing mixing (thermal dispersion),
modification of thermal conductivity (effective thermal conductivity). Moreover, the remarkable
advantage of porous medium is to enhance radiative heat transfer at high temperature or heat
flux. Therefore, some related works prefer to use porous medium to enhance combustion heat
transfer to the heat exchanger [xx] for industrial fluid heating system. However, this present
work will not focus on the radiation aspect.
According to the literatures, some works had been developed the useful correlations related to
forced convective air-side heat transfer performance of single tube embedded in porous medium.
Cheng [xx] concluded that Nusselt number (
D
Nu ) around cylinder in porous medium could be
correlated to a function of Peclet number (
D
Pe ) by using boundary layer approximation
technique: See Eq. (a) in Table A. Nasr et al. [xx] also purposed the related correlation from
their 167 experiment data: See Eq. (c) in Table A. Furthermore, the numerical investigation in
this topic had been carried by Nasr et al [xx], Younis [xx], and Mohammad [xx] with the purpose
of explore the thermal structure and local Nusselt number around tube surface. However, none of
the literature evaluates the influence of effective thermal conductivity (
eff
k ) model on the air-
side heat transfer performance. There are several models used for describe the thermal
conductivity of packed bed as shown in See Eq. (d-i) in Table A. Therefore, in this present work
those model will be used for calculate the (
eff
k ) of spherical alumina pellet bed with particle
diameter (
p
d ) of 10mm and average bed porosity ( ) of 0.6, and the effect of those model will
be evaluated compare with existing correlation of air-side heat transfer performance for
Reynolds number ( Re
D
) varied from 2 to 1200.
Governing Equation and method of solution
The schematic diagram of the solution domain (model) is illustrated in Fig. A. The domain is
assumed to be 5D high in order to neglect the wall effect. The uniform velocity ambient air with
U

and 303 T K

= flow past thought a non-slip isothermal cylindrical tube having 423


wall
T K =
embedded in packed bed of porous medium. The upper and lower domain assumed to be no-slip
wall with adiabatic condition. The exit domain is assumed to be a semi-infinite length with allow
heat transfer past thought only by convection (Convective flux). The fluid flow is assumed to be
two-dimensional steady flow. The Brinkman equation [xx] and the corresponding energy
equation will be used in the modeling.
Continuity: 0 u V = (1)
Momentum:
2
0 u p u
K

+V V = (2)
Energy:
2
,
0
f p f eff
C u T k T V V = (3)
where K is permeability obtained from Ergun [xx]:
| )
3 2
2
1
p
d
K
A

=

(4)
Nasr et al. [xx] suggest that the value employ for Ergun coefficient ( A) is 220.
A finite different method with non-uniform elements is used for the solution method. Elements
are clustered around cylinder and in-high gradient regions. Moreover, the fluid flow and heat
transfer model are tested with different element sizes, 7600 and 36400 elements. The results of
flow field and temperature prediction using both element sizes are not that significant with less
than 1% of discrepancy. Therefore, it is assumed that using 7600 elements is sufficient as far as
accuracy and computer resources are concerned. Moreover, the prediction of air-side heat
transfer coefficient is compared with published correlation in case of without porous medium
[xx] and found to be agreeing well with less than 2% of discrepancy. Therefore, it is ensured that
results are not element sizes dependent.
Data reduction
The temperature profile along bed axis at 0 y = will be illustrated in term of dimensionless
temperature ( O), which could be expressed as
wall
T T
T T

O =

(5)
Moreover, air-side heat transfer performance will be evaluated in term of air-side heat transfer
coefficient, which could be obtained from
''
eff
n
o
wall wall wall
k
q T
h
T T T T n

c
= =
c
(6)
where
''
n
q is normal total heat flux at tube surface, and the subscript n denotes outward normal
on the tube surface at a given angle from the front of the tube.
The air-side heat transfer coefficient based on the previous work [xx] could be obtained from
D eff
o
Nu k
h
D

= (7)
where
D
Nu is achieved from Eq. (a, c) in Table A. It must be noted that
o
h in Eq. (7) is based
on effective thermal conductivity of porous medium (
eff
k ), which could related on the chosen
model (See Table B). However, Nasr et al. [xx] recommends using ZBS Model [xx] for their
correlation (Eq. (c) in Table A). Moreover, in this present work, the same expression will be
used with Cheng [xx] correlation.
Results and Discussion
The results are compared with the numerical data for single tube in free stream (without porous
medium), and show that the presence of porous medium can stabilized the flow and eliminate
Von Karmen votices at downstream of the tube as illustrated in Fig. B, which tends to increase
the air-side heat transfer performance (
o
h ). Moreover, not only the stabilized fluid flow but the
modification of thermal conductivity from to is also enhanced the air-side heat transfer
coefficient up to about 4 times higher than that of the bare tube. According to Fig. B, in case of
porous medium with Re 100
D
= , the numerical simulation predicts 46.68
o
h W m K = using
ZBS model for effective thermal conductivity. However, in case of bare tube, 11.9
o
h W m K =
is obtained from both numerical prediction and correlation [xx].
The influence of Re
D
on dimensionless temperature profiles obtained from using ZBS model in
the numerical prediction is illustrated in Fig C. It is found that the thermal boundary layer in
front of the tube is thinned with the increasing Re
D
, which should result for the increasing in
o
h .
Moreover, the thermal boundary layer behind the tube is also thinned, which could be observed
from the decreasing of dimensionless temperature with increasing Re
D
.











Fig. A: The calculation domain.



Table A: The Nusselt number correlation of single tube embedded in porous medium.
No. References The correlations
a

Cheng [xx]

| )
1 2
1 0157
/
D D
Nu . Pe =
b Fand and Phan [xx]

| )
0 5 0 29 0 179
1 48 0 255 0 699
. . .
D D p
Nu . . . Re Pr Re = +

c

Nars et al. [43]

| )
0 0174
0 114
0 66
0 53
.
s f
.
. k k
D D
p
D
Nu . Pe
d

|
]
=
|
] |
]
' .








5D , U T


wall
T
x
y
D
Table B: The effective thermal conductivity model of porous medium.
No. References
Effective thermal conductivity (
eff
k )

d Deissier and Boegli [xx] | ) 1
eff g s
k k k = +
Parallel
Model
e Deissier and Boegli [xx]

| )
1
1
eff
g s
k
k k

=

+

Series Model
f

Lichteneker [xx]
| ) 1
eff g s
k k k

=
Geometric
Mean Model
g

Gelperin and Ainshtein [xx]
| )
0 18
0 63
1 1
1
0 28
.
g
s
g
s
eff g
k
.
k g
s
k
k
k k
k
.
k


|
|
|
' .
|
|
|

|
|
' .
= +
|
|
+ |
|
' .
-
h Kunni and Smith [xx]
| ) 1
eff f
g
s
k k
k
k

|
|

|
= +
|
|
| + +
|
|
' . ' .

-
i
Zehner and Schlunder [xx]


2 1
1 1
1
eff g
k k
B


= +


| )
| )
2
1
1 1 1
ln
2 1
1
B
B B
B B
B

+ |
]
`
|

' . ]

])

ZBS Model


Dimensionless bed length
0.0 .1 .2 .3 .4 .5 .6 .7 .8 .9 1.0
D
i
m
e
n
s
i
o
n
l
e
s
s

T
e
m
p
e
r
a
t
u
r
e
0.0
.1
.2
.3
.4
.5
.6
.7
.8
.9
1.0
Re
D
= 2
Re
D
= 50
Re
D
= 100
Re
D
= 200
Re
D
= 500
Re
D
= 1000
Increase Re
D
ZBS Model [xx]
d
p
/D
o
= 0.74, c = 0.6
T
wall
= 150
o
C, T
air,in
= 30
o
C

Angle, (deg)
0 45 90 135 180
H
e
a
t

T
r
a
n
s
f
e
r

C
o
e
f
f
i
c
i
e
n
t
,

h
o

(
W
/
m
2
-
K
)
0
10
20
30
40
50
60
70
80
without porous [xx]
Serie Model [xx]
Kunni and Smith [xx]
Gel'perin & Ainshtein [xx]
ZBS Model [xx]
d
p
/D
o
= 0.74, c = 0.6
T
wall
= 150
o
C, T
air,in
= 30
o
C

Reynolds number, Re
D
0 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300
H
e
a
t

T
r
a
n
s
f
e
r

C
o
e
f
f
i
c
i
e
n
t
,

h
o

(
W
/
m
2
-
K
)
1
10
100
1000
Region A Region B
without porous [xx]
Cheng [xx]
Nasr et al. [xx]
ZBS Model [xx]
Licheneker [xx]
Gel'perin & Ainshtein [xx]
Kunni and Smith [xx]
Serie Model [xx]
Parallel Model [xx]
d
p
/D
o
= 0.74, c = 0.6
T
wall
= 150
o
C, T
air,in
= 30
o
C

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy