Journal of Kirkuk University –Scientific Studies , vol.4, No.

2,2009

Convection Concentric Annulus Vertical Cylinders Filling

Porous Media

Adnan M.Hussein* Tahseen A.Tahseen ** Attala H.Jasim**

* Technical Institute- Al-Hawija
**Collage of Engineering - University of Tikrit

Abstract

An experimental and theoretical study of convection vertical concentric annulus two

cylinders filling with porous medium was carried out . The experimental part was
contained manufactured test rig and connected thermocouples then power on and
reading temperatures until steady state . The theoretical part was contained solved
governing equations by using Fluent software which drawing temperature profiles and
stream functions for different ratio (r/R) and Ra . The results contained the relations
between Nu with Ra and Nu with ratio (r/R) then experimental Nu with theoretical Nu
so there was a good agreement between them.

Introduction
Jack(2006) was presented a results of heat transfer measurements in a
differentially heated annulus of fluid for both the non-rotating .The
resulting main effects appear to be exerted through a decrease in the
thickness.The steady conjugate heat transfer problem in vertical open
ended concentric annuli under the laminar natural convection flow regime
was presented by El-Shaarawi(1995). A finite difference technique was
used to solve the governing equations, Numerical results are presented for a
Newtonian fluid of Pr = 0.7 in a fluid annulus of radius ratio 0.5.the effect
of solid fluid conductivity ratio on the induced flow behavior and some
engineering parameters, such as the heat absorbed and the miring-cup
temperature. A finite-difference scheme to solve the transient conjugated
heat transfer problem in a concentric annulus with simultaneously
developing hydrodynamic and thermal boundary layers presented by
El-Shaarawi (1999). The effects of solid-fluid conductivity ratio and
diffusivity ratio on the thermal behavior of the flow have been investigated.
Numerical results were presented for a fluid flowing in an annulus of radius
ratio 0.5 with various values of inner and outer solid wall thicknesses.
Dirker(2000) were presented a comparative study of literature involving

55
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

convective heat transfer in annuli .Which shown that more research is

needed in the area of convective heat transfer correlations in concentric
annuli,as little agreement are found among existing correlation. A
correlation predicting Nu in annuli with ratios ranging from 1.7 to 3.2 was
developed experimentally for water as fluid .The correlation has an
accuracy of 3% in terms of experimental values for Re range. Projahn &
Beer(1985) was studied numerically the various Pr on the laminar
convection flow between concentric and vertically eccentric cylinders. Two
independent computer programs of the governing equations were used local
heat transfer rates are found to depend on the Pr in addition to the Ra
dependence. A numerical study of the natural convection heat transfer of
cold water ,having the density inversion between two isothermal eccentric
horizontal cylinder is studied by Raghavarao & Sanyasiraju (1996).The
numerical solution are obtained for Ra ranging between 1000-100000 a Pr
12 and inversion parameter 0,-1,and -2 .The affect of the radius ratio on the
flow patterns and heat transfer coefficient was studied by taking the radius
ratio as 1.5 and 2 .It is again for Y=-1 ,the minimum heat transfer is
observed like in the case of concentric annulus. Numerical solutions of
transient natural convection heat transfer problem in horizontal isothermal
cylindrical annuli were presented by Hassan & Al-lateef (2007).The results
showed that Grashof number was changed with influence of variation in
Prandtle number and diameter ratio .Good agreement with previous data
was obtained. An experimental and analytical model for natural convection
in a horizontal circular annulus with isothermal boundary condition was
presented by Teertstra & Yovanovich (2000).Comparisons between the
available data and models were carried with specific recommendations
concerning. In this work an experimental and theoretical study of
convection concentric annulus cylinders filling with porous medium was
carried out.The aim of this study to present the effect of ratio (r/R) and heat
flux or Ra on heat transfer coefficient with porous medium, then finding a
relation between Nu with (r/R) ratio for different Ra and Nu with Ra for
different ratio (r/R).

Experimental Part
The experimental investigation was carried out on annulus two
cylindrical concentrically rig, especially designed for the present study
covering all tests. The test rig was designed and manufactured to fulfill the
requirements of the test system for annulus concentric cylinders .The main

56
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

rig and the apparatus of the system illustrated in fig (1) , which contained
of inner steel cylinder (r=0.012,0.02,0.026m) thickness wall (0.002m) and
length (0.26m) put on outer C.I cylinder (R=0.12m), thickness wall
(0.008m) and length (0.26m) with heater inside the inner cylinder , closed
by Gipson and the annulus between heater and inner cylinder filled with
sand to ignore the natural convection in heater and getting constant
temperature around cylinder. The diameter ratio was used
(r/R=0.10,0.16,0.22) but the length was constant . The annulus between
two cylinders filling with porous media (Gravel) the thermal properties was
determined by Al-Kamil(1989).For measuring temperature12 thermocouples
type T were placed longitudinally along the section at points located on the
outer surface of inner cylinder for three levels as shown in fig(2).

Heater
Inner cylinder
Gypsum
Outer cylinder

Fig .1: schematic diagram of the test rig

Ta Tb
T1 T2

Tc T3 T4 Td

Te T5 T6 Tf

Fig.2:Thermocouples location

57
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

1. Measurements system:
The voltage regulator type (HSN 1.25KVA (0-250V)) was used to
control the supply voltage to heating elements A voltmeter was used to
measure voltage provided to heater(1500w) with accuracy (0.02v) , clamp
meter (mode no.266 (100mA+2%)) to measure alternate current passing the
circuit . A thermometer (digital multimeter MY-69 accuracy of 0.01oC )
was used to measure temperatures by thermocouples .
2.Experimental work:
Assembly two cylinders with heater vertically by steel table and
providing with electrical current and reading temperature until steady state .
3.Experimental Calculations:
The calculation of an experimental part was making as following:
Q=I*V …(1)
Q(input power) , I=current , V=voltage
Ai  (Ti 4  To4 ) …(2)
QR 
1 Ai 1
[  (  1)
ei Ao eo

Boltzmann constant =5.67x10-8), e (emissivity), A (area), T (temperature)

Qc=Q-QR …(3)
h
Qc …(4)
DL(Ti  To)

h(convection coefficient) , D(diameter) ,L(length)

Nu 
h*D …(5)
k

Ti= (T1+T2+T3+T4+T5+T6)/6 …(6)

To= (Ta+Tb+Tc+Td+Te+Tf)/6 …(7)
Tm= (Ti+To)/2 …(8)
1
 …(9)
Tm

Pr  …(10)

(kinematic viscosity) (thermal diffusivity), Pr (prandtl number)
Kg (Ti  To) H
Ra   …(11)

K (permeability), g(acceleration) ,  ( thermal expanded ) , H(height)

58
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Theoretical Part
1. Mathematical model:
The governing equations for natural convection heat transfer porous
medium problems are the continuity, momentum and energy equations.
Consider steady, Newtonian, and incompressible. Heat transfer phenomena
are described as follows:
i-Continuity Equation:(Bejan,1984)
u v
 0 ...(12)
x y
ii-Momentum equations:[ A.Bejan(1984)]
u u u  2u  2u 1 p
u v  ( 2  2 )  …(13)
t x y x y  x
density) , u(velocity with x-direction) , v(velocity with y-direction)
v v v  2v  2v 1 p
u v  ( 2  2 )   g[1   (T  TO ) …(14)
t x y x y  y
iii-Energy equation: (Bejan, 1984)
T T T   2T  2T 
u v  2  2  …(15)
t x y  x y 

2.Solution Using a CFD Package (FLUENT)

FLUENT is a state-of-the-art computer program for modeling fluid flow
and heat transfer in complex geometries. In FLUENT, the full set of mass,
momentum and energy equations are solved for two- or three-dimensions.
FLUENT allow us to include heat transfer within the fluid or solid in our
model. When heat is added to a fluid and the fluid density varies with
temperature, flow can be induced due to the force of gravity acting on the
density variations. Such flows are termed natural-convection flows and can
be modeled by FLUENT. The cylinder which equipped by porous media in
this paper had been modeled using a commercial computational fluid
dynamics package, Fluent, and its accompanying mesh generation
software, Gambit. Gambit was used for building the grid as shown in fig.
(3), the type of element used was Quadrilateral element. The model uses
steady solver formulation. The momentum and energy equations are solved
using the first order scheme. The standard discretization is recommended
for solving the natural convection . Under relaxation factors are applied in
order to control the changes of variable values between successive iterations
and avoiding divergence of the solutions.

59
Journal of Kirkuk University – Scientific Studies , vol.4, No.2,2009

Fig.3:Grid of mesh in gambit

Results and Discussion

The temperature profiles and stream functions were plotted for two heat
fluxes(250,400 w/m2) and 3 different ratio of diameter(r/R=0.10,0.16,0.22).
Fig(4-a) showed the temperature profiles for ratio r/R=0.1 , each closed
curve in the plots represented a contour of temperature profiles which
showed the red lines closed hot side and the blue lines closed cold side and
the max. value of temperature was 165oC for heat fluxes 250w/m2 but the
fig.4-b showed the eddy represented the stream function which showed the
red lines around the center of gap and blue lines closed walls , the max.
value was 0.84 for heat fluxes 250w/m2. Fig.5-a showed the temperature
profiles for ratio r/R=0.10 had the same behavior and the max. value of
temperature was 294oC and the fig.5-b showed the stream function which
the max. value was 0.044 for heat fluxes (400w/m2) .Fig(6-a) showed the
temperature profiles for ratio (r/R=0.16) had the same behavior and the
max. value of temperature was 105oC and the fig.6-b showed the stream
function which the max. value was 1.47 for heat fluxes (250w/m2) . Fig.7-a
showed the temperature profiles for ratio (r/R=0.16) had the same behavior
and the max. value of temperature was 178 oC and the fig(7-b) showed the
stream function which the max. value was (0.735) for heat fluxes
(400w/m2).Fig(8-a) showed the temperature profiles for ratio (r/R=0.22)
had the same behavior and the max. value of temperature was (91.9 oC) and
the fig(8-b) showed the stream function which the max. value was (2.0) for
heat fluxes (250w/m2) .Fig(9-a) showed the temperature profiles for ratio
r/R=0.22 had the same behavior and the max.

60
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

value of temperature was 232oC and the fig(9-b) showed the stream
function which the max. value was 4.33 for heat fluxes (400w/m2) . From
these figs.4-a,6-a,8-a and figs.5-a,7-a,9-a noted that the hot area was
increased vs. cooled area when increasing the heat flux and gap space , thus
figs.4-b,6-b,8-b and figs.5-b,7-b,9-b showed increasing the value of stream
function when increasing the heat flux and gap space .Fig (10) showed Nu
no. was changed with the ratio (r/R) for different heat fluxes so different Ra
no. and the equation between them:

Nu=1.7558e4.6635(r/R) …(16)

Fig (11) contained the relation between Nu and Ra for 3 diameter ratio
(r/R) showed the increasing of Nu with increasing of Ra and the relation
was found between them by the equation:
Nu = 1.6264e0.0027Ra …(17)
Note that the increasing of convection heat transfer came from the
increasing of heat flux which acting by Ra number.Fig (12) showed the
relation between experimental Nu and theoretical Nu which showed a good
agreement between them. Note that the behavior of heat transfer
experimentally was the same of the behavior of heat transfer numerically.

Fig.4-a:Contour of Temperature Profiles for (r/R=0.1) with heat flux

(250w/m2)

61
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Fig.4-b:Contour of stream function for (r/R=0.1) with heat flux (250w/m2)

Fig(5-a):Contour of Temperature Profiles for(r/R=0.1)with heat flux

(400w/m2)

62
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Fig(5-b)Contour of stream function for (r/R=0.1) with heat flux (400w/m2)

Fig(6-a):Contour of Temperature Profiles for (r/R=0.16) with heat flux

(250w/m2)

63
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Fig.6-b: Contour of stream function for (r/R=0.16) with heat flux

(250w/m2)

Fig.7-a: Contour of Temperature Profiles for (r/R=0.16) with heat flux

(400w/m2)

64
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Fig.7-b:Contour of stream function for(r/R=0.16)with heat flux(400w/m2)

Fig.8-a:Contour of Temperature Profiles for(r/R=0.22)with heat flux (250w/m2)

65
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Fig.8-b:Contour of stream function for (r/R=0.22) with heat flux (250w/m2)

Fig.9-a:Contour of Temperature Profiles for (r/R=0.22) with heat flux

(400w/m2)

66
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Fig.9-b:Contour of stream function for (r/R=0.22) with heat flux (400w/m2)

5.5

5 Nu = 1.7558e4.6635(r/R)

Ra=250
4.5

Ra=300
Nu
4
Ra=350

3.5

2.5

2
0 0.05 0.1 0.15 0.2 0.25

r/R

Fig.10: Nu with ratio (r/R) for different Ra

67
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

5.5

Nu = 1.6264e0.0027Ra
4.5
r/R=0.1

r/R=0.16
4
r/R=0.22

Nu
3.5

2.5

2
200 220 240 260 280 300 320 340 360

Ra

Fig.11: Nu with Ra for different ratio (r/R)

5.5 r/R=0.1
r/R=0.1
r/R=0.16
r/R=0.16
5
r/R=0.22
r/R=0.22

4.5
Nu th.

3.5

2.5

2
1 2 3

Nu exp.

Fig.12: The relation between theoretical Nu and experimental Nu

68
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

Conclusions
1.The behavior of temperature profiles were the same for any ratio of
diameter and different heat fluxes.
2.The behavior of stream functions were the same for any ratio of diameter
and different heat fluxes.
3.The relation between Nu and the ratio (r/R) was :
Nu = 1.7558e4.6635(r/R)
4.The relation between Nu and Ra was:
Nu = 1.6264e0.0027Ra
5.There was a good agreement between experimental and theoretical
studies, which showed the experimental part proved that the numerical part
was true with a few errors. And the experimental was correlated with
numerical studies.

References
 AL-Kamil.M.T.,(1989):A Packed Bed Solar Collector In Thermo-
siphon Water Heating System.Journal of Solar Energy Research
Vol.7, 17p.
 Bejan. A.,(1984): Convection Heat Transfer. New York , Ch.10 .
 Dirker.J. and Vyver.H.V.,(2000):Convection Heat Transfer in
Concentric Annuli.Department of Mechanical Engineering, South
Africa.
 El-Shaarawi.H., (1995):Transient Conjugated Heat-Transfer in
Concentric Annuli.I.J. of Numerical Methods for Heat Fluid Flow ,
Vol.5, pp: 459-473.
 El-Shaarawi .N.,(1999):Conjugate Natural Convection Heat Transfer
In An Open-Ended Vertical Concentric Annulus.I.J.N.M. Heat Fluid
Flow; Vol.36, pp: 639-655.
 Hassan.A.K. and Al-lateef .J.M. A.,(2007):Numerical Simulation of
Two Dimensional Transient Natural Convection Heat transfer From
Isothermal Horizontal Cylindrical Annuli.Eng.&Technology ,Vol.25.
 Jack.A.C.,(2006):Heat Transfer by Symmetrical Rotating Annulus
Convection.Journal of the Atmospheric Sciences, pp: 929-932.

69
Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009

 Projahn.U. and Beer.H.,(1985):Prandtl number effects on convection

heat transfer in concentric and eccentric horizontal cylindrical annuli.
J.of Heat and Mass Transfer Vol.19,pp.249-254 .
 Raghavarao .C.V. and Sanyasiraju.Y.V.S.S.,(1996):natural
convection heat transfer of cold water in an eccentric horizontal
cylindrical annulus – a numerical study.department of mathematics,
Indian institute of technology.
 Teertstra.P. and Yovanovich.M. M.,(2000):Comprehensive Review of
Natural Convection in Horizontal Circular Annuli" University of
Waterloo, Waterloo, Canada N2L 3G1 .

Table .1: lists of Symbols and units

Symbol meaning unit
A Area m2
E emesivity
H height m
I current A
K permeability m2
K Thermal conductivity W / m.oC
Nu Nusselt number -
Q Thermal Power W
R Outer cylinder radius m
R Inter cylinder radius m
Ra Raleigh number -
o
Ti Temperature of internal cylinder C
o
To Temperature of external cylinder C
U Velocity in x direction m/s
V voltage V
V Velocity in y direction m/s
 Thermal diffusivity m2/s
 Thermal expansion 1/oC
 Kinematics viscosity m2/s

70
‫‪Journal of Kirkuk University –Scientific Studies , vol.4, No.2,2009‬‬

‫الحمل الحر خالل فجوة مملوءة بوسط مسامي بين اسطوانتين عموديتين متحدتي‬
‫المركز‬

‫عطااهلل حسين جاسم**‬ ‫تحسين احمد تحسين**‬ ‫عدنان محمد حسين*‬

‫*المعهد التقني ـ الحويجة‬
‫** كلية الهندسة ـ جامعة تكريت‬

‫الخالصة‬

‫أجريت دراسة عملية ونظرية للحمل خالل فجوة مملوءة بوسط مسامي بين اسطوانتين عموديتين متحددتي‬
‫المركز‪ .‬تضمن الجزء العملي تصنيع مقطع االختبار وتثبيت المزدوجات الحرارية ثم تشغيل المنظومة وقدراءة‬
‫درجات الحرارة حتى الوصول إلى حالة االستقرار‪ .‬أما الجزء النظري فقد تضمن حدل المعدادالت الحاكمدة‬
‫باستخدام برنامج ‪ Fluent‬والذي تمكن من رسم التوزيع الحراري ودالة التدويم لنسب (‪ )r/R‬مختلفة وعدد ‪. Ra‬‬
‫النتائج تضمنت ربط عالقة بين عدد ‪ Nu‬وعدد‪ Ra‬وكذلك ‪ Nu‬و‪ r/R‬كما تضمنت العالقة بين عدد نسلت العملي‬
‫وعدد نسلت النظري وكان هناك توافق جيد بين الدراستين ‪.‬‬

‫‪71‬‬