Numerical simulation and experimental validation of a helical double-pipe

vertical condenser
D. Colorado
a
, J.A. Hernndez
a,
, O. Garca-Valladares
b
, A. Huicochea
a
, J. Siqueiros
a
a
Centro de Investigacin en Ingeniera y Ciencias Aplicadas (CIICAp), Universidad Autnoma del Estado de Morelos (UAEM), Av. Universidad 1001,
Col. Chamilpa, C.P. 62209 Cuernavaca, Morelos, Mexico
b
Centro de Investigacin en Energa (CIE), Universidad Nacional Autnoma de Mxico(UNAM), Privada Xochicalco S/N, Temixco, 62580 Morelos, Mexico
a r t i c l e i n f o
Article history:
Received 2 December 2009
Received in revised form 23 August 2010
Accepted 15 December 2010
Available online 7 January 2011
Keywords:
Absorption heat transformer
Heat transfer
Thermal performance
Water purication
Two-phase ow
a b s t r a c t
A predictive model is developed to describe heat transfer and uid dynamic behavior of a helical double-
pipe vertical condenser used in an absorption heat transformer integrated to a water purication process.
The condenser uses water as working uid connected in countercurrent. Heat transfer by conduction in
the internal tube wall is considered; in addition the change of phase is carried out into the internal tube.
The dynamic model considers equations of continuity, momentum and energy in each ow. The discret-
ized governing equations are coupled using an implicit step by step method. Comparison of the numerical
simulation over range of experimental data presented in the heat device is applied to validate the model
developed. The model is also evaluated of form dynamic to determine the principal operation variables
that affect the condenser with the main objective to optimize and control the system. A variation of mass
ow rate in the internal pipe induces important changes on the heat ux that the pressure and
temperature.
2010 Elsevier Ltd. All rights reserved.
1. Introduction
Condensation is a important phenomenon that occurs in many
applications [1]. Helical coiled tubes are extensively used in steam
generators, refrigerators, nuclear reactors and chemical plants, etc.,
due to their practical importance for high efciency heat transfer,
compactness in structure, ease of manufacture and arrangement
[2].
The condensers and evaporators devices model have been stud-
ies by different authors. Wang [3] presented a numerical method
to analyze the two-phase condensing ow in a double-pipe con-
denser using a semi-implicit nite-difference scheme used to solve
the governing equations. Morales et al. [4] presented a numerical
study of the thermal and uid-dynamic behavior of the two-phase
ow in ducts under condensation or evaporation phenomena. The
authors discretized governing equations and solved using the
semi-implicit method for pressure linked equations. Colorado
et al. [5] developed a predictive model to describe heat transfer
and uid-dynamic behavior of a helical double-pipe vertical evap-
orator for a heat transformer. Ali [6] performed analytically a de-
sign of a compact plates-and-frames absorber possessing a
hydrophobic microporous membrane contactor at the aqueous
solution water vapor interface. The absorber is a component of a
5 kW cooling capacity single-effect lithium bromide-water absorp-
tion chiller that incorporates a hot water thermally driven genera-
tor and a water-cooled absorber and condenser. Voyiatzis et al. [7]
developed a transient one-dimensional model, capable of describ-
ing the performance of a newly introduced adsorption chiller with
continuous operation. Gao et al. [8] carried out the numerical and
experimental assessment of thermal performance of vertical en-
ergy piles. Lee and Lu [9] evaluated empirically models for predict-
ing energy performance of vapor-compression water chillers. Li
et al. [10] developed and experimentally validated a special simu-
lation module for variable refrigerant ow system with a water-
cooled condenser (water-cooled VRF). Also, the ow patterns of
two-phase condensation ow in tubes have been studied experi-
mentally by Wang [11].
Helical double-pipe vertical condenser is also used in the
absorption heat transformer integrated to a water purication pro-
cess. The water purication process used in the absorption heat
transformer is a simple distillation process where impure water
is heated to obtain vapor which is immediately condensed. The
condenser releases heat and pure water. The absorption heat trans-
former is a system that consists of a thermodynamic device capa-
ble of producing useful heat at a thermal level superior to the one
at the source [12,13]. This integration of both systems enables to
increase the temperature of the impure water system, and thus ob-
tain pure water and useful heat. Thermodynamic models to
absorption system have been studied by different authors. Sencan
et al. [14] presented a thermodynamic and exergy analysis for sin-
gle-effect lithium bromide/water absorption system for cooling
0306-2619/$ - see front matter 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.apenergy.2010.12.026

Corresponding author. Tel.: +52 777 3 29 70 84; fax: +52 777 3 29 79 84.
E-mail address: alfredo@uaem.mx (J.A. Hernndez).
Applied Energy 88 (2011) 21362145
Contents lists available at ScienceDirect
Applied Energy
j our nal homepage: www. el sevi er. com/ l ocat e/ apenergy
and heating applications. Sozen and Serder Yucesu [15] showed a
mathematical model of absorption heat transformer operating
with the aqua/ammonia coupled to a solar pond and used a special
ejector. Therefore, these thermodynamic models are considered to
estimate the Coefcient of Performance (COP), which are based on
some assumptions that might be to difcult fulll in practice. The
heat ux to condenser is calculate under steady-state conditions.
Nevertheless, it is necessary to calculate heat ux in dynamic state
to control the on-line system.
According to Sozen and Arcaklioglu [16] for an ejector-absorp-
tion heat transformer, increasing condenser temperature increases
evaporator temperature for working heat transformer and de-
creased COP. The heat ux of the condenser is key to the perfor-
mance of the evaporator in a heat transformer. Consequently, the
objective of this paper is to develop a physical model to describe
heat transfer and uid-dynamic behavior inside a helical double-
pipe vertical condenser (CO). Therefore, study of helical condenser
is important because is a part of absorption system and inuence
COP.
2. Experimental data
Experimental database provided by Huicochea and Siqueiros
[17], consist of different COP values with energy recycling, ob-
tained from a portable water purication process coupled to an
absorption heat transformer. The experimental data set was the re-
sult of different initial concentrations of LiBr + H
2
O in AB and GE,
different temperatures in AB, GE, EV, CO and different pressures
in AB and GE. In addition to the experimental data of each compo-
nent, the steady-state is taken into account for each initial concen-
tration used in the process. From this database only the
experimental database of the condenser (CO) is considered in this
work.
The condenser has a design of helical double-pipe heat exchan-
ger. The whole system was built with stainless steel tubes and was
well isolated by foam insulation. Table 1 describes the dimensions
of the helical condenser. In the internal pipe, working uid ow
(water), which changes from vapor phase to liquid phase, reached
the heat to the cooling water (annulus).
Nomenclature
A cross area section [m
2
]
AB absorber
b coil pitch [m]
CO condenser
COP coefcient of performance [dimensionless]
Cp specic heat at constant pressure [J/kg C]
CV control volume
D helical diameter [m]
Dn Dean number Dn = Re
d
D
_ _1
2
_ _
d diameter [m]
EV evaporator
Fr concentration relation [dimensionless]
f friction factor [dimensionless]
g acceleration due to gravity [m/s
2
]
GE generator
G mass velocity [kg/m
2
s]
H height of helical coils[m]
h enthalphy [J/kg]
He Helical number He =
Dn
[1(b=2p(d=2))
2
[
1
2
_ _
_ m mass ow rate [kg/s]
L length of helical coil [m]
m mass [kg]
n
z
number of control volumes
Nu Nusselt number Nu =
ad
k
_
p pressure [bar]
P perimeter [m]
Pr Prandtl number Pr =
Cpl
k
_ _
_ q heat ux per unit of area [W/m
2
]
Q heat ux [W]
%RSD relative standard deviation
Re Reynolds number Re =
Gd
l
_ _
s deviation standard
t time [s]
T temperature [C]
v velocity [m/s]
X solution concentration in %wt
X
tt
Martinelli parameter X
tt
=
1xg
xg
_ _
0:9
q
g
q
l
_ _
0:5
l
g
l
l
_ _
0:1
_ _
[dimensionless]
x
g
vapor mass fraction [dimensionless]
z axial coordinate
Greek letters
h angle [rad]
q density [kg/m
3
]
d convergence criterion [dimensionless]
r supercial tension [N/m]
U two-phase frictional multiplier [dimensionless]
a heat transfer coefcient [W/m
2
C]
s shear stress [Pa]
k thermal conductivity [W/m C]
l dynamic viscosity [Pa s]

g
void fraction [dimensionless]
Dt temporal discretization step [s]
Dz axial discretization step [m]
Subscript
Anu annulus
c coiled
exp experimental
g vapor
Int internal pipe
l liquid
s straight
sim simulation
tp two-phase ow
Superscripts
previous instant
previous iteration
Table 1
Dimensions of the helical double-pipe condenser.
Internal pipe (mm) External pipe (mm)
External diameter (d) 17.1 26.67
Internal diameter (d) 13.8 22.45
Helical diameter (D) 194 194
Turns 5 5
Length (L) 3000 3000
Height (H) 330 330
Coil pitch (b) 60 60
D. Colorado et al. / Applied Energy 88 (2011) 21362145 2137
Fig. 1 shows inlet and outlet ow in the condenser. The vapor
working uid in the condenser inlet (E2), that comes from genera-
tor (GE) changes of phase, and it is transformed in outlet liquid
(S2). This outlet liquid in the condenser goes to the evaporator
(EV), which is now converted in inlet working uid in the evapora-
tor to continue the energy cycle. Fig. 2 shows a schematic diagram
of the absorption heat transformer integrated to water purication
process. The absorber gives a useful heat quantity Q
AB
, produced
by the heat transformer from the evaporator, condenser and
generator.
The experimental information is as follows: the system mass
ow rate was not measured for internal pipe, only mass ow rate
is registered for the annulus pipe. A transducer pressure was em-
ployed to measure the pressure in the outlet section the generator.
The general assumption is that the inlet pressure from the con-
denser is equal to the registered at the outlet section of the gener-
ator. In order to register the bulk temperature, four thermocouples
type J (iron-constantan) were installed at the inlet and outlet sec-
tions of ows (in positions E1, S1, E2 and S2) of the condenser (see
Fig. 1).
2.1. Experimental uncertainty analysis
In the present work, the quantities measured directly were:
mass ow rate, pressure and bulk temperature. The rotameter
had an accuracy of 3% of full scale, for external ow rate. The
transducer pressure uncertainty was estimated to be less than
0.25% on the measurement, the scale of transducer pressure is from
0.0541 bar to 0.8181 bar. Calibration of this system decreased the
error to 0.0169 bar for design condition according Corro et al.
[18]. The measurements of thermocouples had an accuracy of
0.2 C.
Based on the measurement of different variables, the uncer-
tainty of external heat transfer, in annular section, was estimated
in 11.56% according Bevington and Robinson [19].
The quantities measured indirectly was the mass ow rate in
the internal pipe given by:
_ m
CO
=
_ m
AB
Fr
(1)
where:
Fr =
X
GE
X
GE
X
AB
(2)
The LiBr concentration (X) is given by Huicochea et al. [20]:
X
LiBr
= 1144:335IR
2
3647:354IR 2820:451 (3)
The refraction index (IR) had uncertainty to 0.0002 on the
measurement. The rotameter had an accuracy of 2% of full scale
for _ m
AB
measurement. Based on the measurement and errors of dif-
ferent variables, Eqs. (1)(3) and the error propagation describe by
Bevington and Robinson [19], the uncertainty of internal mass ow
rate, was estimated in %RSD =
s
_ m
= 5:9049.
3. Mathematical formulation of two-phase ow
In this section, mathematical formulation of two-phase ow
over a control volume (CV) is described. A control volume is a nite
E2
S1
E1
S2
D
H
d4
d2
b
Fig. 1. Helical double-pipe condenser used in the heat transformer.
Generator Absorber
Condenser Evaporator
Impure water
Heat source
Auxiliary
condenser
Pure water
Phase
separator
System under
study
Fig. 2. Schematic diagram of absorption heat transformer integrated to water purication process.
2138 D. Colorado et al. / Applied Energy 88 (2011) 21362145
volume that delimits a physical space corresponding to a partial or
global zones of a thermal unit [21], in this case, the helical con-
denser. A control volume is shown in Fig. 3, where i and i + 1
represent the inlet and outlet sections, respectively.
Taking into account the characteristic helical coils, the govern-
ing equations have been integrated following these assumptions
[21]:
v Fluid: water.
v One-dimensional ow: p(z, t), h(z, t), T(z, t), etc.
v Non-participant radiations medium and negligible radiant heat
exchanger between surfaces.
v Axial heat conduction inside the uid is neglected.
v Internal and coiled diameter constant and uniform roughness
surface.
The semi-integrated governing equations (mass, momentum,
energy and entropy) over control volume were presented by Gar-
ca-Valladares et al. [22]. For each control volume, a set of algebraic
equations is obtained by a discretization of the governing equa-
tions. The transient terms of the governing equations are discret-
ized using the following approximation
d/
dt

(//
o
)
Dt
where /
represents a generic dependent variable (/ = h, p, T, etc.); super-
script o indicates the value of the previous instant. The average
of the different variables has been estimated by the arithmetic
mean between their values at the inlet and outlet sections, that
is:
~
/
i


/ =
(/
i
/
i1
)
2
.
Based on the numerical approaches indicated above, the gov-
erning equations can be discretized to obtain the value of the
dependent variables (mass ow rate, pressure and enthalpy) at
the outlet section of each control volume.
The mass ow rate is obtained from the discretized continuity
equation,
_ m
i1
= _ m
i

ADz
Dt
q
tp


q
o
tp
_ _
(4)
where the two-phase ow density is obtained from: q
tp
=
g
q
g
+
(1
g
)q
l
.
In terms of the mass ow rate, gas and liquid velocities are cal-
culated as,
v
g
=
_ mx
g
q
g

g
A
v
l
=
_ m(1 x
g
)
q
l
(1
g
)A
(5)
The discretized momentum equation is solved for the outlet
pressure:
p
i1
= p
i

Dz
A
U
2
lo

f
l
4

_ m
2
2 q
l
A
2
P q
tp
Ag sinh
_

_ m(x
g
v
g
(1 x
g
)v
l
)
_
i1
i
Dz

_ m

_ m
o
Dt
_
(6)
where the subscript and superscript in the brackets indicate
[x[
i1
i
= x
i1
x
i
, i.e., the difference between the quality x at the out-
let section and the inlet section. With the inlet and outlet pressure
obtained from the momentum Eq. (6), the pressure drop in each
control volume is also evaluated.
From the energy equation and the continuity equation, the fol-
lowing equation is obtained for the outlet enthalpy:
h
i1
=
2
_
qPDz _ m
i1
a _ m
i
b
ADz
Dt
c
_ m
i1
_ m
i

~
o
g
~ q
o
g
1~
o
g ( )
~ q
o
l
( )ADz
Dt
(7)
where:
a = (x
g
v
g
(1 x
g
)v
l
)
2
i1
g sinhDz h
i
b = (x
g
v
g
(1 x
g
)v
l
)
2
i
g sinhDz h
i
c = (p
i
p
i1
) p
o
i
p
o
i1
_ _

o
g
~ q
o
g
1
~

o
g
_ _
~ q
o
l
_ _
h
o
i
h
o
i1
_ _
h
i
_ _
~ q
v
i
v
i1
2
_ _2
~ q
o
v
o
i
v
o
i1
2
_ _
2
_ _
(8)
Temperature, mass fraction, density, and all thermophysical
properties are obtained from [23] which were based on the limits
of operation of the helical condenser.
The one-dimensional model requires the knowledge of the two-
phase ow structure, which is evaluated by means of void fraction
e
g
. The evaluation of the shear stress is performed by means of a
friction factor f. This factor is dened from the expression:
s
w
= U
2
lo

f
l
4
G
2
2 q
l
where U
2
lo
is the two-phase factor multipler [21].
The inertial force, centrifugal forces and gravity inuences are em-
broiled in the calculation of friction factor for helical coil. A heat
transfer through the helical tube wall and uid temperature are re-
lated to the convective heat transfer coefcient a, which is dened
as:
a =
_
q
wall
T
wall
T
fluid
(9)
In the same way, the calculation of convective heat transfer in-
volves previously mentioned helical effects.
The above mentioned conservation equations are applicable to
transient two-phase ow. Situations of steady ow and/or single-
phase ow (liquid or gas) are particular cases for this formulation.
In the internal pipe of the helical condenser, different regions
are present in condensation process. The differentiations between
the three principal regions are given by the enthalpy, pressure and
vapor quality, then following these criteria:
v Liquid region: when h(p) < h
l
(p), then x
g
= 0.
v Two-phase region: when h
l
(p) 6 h(p) 6 h
g
(p), then 0 < x
g
< 1.
v Vapor region: when h(p) > h
g
(p), then x
g
= 1.
where h
l
(p) and h
g
(p) represent the saturation liquid and gas
enthalpy for a given pressure p.
The vapor quality is determined by:
x
g
=
h h
l
(p)
h
lg
(p)
(10)
where the pressure and the enthalpy are determined in each control
volume.
z
z
i
i+1
g
m
.
Fig. 3. Flow inside a helical CV.
D. Colorado et al. / Applied Energy 88 (2011) 21362145 2139
3.1. Heat conduction in the internal and external tube wall
In the internal tube, the physical space that separates the ows,
tube wall, is treated assuming the following hypotheses: one-
dimensional transient temperature distribution and negligible heat
exchanger by radiation. The energy balance used to describe heat
transfer in the wall is obtained (see Fig. 4):
(
~
_
q
ns
P
ns

~
_
q
nn
P
nn
)Dz (
~
_
q
nw

~
_
q
ne
)A = m
@
~
h
@t
(11)
where (
~
_ q
ns
) and (
~
_ q
nn
) are evaluated using the respective convective
heat transfer coefcient in each zone, and the conductive heat
transfer uxes are evaluated from the Fourier law, which is:
~
_ q
ne
=
k
ne
(@T=@z)
ne
and
~
_ q
nw
= k
nw
(@T=@z)
nw
.
In the external tube, heat conduction in tube, insulation and
natural convection with the environmental are considered in the
model according to Garca-Valladares et al. [21]. It is necessary to
know the heat losses of each component (AB, EV, CO, GE) of the
absorption heat transformer. The conduction equation for the insu-
lation has been written assuming transient asymmetrically tem-
perature distribution and negligible heat exchanged by radiation
with the external ambient. The north and south interfaces are eval-
uated from the Fourier law, except in the tube-insulation interface
(where a harmonic mean thermal conductivity is used) and in the
insulation-ambient interface (where the heat transfer by natural
convection is introduced) [22].
4. Evaluation of empirical coefcients
A comparison of different empirical correlations presented in
the literature [1,2,2427] was realized in order to select the follow-
ing ones.
4.1. Single-phase region, liquid and vapor in internal pipe.
The friction factor is evaluated from the following expression
proposed by White cited in [26] in the laminar region for
Re 6 2300 in curved pipes:
f
c
f
s
=
0:25
1 1
11:6
Re
d
D
( )
0:5
_ _
0:45
_ _ 1
0:45
_
_
_
_
_
_
(12)
In the case of turbulent region the friction factor is evaluated
with the equation proposed by Ito for Re > 2300:
f
c
= 1:216Re
0:25
0:116
d
D
_ _
0:5
(13)
Guo et al. [26] recommend this equation as the standard for-
mula for the turbulent region. The convective heat transfer coef-
cient is calculated using the equation proposed by Churchill cited
in [28], for the laminar region in helical coiled tubes (Re 6 2300)
as follows:
Nu =
48
11

51
11
1
1342
PrHe
2
_ _
3
1:816
He
1
1:15
Pr
_ _3
2
_
_
_
_
1
3
(14)
For turbulent region, the equation proposed by Rogers and May-
hew [29] is used, for Re > 2300:
Nu = 0:023Re
0:8
Pr
0:4
Re
d
D
_ _
2
_ _
0:1
(15)
4.2. Two-phase ow region, in internal pipe
The void fraction is evaluated according to semi-empirical rela-
tions proposed by Premoli [30]:

g
=
1
1
1xg
xg
_ _
.
l
.
g
_ _
S
(16)
where:
S = 1 F1

y
1 yF2
yF2
_
F1 = 1:578Re
0:19
l
.
l
.
g
_ _
0:22
F2 = 0:0273We
l
Re
0:51
l
.
l
.
g
_ _
0:08
y =

ghomog
1
ghomog

ghomog
=
1
1
1xg
xg
_ _
.
g
.
l
_ _ (17)
Using correction factor (two-phase frictional multiplier) accord-
ing by Guo et al. [26].
/
2
lo
= 1 4:25 2:55x
1:5
g
_ _
G
0:34
(18)
The convective heat transfer coefcient is calculated by Tang
et al. [31] correlation. This correlation was modied involved heli-
cal parameter, it is used for the laminar and turbulent region.
Nu
tp
= 0:023Re
0:8
Pr
0:4
Re
d
D
_ _
2
_ _
0:1
1 4:863 ln p
red
( )
x
g
1 x
g
_ _
0:836
_ _
(19)
4.3. Annulus
Due to the fact that no information is available in technical lit-
erature about a specic correlation used to get the friction factor
and the heat transfer coefcient in the annulus of helical tubes,
they are calculated using the expressions corresponding to sin-
gle-phase ow evaluated with the hydraulic diameter.
5. Numerical solution
The main objective is the determination of the heat transfer and
uid ow behavior through the helical condenser. The discretized
. .
.
nW
z
nP
nE
nn
ns
nw
ne
i-1
i+1
Fig. 4. Heat uxes in a CV of a tube.
2140 D. Colorado et al. / Applied Energy 88 (2011) 21362145
equations have been coupled using a fully implicit step by step
method in the ow direction. From the known values at the inlet
section, the variables values at the outlet of each CV are iteratively
obtained. This solution (outlet values) is the inlet values for the
next control volume. The procedure is carried out until the end
of the helical condenser is reached.
With the conditions of differentiation between regions men-
tioned in mathematical formulation of two-phase ow CVs where
the transition occurs, are detected. The criterion of transition used
in internal pipe, is that when control volume is detected, the begin-
ning of this transition is associated to the outlet of the control vol-
ume. This criterion is important to calculate the appropriated
empirical coefcients in each zone.
Inside each CV was veried the convergence working with the
following criteria (Eq. 20):
/
+
i1
/
i
D/

< d (20)
Here, / is the mass ow rate, pressure and enthalpy. The super-
index + represents the values at the previous iteration. The referent
values D/ are evaluated in each control volume: /
i+1
/
i
. When
this value tends to be zero, it is substituted by /
i
.
Heat conduction in the tube wall and tube wall with insulation
were discretized in accordance by Garca-Valladares et al. [22]. The
set of heat conduction discretized equations is solved using the tri-
diagonal matrix algorithm.
The numerical model to describe heat transfer and uid-
dynamic ow behavior is as follows: the dominium of helical con-
denser is discretized in a xed number of CVs; one-dimensional
ow was adopted in internal and annulus ow; in the wall, solid
element to separate the ows, a displaced mesh was used. The
algorithm follows next sequence:
(1) Initial conditions of ows were specied, mass ow rate,
pressure and temperature. If ow quality is known in the
case of the inlet two-phase ow, this data is used.
(2) Suppose a temperatures distribution in the wall.
(3) The ow in internal pipe is solved. In this section two-phase
ow is presented.
(4) Flow in the annulus is solved.
(5) The external tube wall and insulation is calculated in order
to considered heat losses to the environment.
(6) Re-calculate the temperature distribution in the wall with
the temperature distribution in the ows and heat transfer
coefcients using a tri-diagonal matrix algorithm.
These steps are repeated themselves up to verify a strict conver-
gence criterion (Eq. 20). The steady-state is a particular form for
this model, where it is not considers the temporary derivative in
the discretized equations.
6. Numerical results and experimental validation
Experimental conditions (Test B, see Table 3) are used to deter-
mine the inuence of the number to CV over numerical results ob-
tained for the outlet temperature, pressure and enthalpy in
internal pipe and annulus. Table 2 shows to results independence
of spatial mesh obtained with n
z
P200 CVs with a convergence
criterion of d = 10
5
.
The model in steady-state, with 600 CVs, Dz = 0.0050 m and
convergence criterion d = 10
5
, is compared with experimental
data by Huicochea and Siqueiros [17,32]. This model used initial
experimental values in steady-state in the input ows. Table 3
shows the results compared to outlet values of ows.
Experimental operating conditions by Huicochea and Siqueiros
[17] are results of experimental modications over heat trans-
former with aim of incrementing COP of heat transformer and eval-
uate a compact device. Tests A-J are experimental operating
conditions helical condenser with recycling heat from auxiliary
condenser to a compact heat transformer. An important character-
istic of the model is that no parameter was adjusted to describe heat
transfer in condenser.
Fig. 5 compare temperature and pressure distribution in the
ows obtained with the model with experimental conditions in
Table 2
Outlet temperature, pressure and enthalpy obtained for different numbers of CV and convergence criterion in internal pipe and annulus.
Numerical parameters Internal pipe Annulus
n Dz =
L
n
d T
out
(C) p
out
(bar) hout (kJ/kg) T
out
(C) p
out
(bar) hout (kJ/kg)
200 0.0150 m 10
3
23.62 0.06 99.00 23.17 0.65 97.18
200 0.0150 m 10
5
23.81 0.06 99.78 23.18 0.65 97.21
500 0.0060 m 10
3
24.30 0.06 101.82 23.22 0.65 97.38
500 0.0060 m 10
5
24.49 0.06 102.62 23.18 0.65 97.22
1000 0.0030 m 10
3
24.37 0.06 102.11 23.22 0.65 97.38
1000 0.0030 m 10
5
24.77 0.06 103.80 23.18 0.65 97.21
Table 3
Experimental database of the helical condenser and simulated results with d = 10
5
.
Test Internal pipe Annulus
Exp Exp Exp Sim Sim Sim Exp Sim
_ m
inlet
10
4
(kg/s)
T
outlet
(C) p
inlet
(bar) _ m
inlet
10
4
(kg/s)
T
outlet
(C) p
outlet
(bar) T
outlet
(C) T
outlet
(C)
A 4.12 0.24 23.41 0.2 0.09 0.01 4.30 25.03 0.06 22.64 0.2 22.69
B 3.76 0.22 24.42 0.2 0.09 0.01 3.76 24.82 0.06 23.01 0.2 23.18
C 3.52 0.20 22.70 0.2 0.10 0.01 3.70 19.96 0.07 21.79 0.2 21.74
D 5.32 0.31 23.00 0.2 0.12 0.01 5.10 20.17 0.09 22.99 0.5 22.86
E 5.20 0.30 23.69 0.2 0.12 0.01 4.90 21.05 0.09 23.54 0.5 23.69
F 5.20 0.30 23.36 0.2 0.11 0.01 4.88 20.87 0.08 23.12 0.5 23.37
G 3.90 0.23 25.51 0.2 0.10 0.01 3.90 21.39 0.07 23.03 0.2 22.23
H 3.78 0.23 23.06 0.2 0.09 0.01 3.99 20.25 0.07 21.94 0.2 21.83
I 3.54 0.20 23.82 0.2 0.09 0.01 3.54 22.12 0.07 22.96 0.2 22.84
J 4.05 0.23 23.11 0.2 0.11 0.01 4.05 19.80 0.08 22.19 0.2 22.13
D. Colorado et al. / Applied Energy 88 (2011) 21362145 2141
steady-state for the test B, D and E. These Figures, denote that the
model proposed represented the temperature and pressure behav-
ior across the pipe in a good way.
Experimental data present variations in operations conditions
of helical double-pipe condenser from experimental data by Huico-
chea and Siqueiros [17]. The same working uid (water) and helical
geometry were applied. Variations in inlet pressure, inlet and out-
let temperature and mass ow rate are reported. The standard
deviation for external heat ux calculated for these tests, are esti-
mated in 0.1188 kW. The numerical results for the calculation of
the external heat ux, for these tests in steady-state, had a discrep-
ancy lower than 10.9% with regard to experimental results. Fig. 6
shows experimental and simulated heat ux for all tests. This
Fig. 6 exhibited a satisfactory ability to estimates the heat ux.
The experimental heat ux has a range from 0.87 to 1.21 kW.
Moreover the model is based on the applications of physical laws,
it is possible to extrapolate other operating conditions with good
condence. Furthermore, Table 4 presents the simulated, experi-
mental heat ux and heat losses of environment for these tests.
In this Table 4, the heat losses for the environment is small with
regard to external heat ux. This helical double-pipe vertical con-
denser model to will be incorporated to the general model which
predict the COP of the absorption heat transformer integrated to
water purication process [5,3335].
7. Dynamic analysis
As showed in the previous section, the model in steady-state
reproduced with good accuracy the experimental data. However,
two-phase flow
liquid
vapor
S2
E2
E1
S1
liquid annulus
o
T internal flow simulated
T annulus flow simulated
experimental
20
30
40
50
60
20
30
40
50
60
Distance (m)
T
e
m
p
e
r
a
t
u
r
e


(

C
)
vapor
two-phase flow
liquid
p internal flow simulated
experimental
0.07
0.08
0.09
0.1
0.11
0.07
0.08
0.09
0.1
0.11
Distance (m)
P
r
e
s
s
u
r
e

(
b
a
r
)
two-phase flow
liquid
vapor
S2
E2
E1
S1
liquid annulus
o
T internal flow simulated
T annulus flow simulated
experimental
20
30
40
50
60
20
30
40
50
60
Distance (m)
T
e
m
p
e
r
a
t
u
r
e


(

C
)
vapor
two-phase flow
liquid
p internal flow simulated
experimental
0.1
0.11
0.12
0.13
0.14
0.1
0.11
0.12
0.13
0.14
Distance (m)
P
r
e
s
s
u
r
e

(

b
a
r
)
two-phase flow
liquid
vapor
S2
E2
E1
S1
liquid annulus
o
T internal flow simulated
T annulus flow simulated
experimental
20
25
30
35
40
45
50
55
60
20
25
30
35
40
45
50
55
60
Distance (m)
T
e
m
p
e
r
a
t
u
r
e


(

C
)
vapor
two-phase flow
liquid
p internal flow simulated
experimental
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0.1
0.11
0.12
0.13
0.14
0.1
0.11
0.12
0.13
0.14
Distance (m)
P
r
e
s
s
u
r
e

(
b
a
r
)
Fig. 5. Experimental vs. simulated in steady-state data for tests B, D and E (see Table 3).
2142 D. Colorado et al. / Applied Energy 88 (2011) 21362145
this model is developed to heat pump control purpose, as a conse-
quence it is important to test the robustness of the model for small
perturbations of the process variables (T, p, etc.).
With the objective to determine the inuence over results of
temporal step in dynamic model, we idealize uid ow in helical
double-pipe condenser with temperature and pressure perturba-
tions. Variation from 61.96 to 66.96 C and variations from
0.1252 to 0.1500 bar was used in the internal inlet pipe (see test
D in Table 3) with different temporal steps (Dt). Table 5 shows
the liquid temperature in different times, the results are indepen-
dent of temporal step if Dt > 10s, it presented variations on the sec-
ond decimal.
Illustrative dynamic thermal results with Dt = 10 s, n
z
= 600 and
perturbation mentioned was shown in Fig 7.
A dynamic model was used to describe heat transfer and ow in
function of the time in the helical condenser. Simulated result in
steady-state used to experimental conditions in Test D was also ap-
plied for this analysis. Variations of operation variables in internal
pipe was carried out, as consequence of the desorber performance.
A variation of inlet internal pipe pressure, from 0.12 to 0.15 bar
induced important changes on the saturation temperature of work-
ing uid (see Fig. 8). External heat ux for 0.12 bar in steady-state
is determined to 1.2931 kW, with pressure perturbation to 0.15 bar
external heat ux is determined to 1.2924 kW. Increased of inlet
pressure working uid present decreased to heat ux in 0.05%, this
not is a signicative change in the heat ux.
A variation of inlet mass ow rate in internal pipe, from 0.00051
to 0.00056 kg/s presents important changes on the heat ux (see
Fig. 9). These variations represent increases of 10% in the internal
inlet mass ow rate, possibly produced by the desorber of heat
transformer. Increased to 10% in internal inlet mass ow rate in-
creased external heat ux from 1.29 to 1.42 kW, representing an
increment of 10.01%.
A variation of the inlet internal pipe temperature in 5 C, from
61.96 to 66.96 C did not show any effect on the simulated liquid
temperature in short time periods (t = 100 s) (see Fig. 10). These ef-
fect increased external heat ux in 0.40%, this not is a signicative
change in the heat ux.
Recording the concept incorporated in the literature [5] about
recycled heat, variation of inlet temperature had been realized,
maintained at original operating conditions of helical condenser.
This variation possibly, was obtained to heat recycled since auxil-
iary condenser of water purication process to desorber of heat
transformer.
Perturbations realized in this work to helical double-pipe con-
denser not modify outlet liquid temperature of condenser. There-
fore, not increased temperature of evaporator and not inuences
the COP. To recapitulate this sensitivity analysis of the physical
model, this work shows that the principal parameters that deter-
mine an increase in the heat ux is the mass ow rate in internal
pipe keeping constant annular ow conditions.
8. Conclusion
A numerical model of heat transfer and uid-dynamic behavior
of a helical double-pipe condenser has been developed by means of
a transient one-dimensional analysis of the ows together with a
detailed analysis of heat conduction in the internal wall and exter-
nal wall with insulation. The empirical coefcients used in the
model to evaluate the shear stress, heat ux and two-phase struc-
ture had been selected in technical literature for helical systems.
Finite volume formulations of the governing equations in two-
phase ow were used. The model implements an implicit step by
step numerical scheme for the working uid and uid in the annu-
lus and implicit central difference numerical scheme in walls.
0.8 1 1.2 1.4
0.8
1
1.2
1.4
simulated Q (kW)
e
x
p
e
r
i
m
e
n
t
a
l

Q

(
k
W
)
Fig. 6. Experimental and simulated values of heat ux (Q).
Table 4
Comparison of simulated, experimental and environment values of external heat ux.
Test Q
exp
(kW) Q
sim
(kW)
%Error =
[simexp[
exp
100
Q
environment
(W)
A 1.05 1.08 2.85 5.92
B 0.87 0.94 8.04 4.80
C 0.97 0.94 3.09 7.09
D 1.21 1.29 6.61 6.60
E 1.16 1.23 6.03 5.16
F 1.11 1.23 10.81 5.62
G 0.91 0.98 7.69 4.77
H 1.06 1.00 5.66 7.00
I 0.95 0.89 5.00 6.31
J 1.06 1.02 3.77 6.78
Average 5.95
Max 10.81
Table 5
Outlet temperature (C) obtained for different temporal steps and instants (Dt) with
n
z
= 600 and d = 10
5
for helical double-pipe condenser.
Dt t = 0 t = 50 t = 100 t = 150 t = 200
5 20.17 20.09 20.07 20.06 20.05
10 20.17 20.08 20.04 20.00 19.98
25 20.17 20.09 20.04 20.01 19.99
50 20.17 20.12 20.09 20.06 20.04
100 20.17 20.12 20.07
Annulus
working fluid
t = 0s
t = 200s
t = 300s
0 1 2 3
0 1 2 3
20
30
40
50
60
20
30
40
50
60
Distance (m)
T
e
m
p
e
r
a
t
u
r
e


(


C
)
o
Fig. 7. Illustrative dynamic thermal behavior of helical double-pipe condenser.
D. Colorado et al. / Applied Energy 88 (2011) 21362145 2143
Annulus
working fluid
t = 0s
t = 200s
t = 300s
20
30
40
50
60
20
30
40
50
60
Distance (m)
working fluid
t = 0s
t = 200s
t = 300s
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0.1
0.11
0.12
0.13
0.14
0.15
0.1
0.11
0.12
0.13
0.14
0.15
Distance (m)
p
r
e
s
s
u
r
e

(
b
a
r
)
T
e
m
p
e
r
a
t
u
r
e

(

C
)
Fig. 8. Variation in the inlet internal pressure.
Annulus
working fluid
t = 0s
t = 50s
t = 100s
20
30
40
50
60
20
30
40
50
60
Distance (m)
T
e
m
p
e
r
a
t
u
r
e

(


C
)
working fluid
t = 0s
t = 50s
t = 100s
0 1 2 3
0 1 2 3
0 1 2 3
0 1 2 3
0.09
0.1
0.11
0.12
0.09
0.1
0.11
0.12
Distance (m)
p
r
e
s
s
u
r
e

(
b
a
r
)
o
Fig. 9. Variation in the inlet mass ow rate (increment of 10%).
Annulus
working fluid
t = 0s
t = 50s
t = 100s
0 1 2 3
0 1 2 3
20
30
40
50
60
20
30
40
50
60
Distance (m)
T
e
m
p
e
r
a
t
u
r
e

(


C
)
working fluid
t = 0s
t = 50s
t = 100s
0 1 2 3
0 1 2 3
0.09
0.1
0.11
0.12
0.09
0.1
0.11
0.12
Distance (m)
p
r
e
s
s
u
r
e

(
b
a
r
)
o
Fig. 10. Variation in the inlet temperature.
2144 D. Colorado et al. / Applied Energy 88 (2011) 21362145
Comparisons of experimental and simulated data in steady-
state are presented showing a good degree of correlation. Further
studies will try to adapt this dynamic model for heat transformer
technologies, with the objective of future design, optimization,
control and on-line estimation of COP.
According to physical model analysis, mass ow rate in the
internal pipe is very important parameter to determine the in-
crease of the external heat ux. Perturbations realized in this work
to helical double-pipe condenser not modify outlet liquid temper-
ature of condenser. Therefore, not increased temperature in the
evaporator and not inuences the COP.
The numerical model developed is based on the applications of
governing equations and used general empirical correlations; for
this reason, it is possible to make use of it to other uids, mixtures
and operating conditions (including two-phase ow); it allows
using the model developed as an important tool to design and opti-
mize these kinds of systems.
Acknowledgments
The authors thank the Consejo Nacional de Ciencia y Tecnologa
(CONACyT Project SEP-2004-C01-48024) and PROMEP-UAEMOR-
PTC-158 for the economic support received for the development
of this research.
References
[1] Garca-Valladares O. Review of in-tube condensation heat transfer correlations
for smooth ad micron tubes. Heat Transfer Eng 2003;4:624.
[2] Zhao L, Guo L, Bai B, Hou Y, Zhang X. Convective boiling heat transfer and two-
phase ow characteristics inside a small horizontal helically coiled tubing
once-through steam generator. Int J Heat Mass Transfer 2003;46:477988.
[3] Wang Chi-chuan. A numerical method for thermally non-equilibrium
condensing ow in a double-pipe condenser. Appl Therm Eng
1997;17:64760.
[4] Morales-Ruiz S, Rigola J, Prez-Segarra CD, Garca-Valladares O. Numerical
analysis of two-phase ow in condensers and evaporators with special
emphasis on single-phase/two-phase transition zones. Appl Therm Eng
2008;29:103242.
[5] Colorado-Garrido D, Santoyo-Castelazo E, Hernndez JA, Garca-Valladares O,
Siqueiros J, Juarez-Romero D. Heat Transfer of a helical double-pipe vertical
evaporator: theoretical analysis and experimental validation. Appl Energy
2009;86:114453.
[6] Ali AHH. Design of a compact absorber with a hydrophobic membrane
contactor at the liquid vapor interface for lithium bromide water absorption
chillers. Appl Energy 2010;87:111221.
[7] Voyiatzis E, Palyvos JA, Markatos NC. Heat-exchanger design and switching-
frequency effects on the performance of a continuous type solar adsorption
chiller. Appl Energy 2008;85:123750.
[8] Gao J, Zhang X, Liu J, Li K, Yang J. Numerical and experimental assessment of
thermal performance of vertical energy piles: An application. Appl Energy
2008;85:90110.
[9] Lee TS, Lu WCh. An evaluation of empirically-based models for predicting
energy performance of vapor-compression water chillers. Appl Energy
2010;87:348693.
[10] Li YM, Wua JY, Shiochi S. Experimental validation of the simulation module of
the water-cooled variable refrigerant ow system under cooling operation.
Appl Energy 2010;87:151321.
[11] Wang J. Experimental investigation of the transient thermal performance of a
bent heat pipe with grooved surface. Appl Energy 2009;86:20307.
[12] Santoyo-Gutierrez S, Siqueiros J, Heard C, Santoyo E, Holland FA. An
experimental integrated absorption heat pump efuent purication system.
Part I: Operating on water/lithium bromide solutions. Appl Therm Eng
1999;19:46175.
[13] Holland FA. Fundamentos y aspectos econmicos de las bombas de calor,
manual sobre tecnologa de bombas de calor, program of cooperation and
investigation between Instituto de Investigaciones Elctricas. Cuernavaca
(UK): Mxico and Salford University; 1999.
[14] Sencan A, Yakut KA, Kalogirou SA. Exergy analysis of lithium bromide/water
absorption systems. Renew Energy 2005;30:64557.
[15] Sozen A, Serder Yucesu H. Performance improvement of absorption heat
transformer. Renew Energy 2006;32:26784.
[16] Sozen A, Arcaklioglu E. Exergy analysis of an ejector-absorption heat
transformer using articial neural network approach. Appl Therm Eng
2007;27:48191.
[17] Huicochea A, Siqueiros J. Increase of COP for an experimental heat transformer
using a water purication system. Desal Water Treat 2009;12:30512.
[18] Corro FO, Mendiola OF, Muoz RF, Obregon DJ, Santander RL, Torres GJ. Gua
tcnica sobre trazabilidad e incertidumbre en los servicios de calibracin de
manmetros, transductores y transmisores de presin de elementos
elsticos. Centro Nacional de Metrologa; 2008.
[19] Bevington PR, Robinson DK. Data reduction and error analysis for the physical
science. 3rd ed. New York: McGraw Hill; 2003.
[20] Huicochea A. Puesta en marcha y evaluacin experimental de un sistema
porttil de puricacin de agua integrado a un transformador trmico. Master
thesis. Mxico: Centro Nacional de Investigacin y Desarrollo tecnolgico
(CENIDET); 2004.
[21] Garca-Valladares O. Numerical simulation and experimental validation of
coiled adiabatic capillary tubes. Appl Therm Eng 2007;27:106271.
[22] Garca-Valladares O, Prez-Segarra CD, Rigola J. Numerical simulation of
double-pipe condensers and evaporators. Int J Refrig 2004;27:65670.
[23] International Association for Properties of water and steam industrial
formulation (IAPWS IF-97); 1997.
[24] Wongwises S, Polsongkram M. Condensation heat transfer and pressure drop
of HFC-134a in a helically coiled concentric tube-in-tube heat exchanger. Int J
Heat Mass Transfer 2006;49:438698.
[25] Yu B, Han JT, Kang HJ, Lin CX, Awwad A, Ebadian MA. Condensation heat
transfer of R134a ow inside helical pipes at different orientations. Int Comm
Heat Mass Transfer 2003;30:74554.
[26] Guo L, Feng Z, Chen X. An experimental investigation of the frictional pressure
drop of steam-water two-phase ow in helical coils. Int J Heat Mass Transfer
2001;44:260110.
[27] Al-Hajeri MH, Koluib AM, Mosaad M, Al-Kulaib. Heat transfer performance
during condensation of R134a inside helical coils. Energy Convers Manage
2007;48:230915.
[28] Ko TH. Thermodynamic analysis of optimal curvature ratio for fully developed
laminar forced convection in a helical coiled tube with uniform heat ux. Int J
Therm Sci 2006;45:72937.
[29] Rogers GF, Mayhew YR. Heat transfer and pressure loss in helically coiled
tubes with turbulent ow. Int J Heat Mass Transfer 1964;7:120716.
[30] Premoli A, Francesco DD, Prina A. A dimensional correlation for evaluating
two-phase mixture density. La Termotecnia 1971;25:1726.
[31] Tang L, Ohadi M M, Johnson AT. Flow condensation in smooth and micron
tubes with HCFC-22, HFC-134a, and HFC-410 refrigerants. Part II: Design
equations. J Enhanced Heat Transfer 2000;7:31125.
[32] Huicochea A. Estudio Terico-experimental de un sistema para puricacin de
agua integrado a un transformador de calor. PhD thesis. Mxico: Universidad
Autnoma del Estado de Morelos (UAEM); 2009.
[33] Escobar RF, Hernndez JA, Astorga-Zaragoza CM, Colorado D, Jurez D,
Siqueiros J. Analysis of the evaporator heat ow rate in a water purication
process integrated to an absorption heat transformer with energy recycling.
Desal Water Treat 2009;5:5967.
[34] Escobar RF, Jurez D, Siqueiros J, Irles C, Hernndez JA. On-line COP estimation
for waste energy recovery heat transformer by water purication process.
Desalination 2008;222:67783.
[35] Escobar RF, Uruchurtu J, Jurez D, Siqueiros J, Hernndez JA. On-line indirect
measures estimation for the performance of an absorption heat transformer
integrated to a water purication process. Measurement 2009;42:46473.
D. Colorado et al. / Applied Energy 88 (2011) 21362145 2145