Energy and Buildings 133 (2016) 754–769

Contents lists available at ScienceDirect

Energy and Buildings

journal homepage: www.elsevier.com/locate/enbuild

An innovative Trombe wall as a passive heating system for a building

in Athens—A comparison with the conventional Trombe wall and the
insulated wall
Evangelos Bellos ∗ , Christos Tzivanidis, Eleni Zisopoulou, Georgios Mitsopoulos,
Kimon A. Antonopoulos
Thermal Energy Department, School of Mechanical Engineering, National Technical University of Athens, Athens, Greece

a r t i c l e i n f o a b s t r a c t

Article history: Solar energy utilization for covering the heating loads of buildings is an innovative and clean way which
Received 8 June 2016 leads to lower electricity consumption. Passive heating systems are low-cost and simple technologies
Received in revised form 15 October 2016 which are able to cover a great part of the space heating demand. Trombe wall is the most mature
Accepted 21 October 2016
technology for capturing solar irradiation and transferring it inside the building in order to cover a part
Available online 22 October 2016
of the heating loads, especially the hours after the noon. In this study, an innovative Trombe wall with an
extra window in the massive wall is suggested. This wall is compared with the conventional Trombe wall
Keywords:
and the usual insulated wall in order to make a complete investigation. The examined walls are located
Passive heating
Trombe wall
in south direction, in order to maximize the impact of solar irradiation. According to the final results,
Solidworks the new Trombe wall is the most appropriate technology, creating warmer indoor profile than the other
Thermal behavior cases, especially the hours between noon and afternoon. The analysis in performed with the commercial
software Solidworks Flow Simulation and the examined building is located in Athens, Greece.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction energy can be used for covering the domestic hot water demand,
the space heating demand, the space cooling demand via sorption
The energy consumption in building sector consists the 30–40% machines and the electricity production, mainly with photovoltaic
of the global energy consumption [1–3]. About the half of this collectors.
energy amount is responsible for covering the heating and the In this direction, an intelligent solution is the optimization of
cooling loads of the buildings in order to achieve the desirable the building envelope in order to reduce the heating-cooling loads
thermal comfort conditions [1]. The impact of the buildings energy and to manage the incident solar energy to the building properly.
consumption in global greenhouse emissions is high and for this The use of insulation layers in the external structural components
reason, its reduction is a crucial issue of this time period. US fed- is the building protection to the heat losses in the winter and to
eral policy plans to reduce this energy about 70% up to 2020 [4] heat invasion in the summer. Generally, the use of an insulation
and China sets targets for reducing the energy consumption of new layer of 4–6 cm in external structural elements with thermal con-
buildings to 50% [5]. Simultaneously, the increasing rates in fossil ductivity of 0.03–0.04 W/mK can decrease the heating and cooling
fuels and in electricity consumption are extra reasons for creating loads about 2–3 times, in comparison with the uninsulated building
more efficient buildings. cell [6]. The utilization of structural components with high thermal
The use of renewable energy sources in buildings is a sustainable mass increases the storage capacity of the building and it is another
way to mitigate the environmental and the financial issues related way for reducing the heating and cooling loads. This technique is
to high energy consumption. Solar energy, geothermal energy and ideal for warm climates with high daily range ambient temperature
wind energy are the most abundant renewable energy sources with distribution [7], but this technique is not preferable for cold cli-
the first to be the most easily utilized in the building sector. Solar mates with low ambient temperature levels during the day [8]. The
use of phase change materials is an alternative way to manage the
incident solar irradiation properly and to reduce the building loads.
More specifically, by selecting the proper materials and locating
∗ Corresponding author. them in the suitable places in the building cell, a great improvement
E-mail address: [email protected] (E. Bellos).

http://dx.doi.org/10.1016/j.enbuild.2016.10.035
0378-7788/© 2016 Elsevier B.V. All rights reserved.
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 755

operates as absorber for heating air. This air quantity circulates into
Nomenclature the building, due to buoyancy forces, transferring solar energy into
the indoor space [11]. Green walls or roofs are old solutions for
c Specific thermal capacity of the structural material decreasing the indoor temperature during the summer [12], giving
[J/kgK] extra advantages as acoustic protection and better esthetic image
cp Specific thermal capacity [J/kgK] [11].
G Solar irradiation [W/m2 ] In the literature, Trombe wall is the most analyzed passive sys-
g Radiation heat-transfer factor [W/m2 K] tem and many different configurations have been applied. In the
Gr Grashof number classic case, there are holes in the massive wall in order to create
h Heat convection coefficient [W/m2 K] air circulation. During the winter, the circulation is internal, while
IR Infiltration rate [air changes/h] in the summer there is connection to the environment in order to
k Thermal conductivity [W/mK] enhance the ventilation rate [13]. Abbassi et al. [14] studied the use
L Layer thickness [m] of Trombe wall for Tunisian climate and concluded that 4 m2 of this
M Indoor air mass [kg] wall can reduce the annually heating auxiliary energy about 50%
N Number of layers in every structural component and 8 m2 can lead to 77% reduction. Bojic et al. [15] studied a Mozart
NRL Normalized radiative loss [◦ C] house in Lyon-France and proved that the use of a Trombe wall in
Pr Prandtl number the south direction can reduce the yearly heating consumption up
Qin Indoor loads [W] to 20%.
t Time [h] In the literature, there are also many studies related to improved
T Temperature [◦ C] or modified Trombe walls. The use of internal fins in an unvented
U Thermal transmittance [W/m2 K] Trombe was analyzed by Abbassi and Dehmani [16]. They compared
V Building internal volume [m3 ] the innovative Trombe wall with the classical unvented Trombe
x Cartesian coordinate vertical to the structural com- wall and they proved that the use of internal fins increases the
ponent area [m] indoor temperature and decreases the massive wall temperature.
Hu et al. [17] investigated the use of venetian blinds between the
Greek symbols
glazing and the thermal mass wall, and they proved that the tilt
␣ Absorbance
angle of the blinds plays a significant role in the natural convec-
␧ Emittance
tion in the air gap. Hong et al. also studied [18] the use of venetian
␳ Density [kg/m3 ]
blinds in Trombe walls and they optimized their system. Accord-
␶ Transparency
ing to the results, the optimum distance between blind and glass
is 9 cm, when the air gap is about 14 cm. Duan et al. [19] examined
Subscripts and superscripts
the insert of an absorber plate in the middle of the air gap and not
air Indoor air
in the massive wall surface. They finally proved that this config-
am Ambient air
uration performs better energetically and exegetically, compared
av Daily average value
to the usual Trombe wall. Rabani et al. [20] designed an innova-
c Cover-glazing
tive Trombe wall which takes solar irradiation from East, West and
i Indoor
South direction and covers the half part of the south wall. This sys-
max Daily maximum value
tem is low cost because of its lower surface and it leads to acceptable
min Daily minimum value
thermal comfort conditions, according to this study. Shen et al. [21]
out Outdoor
compared the classical Trombe wall with a composite Trombe wall
sw South wall
and they concluded that the second has better performance in cold
trom Trombe
and cloudy weather. Zhou and Pang [22] designed an innovative
v Indoor surfaces
Trombe wall with phase change materials (PCM) in the absorber
w Wall
surface and with vortex generators in this area. The final results
showed real enhancement of the heat transfer in the air gap and
increased heat rate to the building space.
in the building thermal behavior can be achieved [9,10]. However, All these studies have been focused in the thermal enhancement
the high cost of these materials creates difficulties in their extended of the heat transfer phenomena in the air gap and to the suitable
usage. management of the stored energy. In this study, a new Trombe wall
Passive heating and cooling systems exploit the solar irradiation with extra glazing in the massive wall is proposed and compared to
and the ambient temperature fluctuations in order to store energy the classic Trombe wall and to the conventional insulated wall. The
and to transmit it in the indoor space the preferable time and with use of an extra glazing is able to enhance the heat transfer from
the suitable rate. The most established technology in this field is the air gap to the indoor space and to let a part of solar irradia-
the Trombe wall; a massive wall which is covered by an external tion to heat directly the internal space. Moreover, this configuration
glazing area. This system is similar to solar thermal collector and the increases the physical lighting in the space, something that solves
basic idea is the solar energy storage to a massive wall for utilization partially the aesthetics problems of Trombe wall [22]. The simu-
the hours after the sunset. Other prevalent passive systems are the lations are performed with software SolidWorks® FlowSimulation
solar chimney, the unglazed transpired solar façade and the green which is useful design and simulation tool, especially for cases
walls. where solar energy has to be taken into consideration. A small
The solar chimney operation is based on the airflow generation building is designed and simulated for 6 days, one for every winter
due to buoyancy. The thermal energy from the sun is converted to period month, in order to investigate the examined system for all
kinetic energy of the air and natural ventilation is created, giving the preferable period. Weather data for Athens are taken from liter-
solutions during the summer months [11]. The unglazed transpired ature and the specific geographical position (␸ = 38◦ ) was inserted
solar façade is a simple and low-cost system which utilizes the solar in the simulation program in order to make a realistic analysis.
irradiation with a high rate. In this system, a metal sheet with holes
is located in a part of the external building area and this surface
756 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Table 1
Structural component composition and dimensions.

Component Insulated Wall Roof Trombe wall

Dimensions 5m×3m 5 m × 5m 5m×3m

Outer layer 7 cm brick 7 cm heavyweight concreate 20 cmheavyweight concreate
Intermediate layer 6 cm insulation 6 cm insulation
Inner layer 7 cm brick 7 cm lightweight concreate
U-vaue 0.52 W/m2 K 0.50 W/m2 K 1.81 W/m2 K

Table 2 Table 3
Thermal properties of the materials. Thermal transmittance values for the structural components.

Material k (W/mK) ␳ (kg/m3 ) cp (J/kgK) Component Wall Roof Window Trombe New Trombe
2
Brick 0.62 1700 800 U (W/m K) 0.52 0.50 5.20 1.81 2.26
Insulation 0.04 200 670
Heavyweight Concreate 1.13 2000 1000
Lightweight Concreate 0.38 1200 1000
Glass 1.05 2600 670 Table 4
Building structure and simulation parameters.

Parameter Value

2. The examined cases Height 3m

Length 5m
Width 5m
The examined building of this study is a small square room of
East window 3 m2
25 m2 area. There are four external walls oriented in the four direc- West window 3 m2
tions (south, east, north, west), a roof and a ground floor. Windows New Trombe window 2 m2
have been placed in the east and in the west sides, while the south Glazing-wall distance 50 mm
Glass thickness 5 mm
side is parametrically analyzed. Three configurations for the south
Infiltration rate 3 air changes/h
wall are examined; an insulated wall, a classical Trombe wall and a ␧c 0.88
new design Trombe wall with an extra window in the massive wall. ␶c 0.90
This new configuration presents many advantages, fact that makes ␣w 0.70
this analysis very interesting. Firstly, this design lets the heat to be ␧w 0.70
hout 10 W/m2 K
transferred in the internal space easily and by this way the time
Sky clearness 60%
lag can be regulated. Moreover, lighting can invade in the indoor Athens latitude 38◦
space, increasing the esthetic and the living quality for the resi- Ground Adiabatic
dents. The following figures illustrate the examined cases. Fig. 1 Time Solar time
shows the simple building with the insulated south wall and will be
referred as “insulated” in this analysis. Figs. 2 and 3 shows the clas-
sic Trombe wall case and the new Trombe wall case with the extra
window. These cases will be referred as “Trombe” and “Trombe while Table 4 includes other parameters of the building cell and of
new” respectively. the simulation process. It is essential to state that parameters as the
The only difference between the examined cases is the south absorbance and the emittance were selected to have typical values.
wall. The other structure components are the same in order to make It is important to state that the analyzed Trombe walls are
a realistic comparison. Table 1 includes the dimensions and the unvented in this analysis. These systems are simplest compared
composition of the structural elements of this study. It is essential to the respective with holes for natural circulation and this was the
to state that the ground and the columns were selected to be made first reason for selecting them. The second reason is for examining
from concreate. Moreover, the internal window is located in the if the internal window can act as a thermal-bridge to the stored
upper part of the wall because of the sun altitude in the winter. energy of the massive wall. In the case that this comparison is pre-
The properties of the materials are presented in Table 2. These sented for the vented system, this operation of the window will not
values have taken from the material library of Solidworks and they be examined properly, because the natural ventilation will increase
are corresponding to real structural materials. The most important the heat transfer from the massive wall to indoor space.
value is the thermal conductivity of the insulation layer which is
equal to 0.04 W/mK [23] and this parameter determines the qual-
ity of the building cell. This selected value is a usual value for the 3. Methodology
modern buildings and protects the building cell from heat losses.
The most important parameter for the evaluation of the struc- 3.1. Weather data and simulation days
tural components is the thermal transmittance or (U-value) which
determines how easily heat is transferred between ambient and In this section the methodology of this study will be presented.
indoor space. It is important to state that the windows are single The three designed buildings are tested for 6 typical days, one for
and for this reason their U-values are great. Moreover, the total every winter month. More specifically, the 21th day of November,
thermal transmittance of the new glazed Trombe is 25% greater December, January, February, March and April are tested in order
than the thermal transmittance of the classic massive wall without to determine the performance of the examined system in various
glass. This result shows that the use of an extra internal window cases. Fig. 4 exhibits the weather data and Fig. 5 the incident solar
increases the ability of this wall to transfer the heat from the air gap irradiation on the south wall for the examined days. The weather
to the indoor space. By this way, the great superheating of the mas- data are taken from literature data for Athens which have been used
sive wall during the hours of high irradiation levels can be reduced in many studies [9,24–26]. The solar irradiation has been calculated
and the response of this passive system is getting greater. Table 3 by Solidworks results by selecting the sky clearness equal to 60%.
includes the thermal transmittance of the structural components, Table 5 summarizes the weather data of this analysis.
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 757

Fig. 1. Building with insulated south wall.

Fig. 2. Building with the Trombe wall in the south direction.

Fig. 3. Building with the innovative Trombe wall with the extra window.

Table 5
Summarized weather data.

Months Day duration Tmax Tmin Tav

◦
◦ ◦
hours C C C

November 9.672 14.2 12.0 10.7

December 9.363 11.2 9.3 7.8
January 9.860 12.5 9.4 6.9
February 10.928 14.6 11.0 8.4
March 12.084 14.7 12.0 10.0
April 13.343 18.9 15.1 12.1

3.2. Simulation in solidworks environment

Fig. 4. Ambient temperature profile for the examined days.

The examined buildings were designed in Solidworks and were
simulated in its Flow simulation module. The last years, many stud-
ies have been performed with simulating tool. Valančius et al. [27] The simulation is based on solar time and the respective mod-
and Driss et al. [28] analyzed the thermal behavior of buildings with ifications of the weather data have been made. The ventilation of
Solidworks flow simulation in steady state conditions. Moreover, the building is achieved by creating to holes in the building cell and
there are many other simulations of the solar energy utilization determining the respective boundary conditions in the lids which
with Solidworks Flow Simulation [29–31]. The innovative point of cover these holes. A very important point of this study is the defi-
the present study is the simulation in the time-depended mode in nition of the mean indoor air temperature. Nine points inside the
order to simulate the real thermal behavior of the building. building were selected to be analyzed. The mean temperature of
758 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Fig. 5. Incident solar irradiation in the south wall for the examined days.

the indoor space is calculated as the mean temperature of these

nine points. Fig. 6a shows these points, which are the building
central point and all the midpoints between this and the tops of
the building cell. Fig. 6b shows the location of the holes for the
ventilation.
The next step is to present the main boundary conditions of
this study. Their proper determination is essential for taking the
desirable results.

a) Inlet mass flow rate and temperature of the working fluid in “Air
inlet” hole. Fig. 7. Model validation with a developed numerical model a) comparison for the
b) Static pressure in the “Air outlet” hole. insulated wall case b) comparison for the Trombe wall case.
c) Heat convection between the outer surfaces and the ambient air.
d) The ground has selected to be adiabatic.
3.3. Model validation
e) The proper radiation surfaces have been selected according to
the values of Table 4.
The results from the model in Solidworks were compared with
f) The windows have selected to be transparent to solar irradiation.
the results of a developed numerical model. The insulated wall
case and the conventional Trombe wall are investigated with a
The solar position is determined by selecting the proper axis for numerical model. The developed numerical model is based on the
south and zenith directions, as well as the examined day of the year finite-difference method and it is developed in Fortran. This model
in each case. The time step was finally selected to be 5 min, after a has been used in previous studies of building envelopes [32,33]
simple sensitivity analysis which is presented in Appendix A and and more details about this model are given in Appendix B. For the
especially in Fig. A1. The analysis was made for many consecutive present needs, some modifications in the program have been made.
days and finally is proved that 7 days are enough to converge the The properties of the materials, the building envelope dimensions,
indoor air temperature profile, according to Fig. A2 in Appendix A. as well as the air ventilation have been selected to be the same as
The mesh independence is also presented in the Appendix and in Solidworks model.
especially in Fig. A3. Finally, about 6 million cells were selected Fig. 7 illustrates the comparison between the results. The results
for this study and this mesh is given in Fig. A4. It is important to of the Solidworks model are given with red shades, while the results
state that a basic mesh is selected in the program menu and also from the numerical model are given with blue shades. The curves
extra refinements in the solid and fluid cells are inserted by the are close to each other for both cases, fact that proves the valida-
mesh generator of the simulation tool. This mesh is adequate for tion of the Solidworks model. The results are presented for January
this study, according to the sensitivity analysis which is presented which is the most characteristic month of the winter period. More-
in the Appendix A. over, the mean daily difference in the insulated wall case is about

Fig. 6. Simulation in Solidworks environment a) Points for indoor air temperature b) Inlet and outlet for the ventilation.
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 759

Fig. 8. Comparison of the indoor air temperature profile for the examined cases for all the 7-day simulation period.

0.15 K, while for the Trombe wall case is about 0.22 K; small devi- using a Trombe wall in the south direction. By comparing the two
ations which shows accordance between the results. The results examined Trombe walls, the new configuration with the internal
of the numerical model give higher indoor temperature because window leads to higher indoor temperature profile, especially the
the heat losses from the thermal bridges have not been taken into hours after the noon.
account in this model. This important result can be explained by the solar irradiation
invasion into the building the hours close to solar noon, when the
sun rays are mainly reaching to the south wall. The extra window
4. Results
leaves a part of the solar energy to heat the building directly and
to heat the internal air without time lag. Moreover, it is important
4.1. Comparison of the temperature profiles of the examined
to state that the use of Trombe walls can lead the indoor temper-
buildings
ature levels close to the desired thermal comfort limits, which are
between 20 and 22 ◦ C generally. On the other hand, the use of the
In this paragraph the comparison of the indoor temperature pro-
insulated wall leads to low temperature levels, keeping the indoor
files for the examined cases is presented. Fig. 8 depicts the indoor
space far from the desired conditions. Trombe wall aids to the solar
temperature for all the time period of the simulation (7 days) while
energy utilization and management, while the insulated wall blocks
Fig. 9 shows the results for the last day of the simulation. Both fig-
the solar energy transfer into the indoor space. This is the main
ures are consisted from 6 smaller images; each one for every month
reason for the results of Figs. 8 and 9.
of the winter period.
In a more detailed analysis, the Trombe walls are able to keep
Fig. 8 proves that all the examined cases are converged the 7th
the indoor space in high temperature levels the months November,
day for all the months, a very important result for the validity of this
February, March and April, while the insulated wall is far from the
study. Moreover, it is shown that the results in the 5th day are also
standards for all the months. The new Trombe wall gives about
acceptable, an outcome that can be adopted for future studies with
0.5 K greater daily temperature levels compared to the traditional
greater computational cost. Both Figs. 8 and 9 show that the case
Trombe wall and about 3.5 K greater to the insulated wall. These
with the insulated wall in the south gives a temperature profile
results are similar for all the months, a very important result which
far from the other two temperature profiles. This is the first and
enforces the advantages of the presented new configuration dur-
most significant conclusion of this study which proves the gain of
760 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Fig. 9. Comparison of the indoor air temperature profile for the examined cases for the 7th day of the simulation.

ing all the examined winter period. It is essential to state that the both from the ambient temperature and the solar irradiation and
greatest difference is observed in February with the new Trombe they present a smaller monthly variation, leading to more stable
wall case to have significant warmer indoor conditions compared results which are close to thermal comfort standards.
to the other cases.
4.3. Indoor air temperature distribution
4.2. Comparison of south wall and indoor temperature levels
The indoor air temperature profile is presented in this para-
In this paragraph, results for the south wall temperature are graph, and more specifically in Figs. 13–15. These figures exhibit
presented in order to explain better the Trombe wall operation. the indoor air temperature distribution and the structural compo-
Figs. 10–12 present results for the insulated wall case, the classic nents temperature for 8 moments during the day, in the middle
Trombe wall case and the new Trombe wall case respectively. In horizontal cross section. All these figures illustrate the indoor con-
every figure, there are 6 subfigures, one for every month, where ditions for the typical day of January, because this month is in the
the indoor temperature, the ambient temperature and the internal center of the winter period and it is the most representative month.
south wall temperature profiles are compared. It is important to state the left pallet in this figure corresponds to the
According to Fig. 10, the south wall temperature is low and it solid temperature levels and the right to the fluid (air) temperature
takes values in the range from 14 ◦ C to 20 ◦ C approximately. On the levels.
other hand, the internal side of the massive Trombe walls takes The general results of these figures are similar because the
value from 23 ◦ C to 34 ◦ C, according to Figs. 12 and 13. The values examined building is the same in its greatest part. The maximum
for the south wall temperature between the two Trombe walls are temperature is observed at 18:00 and the minimum at 3:00. In the
similar, something that explains the small difference in their indoor hours before and close to the noon, the east wall is warm and in the
air temperature profiles. afternoon the west wall becomes warm. Another interesting result
Another important result for the south wall temperature is is the superheating of the internal side of the north wall because of
extracted by the monthly comparison. In the case of the insulated the solar energy invasion from west and east windows.
south wall, the temperature of the internal wall side varies from Comparing these figures, the new Trombe wall case gives a
month to month because this is influenced mainly from the ambient warmer profile compared to the other two cases. However, the tem-
temperature. On the other hand, the Trombe walls are influenced perature levels for the two Trombe walls are similar because both
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 761

Fig. 10. Comparison among south wall, indoor air and ambient air temperature profiles for the insulated wall case.

these walls (Trombe and new Trombe) utilize the incidental solar 4.4. Summary of the presented results and discussion
energy on the south wall. The difference between them lies on the
management of solar energy with the new configuration to let the In this paragraph, the final comparison among the examined
direct utilization of the solar energy through the internal window. cases is presented. Table 6 includes the final results for the maxi-
Insulated south wall case is the one with the lower temperature mum, minimum and average monthly temperature values for the
levels during the day. The indoor space is very cold except from the indoor air temperature and for the ambient temperature. The most
afternoon hours. This results show that the conventional building important result is that the new Trombe configuration leads to a
structure does not manage the solar irradiation properly in order mean indoor temperature of 18.5 ◦ C, greater than the temperature
to transfer it in the indoor space because the insulation layer acts of 18 ◦ C for the classic Trombe case. The insulated wall presented
as a thermal shield. mean temperature of 14.9 ◦ C, lower than the previous case but
Trombe walls give satisfying indoor temperature profiles the greater than the mean ambient temperature of 11.5 ◦ C.
hours from noon up to the midnight, the most important time The difference of 0.5 K between the two Trombe walls is about
period for the residential buildings. The temperature of the mas- constant during the winter period and this result is very important
sive wall is high in this time period and this result validates the because the new configuration gives better results independently
successful operation of this technology. The worst time period is from the ambient conditions. The daily range in the cases with
during the night where the stored solar energy has been depleted. Trombe walls is greater than in the insulated wall case because
A difference between air temperature profiles of Figs. 14 and 15 the solar utilization leads to great fluctuations of the indoor tem-
is in the region close to the north wall, especially the hours after perature profile during the day.
noon. The extra window let the solar irradiation to reach in the The comparison of the model in Solidworks with the numer-
north wall and to heat it, a result which increases the air temper- ical model (paragraph 3.3) proves that the results are validated.
ature locally (Fig. 15). Moreover, this result proves that the indoor Moreover, it is interesting to state that in the study of Abbassi and
temperature of the building is more uniform in the pictures of Dehmani [16], a modification in the Trombe wall has resulted to
Fig. 15 than in the pictures of Fig. 14, because in the first case there indoor temperature increase of about 1 K. In the present study, the
are two warm regions which heat the indoor space; one close to indoor temperature increase is about 0.5 K, a value which is in the
south and one close to north wall. same levels.
762 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Fig. 11. Comparison among south wall, indoor air and ambient air temperature profiles for the classic Trombe wall case.

Table 6
Final comparison of the results.

T (◦ C) Ambient Insulated Trombe Trombe New

Max Aver Min Max Aver Min Max Aver Min Max Aver Min

Nov 14.2 12.0 10.7 16.4 15.0 13.7 20.8 18.6 16.6 21.5 19.1 16.6
Dec 11.2 9.3 7.8 13.4 12.0 12.0 17.9 15.7 14.0 18.8 16.3 14.1
Jan 12.5 9.4 6.9 14.1 12.4 12.4 18.4 16.1 14.1 19.5 16.6 14.1
Feb 14.6 11.0 8.4 16.4 14.6 14.6 21.0 18.0 15.5 21.5 18.6 16.1
Mar 14.7 12.0 10.0 17.5 15.9 15.9 21.4 18.6 16.5 21.5 19.2 16.9
Apr 18.9 15.1 12.1 21.5 19.5 19.5 23.2 20.9 18.6 23.4 21.2 19.1
Mean 14.4 11.5 9.3 16.6 14.9 14.7 20.5 18.0 15.9 21.0 18.5 16.6

5. Conclusions proves the stability of this satisfying performance. The daily range
in both Trombe walls were found greater than the respective in the
In this study, an innovative modification in the unvented insulated wall case, because the solar energy utilization leads to
Trombe wall is presented and the new system is compared with greater daily fluctuations in the temperature profiles.
the classic Trombe wall and the insulated wall. Three building cells The maximum temperature in the building is observed at 18:00
are designed in Solidworks and they are simulated with its Flow for all the examined cases, because of the time lag between the solar
Simulation module. The simulations were carried out for 6 typical energy capturing and its transferring into the building. For improv-
days, one for every month of the winter period, in order to examine ing this situation, the new Trombe wall let the solar irradiation to
fully the operation of the new design for space heating proposes. be partially transferred into the space during the hours close the
The final results proved that this new configuration increases noon and the result is the better solar energy management.
the indoor air temperature the time period from noon to lately in According to the results, this new configuration is beneficial
the afternoon and the mean daily increase is about 0.5 K. This result because the indoor temperature is greater compared to the con-
is approximately the same for all the examined months, fact that ventional cases and the lighting of the internal space is improved.
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769
Fig. 12. Comparison among south wall, indoor air and ambient air temperature profiles for the new Trombe wall case.

763
764 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Fig. 13. Cross section for the indoor temperature during the January day for the insulated wall case (left pallet for solid and right pallet for fluid).
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 765

Fig. 14. Cross section for the indoor temperature during the January day for the classic Trombe wall case (left pallet for solid and right pallet for fluid).
766 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Fig. 15. Cross section for the indoor temperature during the January day for the new Trombe wall case (left pallet for solid and right pallet for fluid).

For these reasons, this design can be examined more in the future wall and it is similar for the other cases. Fig. A1 illustrates the time
and it will be very interesting to test its performance in the vented independence and it is obvious that the time step of 5 min leads
mode and in the summer period. to a smooth profile. Fig. A2 exhibits that the indoor air tempera-
ture profile is about the same between the 6th and the 7th day of
Acknowledgment the simulation; fact that leads to the selection of 7 days for conver-
gence. According to Fig. A3, 6 million cells lead to a more smooth
The first author would like to thank Onassis foundation for its profile of the indoor air temperature, compared to the less dense
financial support. meshes. This mesh was selected as the suitable, because a greater
will lead to a huge computational time and the analysis will not be
Appendix A.
feasible to be made. The last figure of this Appendix is the Fig. A4,
where the final selected mesh is presented. The fluid cells are illus-
In this section, the mesh independence is shown in Figs. A1–A4,
trated with blue color, while the solid cells with the black one. It
. This analysis is presented for the building with classic Trombe
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 767

Fig. A1. Indoor air temperature for the 7th day of building with Trombe wall for various time steps and 6 million cells.

Fig. A2. Indoor air temperature in the seven simulation days, for building with Trombe wall, for time step equal to 5 min and for 6 million cells.

is essential to state that the fluid cells are about 5 million and the This equation indicates that the convection heat flux between
solid cells are about 1 million. the ambient and the wall is equal to heat conduction in the start of
the wall. In order to take into account the solar irradiation, the next
Appendix B. equivalent temperature has been utilized:
a·G
In this Appendix the basic information about the used numerical Tout (t) = Tam (t) + − NRL, (B3)
model are given. This model is a one-dimensional model which hout
is based on the heat conduction differential equation. This model NRL expresses normalized radiative loss, as analyzed in Ref. [34].
has been described with many details in [32,33] and here the main For the last node if the external layer “N” with thickness LN , there
points of this methodology are given. is the following boundary condition:
For every structural component of the building envelope (exter-
∂T N (t, LN )     
nal walls and roof), the heat conduction differential equation for the −k · = hin · T 1 (t, 0) − Tin (t) + gv · T N (t, LN ) − Tv (t) , (B4)
∂x
layer “j” can be written as: v

2
∂T j (t, x) ∂ T j (t, x) In Eq. (B4), the heat conduction in the internal side of the wall
·c· =k· , (B1) is equal to the convection with the indoor air and the radiation
∂t ∂x2
exchange with the other internal surfaces (v) of the building enve-
For the first node of the external layer “1”, there is the following
lope. The coefficient gv can be calculated according to Eq. (B5) for
boundary condition:
every couple of surfaces [35]:
∂T 1 (t, 0)    0.1
−k · = hout · Tout − T 1 (t, 0) , (B2) gv = 8 · ε · T N (t, LN ) − Tv (t) , (B5)
∂x
768 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769

Fig. A3. Indoor air temperature for various meshes with time step equal to 5 min and for the 7th day of the simulation.

For this wall, the boundary condition of the Eq. (B2) can be mod-
ified using the (hout )Trombe and not the (hout ). Eq. (B8) shows the
calculation of the (hout )trom :

1
(hout )trom = 1
L 1
, (B8)
hout
+ k glass
+ hgap

This methodology has been described in Ref. [38] with more

details.
The infiltration load can be calculated by Eq. (B9):

IR · V · air
Qinf (t) = · cp,air · [Tin (t) − Tam (t)] , (B9)
3600
The indoor temperature can be calculated by the energy balance
in the indoor space. All the loads have to be taken into account in
order to calculate the change in the indoor temperature level, as
Eq. (B10) shows:

Qin
Tin (t + t) = Tin (t) + , (B10)
Mair · cp,air
Fig. A4. Mesh in the computational domain; with blue color the fluid cells and with
black color the solid cells. (For interpretation of the references to colour in this figure References
legend, the reader is referred to the web version of this article.)
[1] H. Omrany, A. Ghaffarianhoseini, A. Ghaffarianhoseini, K. Raahemifar, J.
Tookey, Application of passive wall systems for improving the energy
effciency in buildings: a comprehensive review, Renew. Sustain. Energy Rev.
62 (2016) 1252–1269.
The indoor heat convection coefficient (hin ) is calculated accord- [2] R. Zeng, X. Wang, H. Di, F. Jiang, Y. Zhang, New concepts and approach for
ing to the following equation [36]: developing energy efficient buildings: ideal specific heat for building internal
thermal mass, Energy Build. 43 (2011) 1081–1090.
[3] E. Bellos, C. Tzivanidis, K. Moschos, K.A. Antonopoulos, Energetic and financial
evaluation of solar assisted heat pump space heating systems, Energy
  16

0.25 6   13
6
Convers. Manage. 120 (2016) 306–319.
[4] H.S. Knowles III, Realizing residential building greenhouse gas emissions
T N (t, LN ) − Tin (t)
hin = 1.5 · + 1.23 · T N (t, LN ) − Tin (t) , reductions: the case for a web-based geospatial building performance and
3
social marketing tool. In: Proceedings of the 17th annual international
(B6) emission inventory conference: inventory evolution-portal to improved air
quality the United States Environmental Protection Agency (EPA). www.
epagov/ttnchie1/conference/ei17/session5/knowlespdf;2008.
[5] P.J.F. Nejat, M.M. Taheri, M. Gohari, M.Z.A. Majid, Global review of energy
consumption, CO2 emissions and policy in the residential sector (with an
For the Trombe wall case, the heat transfer coefficient in the air overview of the top ten CO2 emitting countries), Renew. Sustain. Energy Rev.
43 (2015) 843–862.
gap is calculated as [37]: [6] T.K. Papakostas, M.A. Papadopoulos, G.I. Vlahakis, Optimisation of thermal
protection in residential buildings using the variable base degree-days
kair   method, Int. J. Sustain. Energy 24 (1) (2005) 19–31.
hgap = · 0.01711 · (Gr · Pr)0.29 , (B7) [7] D.M. Ogoli, Predicting indoor temperatures in closed buildings with high
Lgap thermal mass, Energy Build. 35 (9) (2003) 851–862.
 E. Bellos et al. / Energy and Buildings 133 (2016) 754–769 769

[8] A. Dodoo, L. Gustavsson, R. Sathre, Effect of thermal mass on life cycle primary [24] D.A. Kouremenos, K.A. Antonopoulos, E.S. Domazakis, Solar radiation
energy balances of a concrete- and a wood-frame building, Appl. Energy 92 correlations for the Athens, Greece, area, Sol. Energy 35 (1985) 259–269.
(2012) 462–472. [25] D.A. Kouremenos, K.A. Antonopoulos, S. Doulgerakis, Direct and diffuse solar
[9] C. Tzivanidis, K.A. Antonopoulos, F. Gioti, Numerical simulation of cooling radiation correlations for tilted surfaces in Athens, Greece, Sol. Energy 38
energy consumption in connection with thermostat operation mode and (1987) 203–217.
comfort requirements for the Athens buildings, Appl. Energy 88 (8) (2011) [26] E. Bellos, C. Tzivanidis, A. Delis, K.A. Antonopoulos, Comparison of a solar
2871–2884. driven heat pump heating system with other typical heating systems with
[10] E.D. Kravvaritis, K.A. Antonopoulos, C. Tzivanidis, E. Bellos, Solar energy TRNSYS, in: ECOS Conference 2015, Pau, France, 2015.
management using phase change materials passive systems in the athens [27] K. Valančius, V. Motuzienė, S. Paulauskaitė, Redeveloping industrial buildings
area buildings, Int. J. Mech. Syst. Eng. 1 (2015) 102, http://dx.doi.org/10. for residential use: energy and thermal comfort aspects, Energy Sustain. Dev.
15344/2455-7412/2015/102. 29 (2015) 38–46.
[11] H.Y. Chan, S.B. Riffat, J. Zhu, Review of passive solar heating and cooling [28] S. Driss, Z. Driss, I.K. Kammoun, Computational study and experimental
technologies, Renew. Sustain. Energy Rev. 14 (2) (2010) 781–789. validation of the heat ventilation in a living room with a solar patio system,
[12] J. Price, Green Façade Energetics, University of Maryland, College Park, Energy Build. 119 (2016) 28–40.
UnitedStates − Maryland, 2010. [29] C. Tzivanidis, E. Bellos, D. Korres, K.A. Antonopoulos, G. Mitsopoulos, Thermal
[13] F. Stazi, A. Mastrucci, C.D. Perna, Trombe wall management in summer and optical efficiency investigation of a parabolic trough collector, Case Stud.
conditions: an experimental study, Sol. Energy 86 (9) (2012) 2839–2851. Therm. Eng. 6 (2015) 226–237.
[14] F. Abbassi, N. Dimassi, L. Dehmani, Energetic study of a Trombe wall system [30] E. Bellos, C. Tzivanidis, K.A. Antonopoulos, G. Gkinis, Thermal enhancement of
under different Tunisian building configurations, Energy Build. 80 (2014) solar parabolic trough collectors by using nanofluids and
302–308. converging-diverging absorber tube, Renew. Energy 94 (2016) 213–222.
[15] M. Bojić, K. Johannes, F. Kuznik, Optimizing energy and environmental [31] E. Bellos, C. Tzivanidis, D. Korres, K.A. Antonopoulos, Thermal analysis of a flat
performance of passive Trombe wall, Energy Build. 70 (2014) 279–286. plate collector with Solidworks and determination of convection heat
[16] F. Abbassi, L. Dehmani, Experimental and numerical study on thermal coefficient between water and absorber, in: ECOS Conference, Pau, France,
performance of an unvented Trombe wall associated with internal thermal 2015.
fins, Energy Build. 15 (2016) 119–128. [32] K.A. Antonopoulos, F. Gioti, C. Tzivanidis, A transient model for the energy
[17] Z. Hu, W. He, X. Hong, J. Ji, Z. Shen, Numerical analysis on the cooling analysis of indoor spaces, Appl. Energy 87 (2010) 3084–3091.
performance of a ventilated Trombe wall combined with venetian blinds in an [33] C. Tzivanidis, K.A. Antonopoulos, F. Gioti, Numerical simulation of cooling
office building, Energy Build. 126 (2016) 14–27. energy consumption in connection with thermostat operation mode and
[18] X. Hong, W. He, Z. Hu, C. Wang, J. Ji, Three-dimensional simulation on the comfort requirements for the Athens buildings, Appl. Energy 88 (2011)
thermal performance of a novel Trombe wall with venetian blind structure, 2871–2884.
Energy Build. 89 (2015) 32–38. [34] American Society of Heating Refrigerating and Air-conditioning Engineers,
[19] S. Duan, C. Jing, Z. Zhao, Energy and exergy analysis of different Trombe walls, ASHRAE Fundamentals, Atlanta, GA, 1993.
Energy Build. 126 (2016) 517–523. [35] R. Karadag, The investigation of relation between radiative and convective
[20] M. Rabani, V. Kalantar, A.A. Dehghan, A.K. Faghih, Experimental study of the heat transfer coefficients at the ceiling in a cooled ceiling room, Energy
heating performance of a Trombe wall with a new design, Sol. Energy 118 Convers. Manage. 50 (2009) 1–5.
(2015) 359–374. [36] F. Alamdari, G.P. Hammond, Improved data correlations for buoyancy-driven
[21] J. Shen, S. Lassue, L. Zalewski, D. Huang, Numerical study on thermal behavior convection in rooms, Build. Serv. Eng. Res. Technol. 4 (1983) 106–112.
of classical or composite Trombe solar walls, Energy Build. 39 (2007) 962–974. [37] K.R. Randal, J.W. Mitchel, M.M. Wakil, Natural convection heat transfer
[22] O. Saadatian, K. Sopian, C.H. Lim, N. Asim, M.Y. Sulaiman, Trombe walls: a characteristics of flat-plate enclosures, J. Heat Transf. 101 (1979) 120–125.
review of opportunities and challenges in research and development, Renew. [38] S. Kalogirou, in: Soteris A. Kalogirou (Ed.), Solar Energy Engineering, Academic
Sustain. Energy Rev. 16 (8) (2012) 6340–6351. Press, Boston, 2009.
[23] C. Tzivanidis, E. Bellos, G. Mitsopoulos, K.A. Antonopoulos, A. Delis, Energetic
and financial evaluation of a solar assisted heat pump heating system with
other usual heating systems in Athens, Appl. Therm. Eng. 106 (2016) 87–97.