Construction and Building Materials 178 (2018) 360–371

Contents lists available at ScienceDirect

Construction and Building Materials

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

Experimental and numerical study on heating performance of the mass

and thin concrete radiant floors with ground source systems
Mohammad Tahersima a,⇑, Paul Tikalsky b
a
315 Advanced Technology Research Center, Stillwater, OK, USA
b
201 Advanced Technology Research Center, Stillwater, OK, USA

h i g h l i g h t s

Heating performance of mass and thin concrete radiant floors.

Comparing the temperature variation in both radiant floors.

Storing considerable thermal energy in the mass floor by three heating cycles.

A 3D finite element (FE) model for temperature profile in the mass concrete radiant floor.

Good agreement between FE modeling and experimental results.

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

Article history: The thickness of the slabs is imperative when considering the method of heating system due to the ther-
Received 15 February 2018 mal mass and thermal lag properties. Thermal properties of concrete allow thermal energy to be stored in
Received in revised form 12 May 2018 the mass floor after heating cycles. In this study, the experimental heating performance of a 1.22 m-thick
Accepted 22 May 2018
mass concrete radiant floor and a 0.18 m-thick concrete floor is investigated. In addition, a full 3D finite
Available online 26 May 2018
element model of the mass concrete radiant floor is performed and validated by the full-scale building
measurements. The mass floor simulation contains embedded pipes, vertical steel anchors, and horizon-
Keywords:
tal reinforcing steel grids at the top and bottom. The experimental results show that the initial temper-
Mass concrete floor
Radiant heating system
ature of the thick concrete slab is increased by a few cycles and the mass heated concrete floor acts as a
Thermal storage thermal storage battery for the building. Also, the modeling results are in good agreement with construc-
Finite element modeling tion site measurement. This model is further expanded to simulate different thermal loadings on the
pipes to predict the temperature development inside of the mass floor. Furthermore, importance of the
vertical steel anchors in conduction the generated heat to the surface of the floor would be seen in the
modeling results.
Ó 2018 Elsevier Ltd. All rights reserved.

1. Introduction
Concrete slabs with a heating/cooling pipe system have been
extensively used for heating and cooling buildings. Nowadays, to
increase efficiency, concrete slabs with heating and cooling pipe
systems are an essential element of modern buildings. The recent
researches indicated that radiant floor system can provide thermal
comfort and low energy consumption, and quiet operation [1–4].
The heated and (or) chilled fluid in the radiant floor system that
originates from a ground source will substantially increase the
energy efficiency. However, the thin or thick concrete radiant
⇑ Corresponding author at: 21 N University Pl, Apt. 8, Stillwater, OK 74075, USA.
E-mail address: mohammad.tahersima@okstate.edu (M. Tahersima).
floors can perform differently because of the thermal mass and
the thermal lag properties of concrete.
The possibility of storing thermal energy in the mass concrete
floor will have a major impact on stabilizing the temperature in
the building floor. The thermal performance and energy storage
density of a system could be optimized by controlling the
storage-restitution periods, thickness of slab and surface radiative
properties which the most practical choice is slab thickness [5–8].
Concrete has a high volumetric heat capacity and low thermal con-
ductivity properties which allows it to charge and discharge effec-
tively for 24 h period [9]. The specific heat capacity Cp, of the
material is ability of the material to absorb or radiate heat defines
the effective thermal capacity of a material [10]. The material with
higher heat capacity such as concrete can store the generated heat
(hydration heat in mass concrete or heat from radiant heating).

https://doi.org/10.1016/j.conbuildmat.2018.05.177
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371
However, thermal cracks in the concrete elements should be pre-
vented due to huge temperature change [11–13].
Part of the thermal energy is stored by the mass concrete slab
and its’ temperature increases. When the concrete floor is covered
by a roof, the indoor temperature is mostly affected by radiant heat
floor. Using the mass heated floor in the covered structures such as
offices and laboratory with controlled condition of internal air
temperature can moderate the heating need at cool seasons. The
slab temperature could be used as a controlled parameter to
increase energy efficiency or prevent discomfort [14,15].
Due to complicated nature of heat transfer through the concrete
and reinforcement combination, some software engineering and
technology may result in higher accuracy temperature distribution
for the modeling. Therefore, modeling the all elements involved in
heat transfer modeling is necessary. Disregarding the steel role in
the heat transfer of the reinforced radiant floor system would
cause inaccurate results. Therefore, simplifying the model should
be reasonable in the accurate range of results. Some numerical
methods such as finite difference and finite element have been
used for modeling the heat transfer in different materials [16–
18]. However, the model validation needs to be verified by the
physical experiment. Jin, X et al. [19] has studied a finite volume
model for a radiant floor cooling system regarding the effects of
the pipe thermal resistance and water velocity on the performance
of the system. It was noted that the effect of water velocity in the
radiant floor system is not great and the performance of radiant
floor will be affected by low thermal conductivity of the piping sys-
tem. Sattari and Farhanieh [20] studied a finite element model of
radiant floor system and they concluded that thickness of the radi-
ant floor system are the most important parameters in the radiant
floor design.
In this study, the heating performance of thick and thin radiant
concrete floors is presented by temperature measurement of slabs
in a covered laboratory building. Steady state heat transfer analysis
of the heated slab with heating pipe systems is completed by a
widely finite element software [21]. A three-dimension full model-
ing of concrete, steel, and pipes is selected to study the thermal
behavior of mass heated concrete floor as well.
2. Experiment
The experimental part of this research has been conducted in a
new full-scale laboratory building over 3000 m2 area [22–24]. The
concrete slabs with embedded pipes were placed as the foundation
of the building. The slabs include the 1.22 m-thick mass concrete
floor (570 m2 area) with heat tube located 0.15 m under the sur-
face of the slab and the 0.18 m-thick slab (570 m2 area) with
embedded heat tube at the bottom. The mass concrete floor has
four sets of vertical steel anchors with diameter of 0.04 m located
at every 1.22 m in X, Y direction. There is 0.025 m of polystyrene
insulation under the thin slab and all-around of the mass floor
perimeter to increase the thermal storage capacity of the radiant
floor. The mass concrete floor was cast on the 0.075 m mud slab
with no insulation underneath.
Before casting the concrete slabs, T-type thermocouples have
been installed at the positions all through the height of the slabs
to measure the temperature at different elevations. T-
thermocouple wire temperature range is 40 to 80 °C with accu-
racy of ±0.1 °C. All thermocouple wires are connected to a data log-
ger to record the data over time. Data logger was selected with
thermocouple measurement accuracy of ±0.3 °C (25° to 50 °C),
±0.8 °C (55° to 85 °C). The thermocouple measurement was
recorded with ±0.01 °C accuracy in the data logger. The aim of set-
ting thermocouples in various elevations is to measure the thermal
profile along the concrete depth when the radiant floor is applied.
361
The thermocouples are located at 0.05 m, 0.15 m, 0.3 m, 0.52 m,
0.82 m, and 1.22 m below the surface of the mass concrete. One
thermocouple is located under the mud slab to measure soil tem-
perature. Thermocouples are concentrated at the top of the slab to
measure more precisely the heat from the radiant pipe system. It
means when the radiant floor system working, the greatest varia-
tions in temperature are expected at the top of the mass floor
around the pipes. There are fewer thermocouples in 0.18 m slab
to record the temperature profile. Thermocouples are located at
0.05 m below the surface, at middle of the slab, and the bottom
of the thin slab. The thermocouples’ position at the 1.22 m floor
and 0.18 m slab are shown in Figs. 1 and 2, respectively.
The average of two thermocouple readings at each location is
used for collecting the temperature data to verify the accuracy of
the results. In addition, there are two sets of thermocouple column
for each slab including the ‘‘edge point” (0.3 m from the building
façade) and the ‘‘center point” (5m from the building façade).
The temperature in the radiant floor system is important because
source of heating system in the building is the key to control the
indoor air temperature.
In this case study, radiant floor system works with ground
source heat pump (GSHP). The supply temperature and return
temperature of the radiant piping system are measured over time
to control the radiant floor system. An ultrasonic flow meter was
used to measure the flow rate of fluid through the pipes. Pressure
gauges were inserted to the p-t ports of pipes to measure the
return fluid. These measurements over step time of 15 min were
used as input parameters for the modeling of the radiant floor
system.
3. Heat transfer analysis
The in-situ temperature gradient by a piping system can be sim-
ulated with a comprehensive finite element modeling. A three-full
modeling of concrete, steel, and pipes is selected to study the ther-
mal behavior of heated concrete floor. Several researches show
that the 3D finite element analysis provides the more detailed
modeling of the concrete pavements [25]. The fundamental for
thermal analysis in this finite element software [21,26] is a heat
balance equation obtained from the principle of energy
conservation.
@2T @2T @2T @T
k þ k þ k þ q ¼ qc
@x2 @y2 @z2 @t
where c is the specific heat coefficient [J/(kg.°C)], k is the thermal
conductivity coefficient [W/(m.°C)], q is the density [kg/m3], q is
the heat generation rate [J/(m3.s)]. The q term can be neglected
regarding there is no heat generation after long time of casting
concrete.
The generated heat from the piping system distributes to the
surrounded area by concrete, steel anchors, top and bottom hori-
zontal mesh. The heat transfer in the slab happens through the
conduction, convection processes. Then, the indoor air would be
affected by the heat of embedded heat pipes by convection and
radiation. The amount of transferred heat through radiation and
convection from the surface of the concrete depends on a lot of
parameters such as the concrete properties, indoor air tempera-
ture, air movement, building insulation, occupants, floor cover,
and less on outside weather conditions. The most effective param-
eters are needed to simplify the model with accurate results.
Regarding the finite element modeling, a three-dimensional 8-
node thermal solid element (Solid 70) was used to model concrete
thermal properties. A 3D element (Fluid 3-D th-fl pipe 116) used to
model the fluid with the ability to conduct heat and transfer fluid
between its two main nodes. Finally, 3D conduction bar uniaxial
362 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371

Fig. 1. Thermocouples location inside of mass floor (dimension is meter).

Fig. 2. Thermocouples location at 0.18 m slab (dimension is meter).

element (Link 33) was applied to model the vertical steel anchors
and steel mesh with ability to conduct heat between the nodes
at a single degree of freedom of temperature [21]. The element
type shape is presented in Fig. 3 [27].
The steel details and all three elements in FE model are shown
in Figs. 4 and 5, respectively.
The concrete slab dimension is 6.9 m long, 3.3 m wide, and 1.22
m height in proportion of 0.2 of the construction site size. The main
part of heat transfer from the pipes happens through the top and
bottom of slab rather than along the length and width of slab.
The dimension of horizontal mesh at the top and bottom of the slab
is diameter of 0.022 m. Also, the dimension of the vertical steel
Fig. 4. Vertical steel anchors and horizontal steel mesh in FE model.
anchors is diameter of 0.0381 m. At last, the pipe diameter is
selected 0.019 m.

Fig. 3. Element types on left, middle, and right including Steel (Link 33), Concrete (thermal solid 70), Pipe (3D th-fl pipe 116).
 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371
Fig. 5. Finite element model (concrete, steel, and heating pipes elements).
Thermal properties of each element during the heat transfer
analysis would be specific based on the thermal conductivity, heat
capacity and density. Material properties used for the finite ele-
ment model are given in Table 1. Because of low range of thermal
difference in the analysis, it is assumed that the thermal conductiv-
ity and heat capacity will be remained constant in this problem.
On the other hand, the boundary condition on the heated floor
is important to model the heated slab accurately. There is 0.025 m
insulation all around the edges of the slab. The bottom of slab is
rested on 0.075 m of mud slab. The indoor and boundary temper-
ature are specified as the initial condition of the slab based on the
experimental measurement (19 °C). The convection at the surface
of the slab is assumed h = 5.5 W/m2.K [28]. The emissivity of the
surface of the concrete is assumed 0.92 [29]. The mesh conver-
gence is verified by changing the mesh size to make sure that
results are independent of quadrilateral mesh.
Steady state indicates the heat transfer equilibrium of the sys-
tem. The temperature variations in the field were in close agree-
ment with temperature gradient by the finite element
simulation. The result of FE simulation will be discussed.
4. Results and discussion
4.1. Construction site results
The results for temperature profiles in the mass radiant floor
and the thin slab are presented when the heating system was
turned on. As the edge and center of the slab would potentially
have different thermal behavior, the results of site data are pre-
sented separately. The edge point located 0.3 m away from the
building façade is more affected by the ambient temperature. The
center point located 5 m from the building facade would be less
affected by the outdoor air conditions. The results by temperature
variation across the mass floor at different elevations over time is
shown in Figs. 6 and 7 for edge and center points, respectively.
The vertical axis is temperature in degree Celsius and horizontal
Table 1
FEM Material properties.
Concrete Steel Fluid
Density (kg/m3) 2400 7860 1100
Thermal Conductivity (W/m.K) 1.8 80 0.258
Specific Heat (J/kg.K) 900 450 2500
363
axis is time in hours. The supply fluid temperature was set on
80°F (26.7 °C). The orange curve at top of other temperature trends
shows the thermocouple measurement tied to the heat tube.
Zero time is starting time (Jan, 26th, 00:00) of the temperature
measurements to show the initial condition of the mass floor
before enabling the heating system at 250 h. In addition, the initial
temperature of the slab before enabling the heating system is
about 18–19 °C for edge point and 20–20.5 °C for the center point.
The blue line at bottom of slab temperature trends indicates the
ambient temperature which is in range of 3 to 22 °C.
The top of the slab is warmer than the bottom of the slab
because of embedded heating pipes under the surface of the con-
crete. However, the temperature at the bottom of the slab to some
extend is affected by the heating pipes because of heat conducted
by vertical steel anchors. As we can see, the slab temperature
trends will be almost stable after 9.5 days of heating cycle. Based
on this thermal equilibrium of the system, the next cycle of heating
cycles is selected 9.5 days. After turning the heating system OFF at
490 h, the temperature profiles drop. However, the slab tempera-
ture after disabling the heating system would be higher than the
initial slab temperature.
Furthermore the temperature variations in the thin slabs are
shown in Fig. 8 and Fig. 9 for the edge and center points. The mid-
dle of slab and the 0.03 m below the surface of the slab have almost
the same temperature trends.
The temperature variation in the thin slab at the center point is
the same for different elevations of the thermocouples when the
heated thin floor is deactivated. The heat gain and heat loss from
the thin slab would be faster in comparison with the mass floor.
The heat loss of the mass concrete floor is considerably slower
due to massive concrete and lower thermal conductivity of con-
crete in comparison with steel. Radiant floor heating takes advan-
tage of concrete’s thermal mass to absorb and store the heat to
provide the comfort and efficiency for residents [30].
The temperature gradient in the mass floor after three heating
cycles (one 80°F (26.67 °C) and two 90°F (32.22 °C) heating cycles)
is illustrated in Fig. 10, Fig. 11 for edge and center points. For more
clarity, the legends are not shown in Fig. 10. The legend for Fig. 10
is the same as Fig. 6. Each heating cycle and rest time lasts 9.5 days.
The highest temperature inside of slab is around the heating pipe
and the lowest temperature is under the mud slab temperature.
Every heating cycle leads to temperature rise in the concrete slab
which is a proof of storing heat in the mass floor even regarding
the steel role in conducting heat to the indoor air faster. However,
the amount of heat storage is clearly seen in Fig. 11 for the center
364 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371

Fig. 6. Temperature measurement at mass floor edge point (80°F (26.67 °C) thermal loading).

Fig. 7. Temperature measurement at mass floor center point (80°F thermal loading).

Fig. 8. Temperature measurement at thin slab edge point (80°F thermal loading).
 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371 365

Fig. 9. Temperature measurement at thin slab center point (80°F thermal loading).

Fig. 10. Mass floor edge temperature rise (80°F and two 90°F heating cycles).

Fig. 11. Mass floor center temperature rise (80°F and two 90°F heating cycles).
366 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371

point. The temperature variations for top (0.05 m below the sur-
face), middle (0.52 m under the surface), and bottom (on the
mud slab) are compared in Fig. 12 at center and edge point in
the mass floor. In addition, the temperature trends for top (0.03
m below the surface) and bottom (on insulation) are shown in
Fig. 13 at center and edge point in the thin slab. Figs. 12 and 13
clearly illustrates that the center point has the higher temperature
than the edge point for both slabs.
The statistical results of thermocouple measurement are shown
in Figs. 14 and 15. The temperature range and average of thermo-
couple measurement at top, middle, and bottom of the mass floor
are shown in Fig. 14 at edge and center points. Fig. 14 illustrates
bottom part of the mass floor has lower temperature range. pro-
vided below. In addition, the temperature range and average tem-
perature at top and bottom of thin slab are shown in Fig. 15 at
center and edge points. Also, the maximum, average, and mini-
mum ambient temperature of the measurement period are 33.16
°C, 13.84 °C, and 2.58 °C.
The experiments for the heating cycle were carried out at
known initial condition of the slab. After three cycles of heating
for 80°F, 90°F, and 90°F, the overall temperature of the mass floor
is increased for 3 °C. This temperature increase was not dropped
down after passing the considerable time when the heating system
was disabled. It means the thermal energy can be stored in the
mass floor with just few cycle of heating system. The amount of
thermal energy saved inside of the mass concrete floor based on
the heat capacity of the floor and three-degree temperature rise
would be 4 Million BTU (1172KWH) according to energy equation
(Q = mcDT). It helps to heat pumps efficiency to use less energy to
make the heated fluid for the radiant floor.
Temperature gradient in the mass floor along the height of con-
crete slab is shown in Fig. 16 at different elevation of thermocou-
ples positions. All temperature measurement shows the
temperature of the slab at 00:00 time of each day. The first three
lines at the left side highlighted by an elliptic indicate the initial
temperature of the slab before turning the heating system on.
Then, the radiant floor heating is enabled on the fourth day
(02/08). The next yellow line at 02/08 shows the top of half the slab
is heated by the heating pipe. Then, the slab temperature increases
gradually until the heat equilibrium is reached based on thermal
loading on the heating pipe after 9 days. The mass floor tempera-
ture dropped after disabling the heating system on 02/14.
The temperature development at the center of the mass floor in
Fig. 16 shows that the overall initial and final temperature are
greater than the same temperature at the edge point. The top of
the slab is affected by the heating pipe from 20 °C to 25.5 °C. How-
ever, the temperature at the bottom of slab is increased from 20 °C
to 22.7 °C. The initial temperature of the slab was not returned to
20 °C even after the disabling the heating system.
The temperature variation in the center point of thin slab is
shown in Fig. 17 for the same period. The temperature profile is
almost the same for different three spots through the height of

Fig. 12. Comparison of top, middle, and bottom temperature of mass floor at center, edge.

Fig. 13. Comparison of top and bottom temperature of thin floor at center, edge.
 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371 367

the thin slab. The heat gain and loss for the thin slab would be fas-
ter than mass floor due to lower height of the slab. The bottom of
slab is warmer than top because of heating pipes located at the
bottom.
The temperature distributions in the mass floor and thin slab
before enabling the heating system and initial and final each 9 days
of heating cycle heating are shown in Figs. 18 and 19, respectively.
The average initial temperature of the mass floor at the center
point is 19.7 °C. After every heating cycle, the initial temperature
inside of slab for next heating cycle would be greater.
4.2. Finite element model results
The temperature distribution through the concrete slab from
the steady state analysis of finite element model at 26.67°F supply
temperature is shown in Fig. 20. Every color shows a temperature
range which is indicated by the temperature bar under the slab.
The four dots on the surface of the concrete show the tip of vertical
steel. They are directly in contact with indoor air which means heat
at the surface of the slab can be transferred by the concrete and
vertical steel. The 3D model described in the modeling section with

Fig. 14. Temperature range and average values of mass floor Edge (left), Center (right).

Fig. 15. Temperature range and average values of thin floor Edge (left), Center (right).

Fig. 16. Temperature variation at mass floor center point at 00:00 (80°F ON and OFF).
368 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371

0.18

3-Feb (192 h)
0.15 4-Feb (216 h)
5-Feb (240 h)
6-Feb (264 h)
Height of Thin Slab (m)

0.12 7-Feb (288 h)

8-Feb (312 h)
9-Feb (336 h)
0.09
10-Feb (360 h)
11-Feb (384 h)
0.06 12-Feb (408 h)
13-Feb (432 h)
14-Feb (456 h)
0.03
15-Feb (480 h)
16-Feb (504 h)
0 17-Feb (528 h)
19 20 21 22 23 24 25 26 18-Feb (552 h)
Temperature (°C)

Fig. 17. Temperature variation at 0.18 m floor center point at 00:00 (80°F ON and OFF).

1.2

0.95
5-Feb (240 h) - Initial Condition
Height of Slab (m)

14-Feb (456 h) - 80°F cycle

0.7 23-Feb (672 h) - OFF after 80°F
3-Mar (888 h) - ON at first 90°F

0.45 13-Mar (1128 h) - OFF after first 90°F

23-Mar (1368 h) - ON at second 90°F
7-Apr (1728 h) - OFF after second 90°F
0.2

-0.05 15 18 21 24 27 30
Temperature (°C)

Fig. 18. Final Temperature variation at mass floor center point at 00:00 (before and after three heating cycles).

0.18

0.15
Height of Thin Slab (m)

0.12
5-Feb (240 h) - Initial Condition
14-Feb (456 h) - 80°F cycle
0.09 23-Feb (672 h) - OFF after 80°F
3-Mar (888 h) - ON at first 90°F
0.06 13-Mar (1128 h) - OFF after first 90°F cycle
23-Mar (1368 h) - ON at second 90°F cycle
7-Apr (1728 h) - OFF after second 90°F cycle
0.03

0
15 18 21 24 27 30

Temperature (°C)

Fig. 19. Final Temperature variation at 7 in the slab center point at 00:00 (before and after three heating cycles).

different mesh size range has been analyzed to demonstrate the This makes the bottom of the slab may receive the temperature dif-
heat transfer results do not depend on the mesh pattern. ference due to thermal loading of the pipes. The left side of the slab
Due to higher conductivity of steel, the generated heat by piping has the higher temperature because the hot fluid is inserted to the
system is transferred by vertical steel faster than concrete slab. slab at the left side.
 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371 369

Fig. 20. Temperature distribution in the mass floor (FE model).

1.2

0.95

3-Feb, 00:00 (192 h)

Height of Slab (m)

0.7 4-Feb, 00:00 (616 h)

5-Feb, 00:00 (240 h)
Feb-6, 00:00 (264 h)
0.45
14-Feb, 00:00 (456 h)
FEM-Edge

0.2

-0.05 15 17 19 21 23 25
Temperature (°C)

Fig. 21. Temperature variation for mass floor edge (Actual data vs FEM).

1.2

0.95
Height of Slab (m)

3-Feb, 00:00 (192 h)

0.7 4-Feb, 00:00 (216 h)
5-Feb, 00:00 (240 h)
6-Feb, 00:00 (264 h)
0.45 14-Feb, 00:00 (456 h)
FEM-Center

0.2

-0.05 15 17 19 21 23 25
Temperature (°C)

Fig. 22. Temperature variation for mass floor center (Actual data vs FEM).
370 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371

Fig. 23. Top and Bottom view of temperature distribution in the mass floor (FE model).

To validate the model, the full 3-D model, the result of the FE 1.4
steady state analysis is compared with final temperature profiles
1.2
through the depth of the concrete block for edge and center points
shown in Figs. 21 and 22, respectively. The maximum temperature
Height of Slab (m)

1 Center - FEM
along the height of concrete slab is occurred at the heating pipe Edge - FEM
elevation as expected. First three lines indicate the initial condition 0.8
of mass concrete floor before turning the heating system ON. Good
0.6
agreement between the model results and experimental data ver-
ifies the FE model validation. The temperature prediction at the 0.4
center of mass floor has the better match to the results of site
measurement. 0.2

After verification of the model with thermocouple measure- 0

ment at the job site, the model can be extended for another supply 20 25 30 35 40
temperature by the heat pumps. For example, 110°F (43.3 °C) of Temperature (°C)
supply temperature is used to model the mass floor in the heating
mode. All other parameters are kept the same as the previous Fig. 24. Finite element model results for center and edge point (110°F supply
temperature).
model. Temperature distribution in the mass floor is shown in
Fig. 23 for the top and bottom view.
From the temperature distribution in the mass floor, the tem- 5. Conclusion
perature gradient at the center and edge point of slab are extracted
and shown in Fig. 24. There is greater rate of heat loss in higher In summary, the idea of saving thermal energy by mass heated
supply temperature. The temperature at the surface of the slab is concrete floor has been studied. The experiments show that 4
much less than piping temperature because of surface convection MBTU can be stored in the 34.5  16.5  1.22 m3 mass floor after
to the indoor air. three heating cycle of 80°F (26.67 °C), 90°F, and 90°F (32.22 °C)
 M. Tahersima, P. Tikalsky / Construction and Building Materials 178 (2018) 360–371
with 9.5 days long at each cycle. The thermal storage of mass floor
can be applied in the building regarding the less energy usage. It
provides a source of heating at the floor level which provides the
level of comfort and evenly temperature distribution in the build-
ings. However, the thin slab cannot store heat as much as the mass
radiant floor. Due to low cost of concrete construction, it is recom-
mended to increase the thickness of concrete radiant floor founda-
tion for building over large bay area. Also, the thermal behavior of
the mass concrete slab was investigated by a full 3D finite element
model. FE results show that the temperature development of the
slab after enabling the heating system is in good agreement with
experimental data. The rule of vertical steel for conducting the heat
in the modeling is clearly seen with higher temperature distribu-
tion rather than concrete floor at the surface. This modeling can
be used to predict the temperature development inside of the mass
floor at other different thermal loading of radiant floor system.
Conflict of interest
None.
Acknowledgements
The authors gratefully acknowledge the facility management of
Oklahoma State University – United States for great collaboration
to control and running the heating system based on needs of this
research.
References
[1] J.L. Niu, L.Z. Zhang, H.G. Zuo, Energy savings potential of chilled ceiling
combined with desiccant cooling in hot and humid climates, Energy Build. 34
(2002) 487–495.
[2] T. Imanari, T. Omori, K. Bogaki, Thermal comfort and energy consumption of
the radiant ceiling panel system: comparison with the conventional all-air
system, Energy Build. 30 (1999) 167–175.
[3] C. Stetiu, Energy and peak power saving potential of radiant cooling systems in
US commercial buildings, Energy Build. 30 (1999) 127–138.
[4] A. Kyu-Nam Rhee, Kwang Woo Kim, A 50 year review of basic and applied
research in radiant heating and cooling systems for the built environment,
Build. Environ. 91 (2015) 166–190.
[5] E. Bilgen, M.A. Richard, Horizontal concrete slabs as passive solar collectors,
Sol. Energy 72 (5) (2002) 405–413.
[6] A. Engström, L.O. Andersson, Energy Storage in Concrete Slabs Reduce Energy
Consumption and Peak Cooling Loads at no Increase in Capital Costs, Effstock,
Stockholm, 2009.
371
[7] P. Chuangchid, M. Krarti, Foundation heat loss from heated concrete slab-on-
grade floors, Build. Environ. 36 (5) (2001) 637–655.
[8] M. Tahersima, P. Tikalsky, R. Revankar, An experimental study on using a mass
radiant floor with geothermal system as thermal battery of the building, Build.
Environ. 133C (2018) (2018) 8–18.
[9] R. Walsh, P. Kenny, V. Brophy, Thermal mass & sustainable building, Improving
energy performance and occupant comfort, Irish Concr. Federation (2006).
[10] B. Döring, C. Kendrick, R.M. Lawson, Thermal capacity of composite floor slabs,
Energy Build. 67 (2013) 531–539.
[11] M. Mirsayar, X. Shi, D. Zollinger, Evaluation of interfacial bond strength
between Portland cement concrete and asphalt concrete layers using bi-
material SCB test specimen, Eng. Solid Mech. 5 (4) (2017) 293–306.
[12] M.M. Mirsayar, K. Huang, D.G. Zollinger, New approach to determining
concrete slab lift-off by use of interfacial fracture mechanics concepts,
Transp. Res. Record: J. Transp. Res. Board 2590 (2016) 10–17.
[13] S.Z. Mehrian, S.R. Amrei, M. Maniat, S.M. Nowruzpour, Structural health
monitoring using optimising algorithms based on flexibility matrix approach
and combination of natural frequencies and mode shapes, Int. J. Struct. Eng. 7
(4) (2016) 398–411.
[14] M. Shin, K. Rhee, S. Ryu, M. Yeo, K. Kim, Design of radiant floor heating panel in
view of floor surface temperatures, Build. Environ. 92 (2015) 559–577.
[15] S.H. Cho, M. Zaheer-Uddin, An experimental study of multiple parameter
switching control for radiant floor heating systems, Energy 24 (5) (1999) 433–
444.
[16] Md Azree Othuman Mydin, Simplified explicit model to measure transient
heat transfer in foamcrete panel system, Australian J. Basic Appl. Sci. 7 (7)
(2013) 1–13.
[17] S.V. Patankar, D.B. Spalding, A calculation procedure for heat, mass and
momentum transfer in three-dimensional parabolic flows, Int. J. Heat Mass
Transfer 15 (10) (1972) 1787–1806.
[18] Z.P. Bažant, J.C. Chern, W. Thonguthai, Finite element program for moisture
and heat transfer in heated concrete, Nucl. Eng. Des. 68 (1) (1982) 61–70.
[19] X. Jin, X. Zhang, Y. Luo, R. Cao, Numerical simulation of radiant floor cooling
system: the effects of thermal resistance of pipe and water velocity on the
performance, Build. Environ. 45 (2010) 2545–2552.
[20] S. Sattari, B. Farhanieh, A parametric study on radiant floor heating system
performance, Renewable Energy 31 (10) (2006) 1617–1626.
[21] ANSYS academic research, release 15.0 ANSYS academic research, release 15.0
[22] M. Tahersima, T. Ley, P. Tikalsky, Hydration Heat in a mass concrete and a thin
slab with limestone blended cement, Int. J. Mater. Sci. Eng. 5 (2) (2017) 79–86.
[23] M. Tahersima, P. Tikalsky, Finite element modeling of hydration heat in a
concrete slab-on-grade floor with limestone blended cement, Constr. Build.
Mater. 154 (2017) 44–50.
[24] M. Tahersima, Modeling Heat Transfer in Mass Concrete Floors with Radiant
Heat from Ground Source Systems (Doctoral dissertation),, Oklahoma State
University, 2017.
[25] B. Choubane, M. Tia, Analysis and verification of thermal-gradient effects on
concrete pavement, J. Transp. Eng. 121 (1) (1995) 75–81.
[26] L. Dahmani, Thermomechanical response of LNG concrete tank to cryogenic
temperatures, Strength Mater. 43 (5) (2011) 526–531.
[27] ANSYS Thermal Analysis Guide
[28] K.N. Rhee, B.W. Olesen, K.W. Kim, Ten questions about radiant heating and
cooling systems, Build. Environ. 112 (2017) 367–381.
[29] M.N. Ozisik, Heat Transfer: A Basic Approach, McGraw-Hill, London, 1985.
[30] I. Mattsson, G. Fries, Radiant heat with concrete, Concr. Technol. Today 18
(1997) (PL971. 01B).