Optimal rammed earth wall thickness for a single-family house in Serbia

S. Stevanovi

University of Ni, Faculty of Sciences and Mathematics, Viegradska 33, 18000 Ni, Serbia

ABSTRACT: Sustainability and energy efficiency have become important factors in modern architecture,
with additional momentum generated by the revised European Building Performance Directive 2010/31/EU,
which set the goal that all buildings in European Union constructed after 2020 have to be nearly zero energy
buildings. Sustainability of rammed earth walls is unquestionable, however their large U-value implies that
they have to be combined with external insulation in climates with cold winters, as is the case in Serbia. Our
goal here is to study, through building energy performance simulations in Energyplus, to what extent a combination of rammed earth walls and external insulation may accomplish the goal for a single-family house in
Serbian city of Belgrade to become a nearly zero energy building, provided the house is equipped with properly shaded, high performance windows, and highly insulated floor slab and roof.

1 INTRODUCTION
Increased demand for sustainability in architecture
renewed interest in traditional building methods
which use locally available resources. Building with
earth is one such method, present in different forms
throughout the world since ancient times. Earthen
architecture in Serbia mostly used adobe bricks, followed by wattle and daub technique, while rammed
earth technique, widespread in arid and semi-arid
climates, does not appear to have been present in
Serbian vernacular architecture. However, with the
establishment of brick factories in Serbia, and especially after the World War II, the practice of earthen
building gave way to building with baked bricks and
concrete, due to their better structural characteristics
and ability to build taller structures.
An important characteristic that was neglected at
the time is earth's much higher ability to control humidity by reducing the amplitude of relative humidity fluctuations inside the building and the frequency
of high humidity periods at the wall surface (Allinson & Hall 2010), thus managing to maintain relative humidity levels within the range of 40-60%,
deemed as most comfortable for humans. Combined
with earth's high thermal inertia, which enables it to
store and reradiate heat, this often results in more
satisfactory indoor thermal comfort.
Energy efficiency is another important factor in
modern architecture, emphasized by the influence

that the consumption of energy from nonrenewable

sources within buildings has on climate change, and
put forward by current legislation, especially in regions with cold winters. For example, the most recent Serbian regulations on energy efficiency in
buildings (Government of the Republic of Serbia
2011) prescribe that external walls in new buildings
may have the maximal U-value of 0.40 Wm -2K-1.
Having in mind that rammed earth has a low resistance to heat transfer, with a dry conductivity in the
range of 0.8-1.0 Wm-1K-1 (Hall 2008), meeting this
limit with rammed earth only would result in unacceptably thick walls. Hence, rammed earth walls in
Serbia should be coupled with thermal insulation in
order to attain higher thermal resistance. Fix (2009)
recommends the use of external insulation in combination with rammed earth in cold climates, as it
completely protects rammed earth from freezing
during winter, and thus from spalling damage, while
it leaves unhindered contact with the indoor air, facilitating the use of rammed earth's thermal mass
and humidity control.
Even further, the revised European Building Performance Directive 2010/31/EU (European Parliament 2010) sets the goal for all buildings in
European Union constructed after 2020 to be nearly
zero energy buildings. This means that the building
will have very high energy performance, determined
on the basis of actual or predicted energy consumption for its typical use, which includes energy for

space heating and cooling and sanitary water heating, while the nearly zero or small amount of energy
needed should be, in largest part, produced from renewable energy sources at the building site or in its
vicinity.
Our goal here is to study to what extent a combination of rammed earth walls and external insulation
may be used to accomplish the goal of a nearly zero
energy building, through simulations of building energy performance in Energyplus, for a case study of
a single-family house in Serbian city of Belgrade,
provided the house is equipped with properly
shaded, high performance windows and highly insulated floor slab and roof.
The plan of the paper is as follows. In Section 2
we briefly analyze the climate of Belgrade, while in
Section 3 we present the structure and characteristics
of the case study house. The simulation results for
the different combinations of thicknesses of external
insulation and rammed earth are discussed in Section 4, while establishment of net zero energy goal is
discussed in Section 5.
2 CLIMATE DATA ANALYSIS
Representative Belgrade weather data are analyzed
from the aspect of temperatures, wind and solar radiation.
Figure 1 shows the distribution of average and
absolute minimal and maximal monthly temperatures in Belgrade. The highest average maximal
monthly temperatures are 27.3C in July and August,
while the lowest average minimal monthly temperature is -2.3C in January. The distribution of monthly
temperatures reflects the high demand of heating for
buildings in Belgrade, and the low demand of cooling for buildings with no large amounts of internal 3
heat gains.

Monthly dry bulb temperature (C)

50
40
30
20
10
0
-10
-20

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Abs. Max.
Avg. Max.
Avg. Min.
Abs. Min.

-30
Figure 1. Distribution of monthly temperatures in Belgrade
average and absolute minimal and maximal temperatures.

There are two dominant wind directions in Belgrade: north-west and south-east. The north-west
wind from Atlantic mostly blows in late spring and
early summer and has smaller speeds, while the
south-east wind, so-called koava, is turbulent in
nature and blows at its fullest in February and
March, when the gusts may reach 18 m/s, indicating
the need for wind protection from the south-east.
However, about 93% of wind in Belgrade ranges in
speed from 1 m/s to 5 m/s.
Figure 2 shows the distribution of average
monthly global solar radiation on horizontal surface
in Belgrade (Gburik 2008). Annual total solar radiation on horizontal surface in Belgrade is 1396
kWh/m2 with approximately 2025 hours of sunshine.
The amount of solar radiation is generally high
throughout March-September and is indicative of the
possibility to use solar energy both for sanitary water
heating and electricity production.
200
180
160
140
120
100
80
60
40
20
0
Figure 2. Average monthly global solar radiation on horizontal
surface in kWh/m2 in Belgrade.

CASE STUDY HOUSE

The single-story house under considerations is designed for a four-person family. It has gross area of
83 m2, featuring an open plan for the living, dining
and kitchen area in the southern part of the house,
and two bedrooms and a bathroom in the northern
part. An option exists to further expand the common
areas with an office/studio in the loft above the bedrooms. The house floor plan is shown in Figure 3.
We have already seen that the rammed earth walls
have to be externally insulated in order to meet regulatory demand for U-value of at most 0.40 Wm -2K-1,
while still providing thermal mass and indoor humidity control. The expanded polystyrene (EPS) is
chosen as insulating material for walls in this research, due to its small price and prevalence in Serbia. If sustainability would be the main priority, then
the natural fiber or cellulose insulation have to be
considered instead of EPS, however we have to note
that these materials are not presently available in
Serbian market. The external walls consist of 20 mm

rendering on the outside, EPS ranging in thickness 4 ENERGY PERFORMANCE SIMULATIONS

from 15 cm to 25 cm, stabilized rammed earth
(SRE) ranging in thickness from 30 cm to 50 cm,
The house was modeled in Energyplus as a single
and 13 mm dense plaster on the inside. In order to
zone with an infiltration rate of 0.7 ach. Stabilized
prevent capillary rise of water from soil, the stabilrammed earth walls are supposed to contain 40%
ized rammed earth walls are erected on a concrete
sands. Twelve different external walls types have
foundation with a 25 cm base exposed above ground
been simulated. Their structure, exclusive of 20 mm
level, and a bituminous painted layer on top of the
external rendering and 13 mm internal dense plaster,
base.
is given in Table 1.
The internal walls consist of 30 cm thick stabilTable 1. Structure of simulated rammed earth walls.
ized rammed earth, covered with13 mm dense
Wall
EPS thickness SRE thickness U-value
plaster on both sides.
-2 -1
1
2
3
4
5
6
7
8
9
10
11
12

Figure 3. The house floor plan.

Since our goal is to study the effect of thicknesses

of external insulation and stabilized rammed earth
walls on heating and cooling energy needs, the floor
slab and roofs are highly insulated. The floor slab
consists of 20 cm extruded polystyrene on the
ground and below foundations, 10 cm cast concrete,
5 cm floor screed and ceramic tiles, yielding U-value
of 0.158 Wm-2K-1. The roof consists of 25 mm clay
tile, 25 cm stone wool and 5 mm roofing felt,
covered on inside with gypsum plasterboard, yielding similar U-value of 0.153 Wm-2K-1.
From the same reason, we have chosen for simulation the high performance windows with triple,
low-emissivity glazing. Such windows have been
shown to be best performers in Serbian climate
(Stevanovi et al. 2011). The roof has sufficiently
large eaves overhangs, in line with Serbian vernacular architectural practice, in order to provide shading
to windows during summer.
Finally, the extension of the eastern external wall
towards south is designed to provide protection from
the turbulent south-eastern winds in late winter.

cm
0
0
0
10
10
10
15
15
15
20
20
20

cm
30
40
50
30
40
50
30
40
50
30
40
50

Wm K
2.07
1.76
1.54
0.34
0.33
0.32
0.24
0.23
0.23
0.18
0.18
0.18

The house was first simulated in a free floating

mode, without any HVAC system installed. The average monthly indoor temperatures, together with
average monthly outside temperature, are illustrated
in Figure 4. As the figure shows, there is a clear distinction between uninsulated walls (types 1-3) and
insulated walls (types 4-12), with the uninsulated
houses being 2C to 5C colder than the insulated
houses, both during winter and summer. Interestingly, the range of average temperatures for insulated walls (types 4-12) is narrower than 2C, except
during May and October, when wall types 4 and 5
present outliers.
The house was then simulated with HVAC system
installed, which consisted of underfloor heating system using the natural gas boiler and electric chiller.
As the heating source is at the floor, the air temperature at the lower level is warm and cool at the higher
level (warm feet, cool head), which provides the
same thermal comfort level at lower temperature
than the conventional heating systems. Hence, the
heating setpoint in this case was 18C, while the
cooling setpoint was set at still comfortable 28C.
For all wall types, the cooling energy needs from
May until September are extremely low, ranging annualy from 64 kWh to 72 kWh of electricity, which
is less than monthly lighting electricity use. This can
be contributed to rammed earth's thermal mass work,
as the thickness of external insulation (or its absence) does not appear to have influence on the cooling energy needs. Thus, this fully confirms numerous subjective claims that the indoor temperatures in
earthen houses are very comfortable during the summertime.

35

30
Ambient temperature
no EPS, 30cm SRE
no EPS, 40cm SRE
no EPS, 50cm SRE
10cm EPS, 30cm SRE
10cm EPS, 40cm SRE
10cm EPS, 50cm SRE
15cm EPS, 30cm SRE
15cm EPS, 40cm SRE
15cm EPS, 50cm SRE
20cm EPS, 30cm SRE
20cm EPS, 40cm SRE
20cm EPS, 50cm SRE

Average monthly temperature (C)

25

20

15

10

0
Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Figure 4. Average monthly indoor temperatures in free floating simulations for different wall types, with ambient temperature
shown at the bottom.

2500

Natural gas (kWh)

2000

no EPS, 30cm SRE

no EPS, 40cm SRE
no EPS, 50cm SRE
10cm EPS, 30cm SRE
10cm EPS, 40cm SRE
10cm EPS, 50cm SRE
15cm EPS, 30cm SRE
15cm EPS, 40cm SRE
15cm EPS, 50cm SRE
20cm EPS, 30cm SRE
20cm EPS, 40cm SRE
20cm EPS, 50cm SRE

1500

1000

500

0
Figure 5. Monthly
heating energy needs (kWh) for different wall types.
Table 2. Total heating energy needs (kWh) for different wall types.
Wall type
1
2
3
4
5
Natural gas (kWh)
8340
7596
7004
4361
3412

Total heating energy needs for different wall

types are given in Table 2, while the monthly heating
energy needs from October until April (until May for

6
3366

7
3120

8
3083

9
3047

10
2931

11
2901

12
2870

uninsulated wall types) are illustrated in Figure 5.

The role of external insulation in Serbian climate is
obvious from this figure, as the presence of at least

10 cm of EPS reduces the heating needs by a factor

of two. Still, the thickness of only 10 cm of EPS is
apparently below the critical value, as the increase of
SRE thickness from 30 cm to 40 cm further decreases heating energy needs for 21.8%. On the other hand, the 15 cm of EPS appears to be above the
critical value, since the increase in SRE thickness
then has a marginal effect on heating energy needs
the increase in SRE thickness from 30 cm to 50 cm
decreases the heating energy needs by at most 2.3%.
The increase in EPS thickness from 15 cm to 20 cm
yields further 5.8-6.1% decrease in heating energy
needs and is worthwhile to be considered, taking
into account relative inexpensiveness of EPS and future price escalation of natural gas. Hence, the optimal external wall for the studied house appears to
be type 10, consisting of 20 cm expanded polystyrene external insulation and 30 cm of stabilized
rammed earth.
5 NEARLY ZERO ENERGY GOAL
For the wall type 10, the total heating and cooling
energy needs amount to 3003 kWh, while the simulated annual consumption of electricity for interior
lighting and appliances is 2117 kWh. These amounts
do not include sanitary water heating, for which solar water heating system may be used. It was shown
in Stevanovi & Pucar (2011) that roof-mounted solar collectors with 4.8m2 of aperture area in Belgrade
yield a solar fraction of 61% for the daily hot water
usage of 220 liters. The remaining 39% of hot water
needs, when covered by backup electric resistance
storage water heater, uses approximately 1300 kWh
of electricity. Hence, the case study house total energy needs amount to 6420 kWh.
Solar energy is the most abundant renewable energy source in Belgrade, and Stevanovi & Pucar
(2012) estimated that the roof-mounted photovoltaic
plant, installed at tilt of 35, has annual specific production of 1246 kWh/kWp in Belgrade. Thus, the
house total energy needs may be fully offset by installing 5.15 kWp photovoltaic plant (which takes up
approximately 36 m2 of roof area), in which case the
house would become a net zero energy building.

REFERENCES
Allinson, D. & Hall, M. 2010. Hygrothermal analysis of a stabilised rammed earth test building in the UK. Energy and
Buildings 42: 845-852.
Bui, Q.B, Morel, J.C, Venkatarama, R. and Ghayad, W. 2009.
Durability of rammed earth walls exposed for 20 years to
natural weathering, Buildings and Environment 44: 912919.
European Parliament. 2010. 2010/31/EU Directive. Brussels:
Official European Union Journal.
Ferreira, J. & Pinheiro, M. 2011. In search of better energy performance in the Portuguese buildingsThe case of Portuguese regulation. Energy Policy 39: 7666-7683.
Fix, S. 2009. Viability of Rammed Earth Building Construction
in Cold Climates. Toronto: Ryerson University.
Gburik, V. 2008. Final report of the technological development Project TD-7042B Atlas of solar and wind energy
potential in Serbia. Belgrade: University of Belgrade, Institute of Multidisciplinary Research.
Government of the Republic of Serbia. 2011. Regulations on
energy efficiency in buildings. Belgrade: Official Gazette.
Hall, M.A. 2008. Assessing the moisture-content-dependent
parameters of stabilised earth materials using the cyclic-response admittance method. Energy and Buildings 40: 20442051.
Minke, G. 2006. Building with EarthDesign and Technology
of a Sustainable Architecture. Basel: Birkhuser.
Pacheco-Torgal, F. & Jalali, S. 2012. Earth construction: Lessons from the past for future eco-efficient construction.
Construction and Building Materials 29: 512-519.
Sheweka, S. 2011. Using Mud Bricks as a Temporary Solution
for Gaza Reconstruction. Energy Procedia 6: 236-240.
Soebarto, V. 2009. Analysis of indoor performance of houses
using rammed earth walls. In: Building Simulation 2009;
Proc. Eleventh International IBPSA Conference, Glasgow,
Scotland, July 27-30, 2009. International Building Performance Simulation Association.
Stevanovi, S., Pavlii, M. and Marinkovi, A. 2011. Determining optimal combination of passive solar design parameters in Novi Sad. In: PhIDAC 2011; Proc. of III International Symposium for students of doctoral studies in the fields
of Civil Engineering, Architecture and Environmental Protection, Novi Sad, Serbia, September 21-23, 2011. Novi
Sad: University of Novi Sad, Faculty of Technical Sciences.
Stevanovi, S. & Pucar, M. 2011. Financial measures Serbia
should offer for solar hot water systems. In: ECOS 2011;
Proc. of the 24th International Conference on Efficiency,
Cost, Optimization, Simulation and Environmental Impact
of Energy Systems, Novi Sad, Serbia, July 4-7, 2011. Novi
Sad.
Stevanovi, S. & Pucar, M. 2012. Investment appraisal of a
small, grid-connected photovoltaic plant under the Serbian
feed-in tariff framework. Renewable & Sustainable Energy
Reviews 16: 1673-1682.