Proceedings: Building Simulation 2007

DOUBLE SKIN FAADES CAVITY AND EXTERIOR OPENINGS DIMENSIONS FOR SAVING ENERGY ON MEDITERRANEAN CLIMATE Mauricio Torres1,*, Pere Alavedra1, Amado Guzmn1, Eva Cuerva1, Carla Planas2, Raquel Clemente2, Vanessa Escalona2
1

Department of Construction Engineering, Universidad Politcnica de Catalua, Barcelona, Spain. Av Diagonal, 647. 2 Grup JG Enginyers Consultors, Barcelona, Spain.
* Corresponding author e-mail: [email protected]

ABSTRACT Taking into account Mediterranean climate particularities for Barcelona, Spain, a whole year study using TAS simulation software was carried out for a Double Skin Faade corporative office building. It is a typical office building with an extended working hours schedule for acclimatization. Four different cavity widths were simulated 40cm, 60cm, 80cm, 100cm - as well as three different external opening areas for cavitys natural ventilation purpose. Two Double Skin Faade typologies were simulated a corridor faade and a multistory faade. Simulations showed cooling loads are much greater than heating loads on a double glazed skin faade building south oriented in a Mediterranean climate town like Barcelona. Results demonstrated that a Multistory Faade, depending on its configuration, might save up to 5% on annual cooling loads respect to a Corridor Faade. Different opening areas and cavity depths shall be used for different typologies in order to obtaining good energy results.

Double Skin Faade (DSF) typology is being adopted as faade system on many new corporative buildings in Europe with the promise of saving energy while maintaining a transparent faade. DSF offers many configuration possibilities, like using the cavity as a conduit to fresh air intake and to exhausting vitiated air from offices (Gratia and De Herde 2003). Also, it is possible to use a DSF together with the heating, ventilating and air conditioning (HVAC) system, in a symbiosis process (Stec 2006). An energy performance assessment of a single storey multiple-skin faade was carried by Saelens (2003) and a study about total solar energy transmittance of a DSF with free convection was led by Manz (2003). Combined heat transfer in turbulent mixed convection fluid flows in DSF was done by Zllner (2002). Some studies using Thermal Analysis Software (TAS) simulation software about DSF thermal behavior were done (Gratia 2003, Hien et al. 2004). DSF typologies, advantages, disadvantages, uses and many examples were compiled by Oesterle et al. (2001). In spite of all, investigations about DSF thermal behavior and energy performance for the Mediterranean climate are just starting, while DSF buildings are being constructed adopting recommendations from investigations and experiences from other latitudes. Location data and mechanical conditioning data from Barcelona are shown on table 1. Reference temperatures for cooling and heating refer to working hours of a typical office building (8 AM to 20PM).
Table 1 Location and mechanical conditioning data from Barcelona
Latitude 41.3 Longitude 2.1 Altitude 6 Cooling Degree-hours (24C reference) 3160 Heating Degree-hours (21C reference) 485

KEYWORDS
Building simulation, Double Skin Faade, Energy Efficiency, Cooling Loads, Energy saving.

INTRODUCTION
Climate change all over the world and its immediate and future consequences are making up a responsive ecological conscience that claims for energy responsive buildings. This idea is being reflected on construction standards in some countries, like is happening in Spain. The new construction standard Cdigo Tcnico de la Edificacin (CTE 2006)) shed light on the scope of energy saving on buildings and adopts more restrictive rules for construction, limiting heat gains and losses through the skin of the building. With the application of the new Spanish standard for thermal performance of buildings, a decrease of 30 to 40% on energy consumption with heating and cooling, ventilating and lighting, respect to actual consumption is expected.

OBJECTIVES
The main focuses of these simulations are: Distinguishing the influence of cavity depth and height on annual cooling loads (CAL), cooling peak

- 198 -

Proceedings: Building Simulation 2007

loads (CPL), annual heating loads (HAL) and heating peak loads (HPL) for the DSF building. Verifying the influence of the exterior opening (EOp) dimensions on annual energy demands and peak loads for the DSF building. Verifying the significance of combining different external opening areas with different cavity depths and heights for saving energy.

orientations and internal heat transmissions were minimized focusing results on south faade configuration. The only double glazed faade simulated was south faade. The perspective of the building can be seen on figure 1.

SIMULATION
Simulation by TAS software The TAS software works solving the dynamic fluid equations at the studied building zones using hourly conditions according to local climate data. The equations together describe mass transfer and thermal transmission on each of its process conduction, convection and radiation. The software also takes into account internal loads like equipments (computers, printers, etc), artificial lighting and peoples load for the building. It also considers infiltration through openings, ventilation and air exchange between zones. The simulation software is fed with climatic data and predefined internal loads, plus the construction materials thermal characteristics. Energy demand to supply the mechanical conditioning of the building is obtained from a balance between internal and external loads, construction materials thermal and optical characteristics plus the resulting heat exchange between the building and the external environment. All these parameters and the required level of comfort maintenance will determine the necessary energy to keep the system working properly. The study is focused on the thermal performance of the building by means of its energy demands comparison for heating and cooling annual and peak loads.

Fig. 1 . Perspective of the corridor faade office building

The orientation of the building matches exactly with the 4 main natural solar orientations north, south, east and west, as seen on figure 2.

Fig. 2. Building organization and solar orientation for the complete model(left) and for the simplified model (right)

The corridor faade and its components can be seen on figure 3.

The building The model is an isolated office building, counting on a ground commercial floor plus 6 office floors. There are 8 offices per floor and the building is totally air conditioned based on an extended working hours schedule (weekday from 8 AM to 8 PM). DSF cavity is naturally ventilated 24 h during the whole year through a corridor faade, partitioned on each floor. The air flows diagonally, entering by bottom opening and exiting by top opening on the exterior skin. It prevents vitiated air from lower floor entering the floor above. This typology is called Corridor Faade (CF). The model was simplified to obtaining non dependent results from others solar orientations. In this way, solar gains from other solar

Fig.3. Corridor faade detail

Other solar orientations but south faade are treated as opaque surfaces, icluding ground floor walls. The entire building is air conditioned except faade cavities. Figures 4 shows the CF buidling.

- 199 -

Proceedings: Building Simulation 2007

Roof

Simulation Parameters Input data is the necessary information to feed the simulation software in order to perform the thermal analysis. It is divided in two groups - the variable data and the fixed data. The variable data is DSF cavity and exterior opening dimensions that are being analyzed and compared, as can be seen on table 3.
Table 3 Variable Input data Corridor Multistory faade faade 0.8%=1.54m2 1.6%=3m2 5%=1.6m2 2.4%=4.6m2 2 10%=3.2m 1.6%(7.5m wide)= 3m2 15%=4.8m2 3.2%(7.5m wide)= 6m2 4.8%(7.5m wide)= 9.2m2 3.5m 23.5m

Top opening External Glazing

Bottom opening Slabs External Wall

Ground Floor

Exterior openings dimensions Vertical distance between external openings DSF cavity depth

Fig. 4. Buildings corridor faade south offices

The interior skin of the building is a typical Curtain Wall typology 60% double glazed while exterior skin is tottaly single glazed. Table 2 shows the studied zone dimensions.
Table 2 Studied zone dimensions Office area per floor (south office) Office volume per floor (south office) Cavity area per floor (south cavity) Cavity volume per floor (south cavity) Vertical distance between external openings Transparent external faade area per floor 64m2 236.8m3 6.4m2 24.96m3 3.5 23.5m 32 m2

40cm, 60cm, 80cm, 100cm

Fixed data is the information repeated on every simulation and is not subjected to analysis. These non variable parameters are seen on table 4.
Table 4 Fixed Input Data Parameter Artificial lighting thermal load Equipments thermal load Occupancy sensible gain Occupancy latent gain Working schedule Barcelonas climate database Air leakage Floor Height Ground solar reflectance Buildings Height Mean height of surroundings Exterior glazing width Exterior glazing solar factor Double Glazing interior pane width Double Glazing cavity width Double Glazing cavity filling Double Glazing exterior pane width Exterior glazing U-value Internal floors U-value Ground floor U-value Roof U-value Exterior Walls U-value Cavity partition elements U-Value Windows frame conductivity Value 15 W/m2 10 W/m2 6.5W/m2 4.2W/m2 8am to 8 pm Meteonorm 0.05 0.1ACH 4m 0.2 30 m 30 m 6mm 0.3 6mm 10mm air 6mm 5.73 W/m2C 0.9W/m2C 0.27W/m2C 0.25W/m2C 0.5W/m2C 6W/m2C 0.14W/m2C

A multistory faade (MSF) is obtained by removing horizontal partitions on each floor of the building and by closing all EOp, except the bottom opening at the first floor and the top opening at the sixth floor. The cavity of this typology becomes totally integrated and the air enters the cavity by the bottom opening and exits by the top opening. The simplifications on the models will not provide the exact energy demands for cooling and heating the offices, as they isolate south offices from the influence of others solar orientations and internal asymmetric temperatures for different zones. The results accomplish with the purpose of a qualitative comparison of the different models. Figure 5 shows the MSF simplified model.
Roof Top opening Slabs External Glazing External Wall

Output data will be the CAL, CPL, HAL, HPL, maximum cavity temperature and air flow rate in the cavity. CAL and HAL represent the maintenance energy required to accomplish with users minimum thermal comfort. It means how expensive thermal comfort will be for such building typology. CPL and HPL represent the necessary power for the HVAC plant to accomplishing with maximum HVAC demands for

Bottom opening Ground Floor

- 200 -

Proceedings: Building Simulation 2007

Annual Cooling Loads (KWh/m2)

thermal comfort maintenance. It means how much money will be invested on HVAC equipment acquisition. Simulations were run hourly for the whole year, with 15 days of precognition air-conditioning of the building.

54,50 54,00 53,50 53,00 52,50 52,00 51,50 51,00 40cm 60cm Cavity depth External opening area: 5% 10% 15% 80cm 100cm

d) Simulations sequence External openings area simulation At first, the 1 floor height CF was simulated varying its EOp area, according to table 3. Cavity depth simulation At this moment, simulations were carried out varying faades cavity width among 40, 60, 80 and 100cm. Each of these typologies was tested with the different EOp areas listed on table 3. Cavity height simulation The MSF was then simulated, varying its EOp areas and cavity depths according to table 3. Then, the area of the EOp was doubled for the multistory faade, as shown on figure 3. The width of these openings passed from 3.75m to 7.5m, as the possible problem of warm air reentering the faade through openings does not exist for this case.

Fig 6 Annual Cooling Loads for 1 floor Corridor Faades

Wider cavities with bigger EOp also presented lower CPL, as showed on figure 7. Again, increasing openings area represents a more important influence on decreasing CPL than increasing cavity depth.

50,50

Cooling peak loads (W/m2)

50,00

49,50

49,00

48,50

48,00 40cm 60cm Cavity depth External opening area: 5% 10% 15% 80cm 100cm

Fig 7 Cooling Peak Loads for 1 floor Corridor Faades

HAL for 1 floor height CF are shown on figure 8.

RESULTS
Annual Heating Loads (KWh/m2)

0,52

0,50

Simulations results refer to annual loads and peak loads for the six south facing offices, totalizing 384m2 of analyzed floor area. Cavity temperatures refer to the 6th floor cavity. Air flow volume (Air changes per Hour, ACH) represents the ventilation rate of the cavity 1 floor height in the CF case and 6 floors height in the case of the MSF. The air flow rates also vary according to the cavity widths simulated.

0,48

0,46

0,44

0,42

0,40 40cm 60cm 80cm 100cm

Cavity depth External opening area:

Fig 8 Annual Heating Loads for1 floor Corridor Faades

1 FLOOR CORRIDOR FAADE (CF) CAL results for 1 floor height CF demonstrated that wider cavities perform better than narrower cavities, as can be seen on figure 6. The same figure shows the bigger the opening areas the lower energy demands for CAL. However, it is possible to sense a bigger difference on these loads when changing from 5% to 10% the EOp areas, rather than when changing from 10 to 15%. Increasing EOp has a more important influence on decreasing CAL than increasing cavities depth. By increasing opening areas from 5% to 15% for a 40cm cavity depth, a 2.06% less energy demands is obtained. For a 100cm cavity depth, it lowers demands by 2.48%.

HPL depend more on EOp than on cavity depth, just as HAL, as shown on figure 9. For both of them, the thinner the cavity and the smaller the EOp areas, the less energy for heating is demanded.

- 201 -

Proceedings: Building Simulation 2007

17,80 17,60 Heating Peak Loads (W/m2) 17,40 17,20 17,00 16,80 16,60 16,40 40cm 60cm Cavity depth External opening area:
5% 10% 15%

MULTISTORY FAADE (MSF) MSF with 3.5m wide external openings perform on a different way than 1 floor height CF when concerning CAL. As there is a sharper temperature gradient occurring along the MSF cavity due to its height, the heat tends to accumulate on the higher levels of this space before exiting by the external top opening. Greenhouse effect on these faade will be more accentuated and will help warm air to moving upward inside the cavity, increasing its ventilation rate. Variation on CAL for the 5% EOp model is almost the same for all depths tested. As this EOp percentage can not ventilate the cavity efficiently, most of the upward airflow depends on the stack effect. Larger openings perform better to diminish CAL when used together with narrower cavities, as shown on figure 12. This is because of the increment of the ventilation rate that helps the stack effect on extracting air from the cavity.
56,00 Annual Cooling Loads (KWh/m2) 55,50 55,00 54,50 54,00 53,50 53,00 52,50 52,00 51,50 51,00 40cm 60cm Cavity depth External opening area: 0,8% 1,6% 2,4% 80cm 100cm

80cm

100cm

Fig 9 Heating Peak Loads for 1 floor Corridor Faades

Maintaining the EOp area fixed while increasing cavity depth from 40cm to 60, 80 and 100cm produce an average decreasing on ACH of 32%, 49% and 59%, respectively, as can be sensed on figure 10. On the same figure, it is possible to perceive that increasing exterior opening areas from 5% to 10% and from 5% to 15% produce an average lift on ACH of 83% and 147%, respectively, for the same cavity depth.

100,00 90,00

Air Changes per Hour (ACH)

80,00 70,00 60,00 50,00 40,00 30,00 20,00 10,00 0,00 40cm 60cm 80cm 100cm

Fig 12 Annual Cooling Loads for Multistory Faades

Fig 10 Air changes per hour at the 6th floor of the 1 floor Corridor faades

Maximum cavity temperature at 6th floor cavity demonstrated to be more self dependent on EOp areas variance than on cavity depth changes, as can be seen on figure 11. Increasing EOp areas from 5 to 15% can lead to decrements on these temperatures by 5.6%, while increasing cavity depth from 40 to 100cm, and maintaining the same opening area, may lift 6th floor cavity temperature by 1.1%.
45,50 45,00 44,50 Celsius degree (C) 44,00 43,50 43,00 42,50 42,00 41,50 41,00 40,50 40cm 60cm Cavity depth External opening area: 5% 10% 15% 80cm 100cm

CPL for this typology presents a very similar behavior, as seen on figure 13. The 40cm cavity depth, together with 15% EOp presents the lowest demands for CAL and for CPL. This occurs because the warm air volume is more effectively extracted from the cavity. Furthermore, this faade typology presents the highest maximum temperature rates as can be seen on figure 16, therefore the stack effect helps extracting air from the cavity.

51,00 50,80 Cooling peak loads (W/m2) 50,60 50,40 50,20 50,00 49,80 49,60 49,40 49,20 49,00 48,80 40cm 60cm Cavity depth External opening area: 0,8% 1,6% 2,4% 80cm 100cm

Fig 13 Cooling Peak Loads for Multistory Faades

Fig 11 Maximum temperature for 1 floor Corridor faades

- 202 -

Proceedings: Building Simulation 2007

Air Changes per Hour (ACH)

Maximum cavity temperatures coincide with lower cavity ACH, which is the case of the 5% EOp together with the narrower cavity depth that heats up easily, thus causing lower demands for HAL. Indeed, these loads diminish predominantly because of the variance of the EOp, as seen on figure 14.
0,60 Annual heating loads (KWh/m2) 0,50 0,40 0,30

In this way, as can be seen on figure 17, typologies that depends predominantly on the stack effect to extracting air present the lower rates of ACH.
250,00

200,00

150,00

100,00

50,00

0,00

0,20 0,10 0,00 40cm 60cm Cavity depth External opening area: 0,8% 1,6% 2,4% 80cm 100cm

40cm

60cm

80cm

100cm

Cavity depth External opening area:

Fig 17 Airflow rates at 6th floor cavity for the multistory faades

Fig 14 Annual Heating Loads for Multistory Faades

In this case, when upward airflow is slow and depending predominantly on the stack effect, the cavity works as a thermal buffer, insulating the building and lowering heat losses. It also contributes to lowering HPL, as shown on figure 15.
17,80 17,60 Peak heating loads (W/m2) 17,40 17,20 17,00 16,80 16,60 16,40 16,20 16,00 40cm 60cm Cavity depth External opening area: 0,8% 1,6% 2,4% 80cm 100cm

Furthermore, on wider cavities with small EOp, the airflow inside the cavity tends to adopt a recirculation behavior, ascending and descending along the cavity height before exiting, thus overheating the cavity space.

MULTISTORY FAADE WITH 7.5 METERS WIDE TOP AND BOTTOM OPENINGS This DSF typology doubles the EOp area of the 6 floors height MSF previously analyzed by extending its bottom and top exterior openings. In this way, the openings occupy almost the total width of the analyzed faade. As EOp areas are doubled, an increment on the ACH of the cavity is expected for all typologies, as well as a decrement on maximum cavity temperatures. As a consequence of that, lower CAL and CPL are expected as well as higher HAL and HPL.
56,00 Annual cooling loads (KWh/m2) 55,00 54,00 53,00 52,00 51,00 50,00 49,00 40cm 60cm Cavity depth External opening area: 1,6%W 3,2%W 4,8%W 80cm 100cm

Fig 15 Heating peak Loads for Multistory Faades

Maximum resultant temperatures on the 6th floor cavity depend more on EOp area than on cavity depth, as shown on figure 16. Thinner cavities with bigger openings present a higher air flow rate than ticker cavities with smaller openings, thus lowering the maximum temperatures.
51,00 50,00 49,00

Temperature (C)

48,00 47,00 46,00 45,00 44,00 43,00 42,00 41,00 40cm 60cm 80cm 100cm

Fig 18 Annual Cooling Loads for Multistory wide openings faade typology

Cavity depth External opening area:

Fig 16 Maximum temperature at 6th floor cavity for multistory faades

The chart on figure 18 confirms expectations about CAL, which has lowered in comparison to the 6 floors MSF with 3.5m wide. The same has happened to the CPL, demonstrated on figure 19. The larger openings help warm air extraction from the cavity as higher air flow rates start occurring.

Wider cavities used together with small EOp depend mainly on stack effect to extracting air from cavity.

- 203 -

Proceedings: Building Simulation 2007

51,00 50,50 50,00 49,50 49,00 48,50 48,00 40cm 60cm Cavity depth External opening area: 1,6%W 3,2%W 4,8%W 80cm 100cm
Temperature (C)

52,00 50,00 48,00 46,00 44,00 42,00 40,00 38,00 40cm 60cm Cavity depth External opening area: 1,6%W 3,2%W 4,8%W 80cm 100cm

Cooling peak loads (W/m2)

Fig 19 Cooling Peak Loads for Multistory wide openings faade typology

Fig 22 Maximum temperature at 6th floor cavity for the multistory faades

Air Change per Hour(ACH)

HAL for this typology are a little bit higher than for the 3.5m EOp wide MSF, as can be seen on figure 20. As the higher air flow helps the stack effect on extracting warm air from the cavity, the effectiveness of the cavity space as a thermal buffer is prejudiced and increasing heat losses. The effect of air recirculation inside the cavity also diminishes, as wind driven force is more effective on extracting air from the cavity.
0,60 Annual heating loads (KWh/m2) 0,50 0,40

Thinner cavities with large openings present a larger ventilation rate, as seen on figure 23. Counting on larger openings to ventilate smaller spaces, plus the greenhouse effect heating up smaller spaces, a higher ACH is achieved.
300,00 250,00 200,00 150,00 100,00 50,00 0,00

0,30 0,20 0,10 0,00 40cm 60cm Cavity depth External opening area: 1,6%W 3,2%W 4,8%W 80cm 100cm

40cm

60cm Cavity depth

80cm

100cm

External opening area:

1,6%W

3,2%W

4,8%W

Fig 23 Airflow rates at 6th floor cavity for the multistory faades

Fig 20 Annual Heating Loads for Multistory wide openings faade typology

Minimum loads and its correspondent DSF typology can be seen on table 5.
Table 5 Minimum loads for analyzed typologies for the south offices
DSF Typology Dimensions 40cm - 2,4% 40cm - 2,4% 40, 60, 80, 100cm 0.8% 80cm 0.8% Minimum Value 51.69 49.09 0.41 16.57 Cooling Loads (KWh/m2) Cooling Peaks (W/m2) Heating Loads (KWh/m2) Heating Peaks (W/m2) Multistory 6 floors WEO Multisoty 6 floors WEO Multistory 6 floors Multistory 6 floors

As a result of increases on heat exchanges between faade and exterior environment, HPL also increases, as warm air is easily extracted, as shown on figure 21.
18,50 18,00 17,50 17,00 16,50 16,00 15,50 40cm 60cm Cavity depth External opening area: 1,6%W 3,2%W 4,8%W 80cm 100cm

Heating peak loads (W/m2)

CONCLUSIONS
Simulations showed cooling loads are much greater than heating loads for south oriented DSF building, under Barcelona climate conditions. Thus, preventing overheating of the cavity through adequate ventilation is necessary for energy saving. Optimum cavity ventilation might be achieved by equilibrating its dimensions height, width and depth as well as its elements dimensions, such as the EOp. All these factors are linked and demonstrated on the simulations on this study. Building a DSF involves high costs, including maintenance costs of two additional surfaces. It also requires integrated elements and implies on losing

Fig 21 Heating Peak Loads for Multistory wide openings faade typology

As expected, maximum resultant temperatures on these faades have also diminished (figure 22), because of the higher rates of the ACH in the cavity, as seen on figure 23, due to the larger EOp. These relations are very similar between all cavity depths and opening areas.

- 204 -

Proceedings: Building Simulation 2007

office occupancy area for cavity space purpose. Indeed, costs might increase significantly not only by the building of the second skin and its components, but also by the acquisition of a more powerful HVAC plant and the higher loads for thermal comfort maintenance that it probably implies. Simulations demonstrated that a narrower cavity with no horizontal partitions may demands less energy on CAL and CPL than a wider cavity partitioned on each floor. The lower energy demand for CAL for the 1 floor CF was 52.25 kWh/m2. The 6 floors MSF with wide EOp demands 1.07% less energy for the same purpose. Apparently, this saving is little significant, but it implies on others advantages. It is due to the greater ventilation rate of the cavity and to the accentuated stack effect occurring inside the cavity. The air heats up rapidly in this space due to the great distance between exterior bottom and top openings, causing a higher temperature gradient along the height of the cavity. The larger exterior opening areas proportionate better ventilation for narrower spaces. These advantages would be the less space dedicated to the cavity space, which could be used as office area; the less material used on the cavity, as well as the lower weight of the additional structure. A better sound insulation would be another advantage due to the less exterior openings, as it can be significantly helpful on noisy areas of a town. For using a 40cm wide multistory faade, the cavity space should be accessed by the offices through interior openings, for maintenance reasons. If maintenance service is to be made by walking over rails inside the cavity space, then its depth should be wider than 40cm. Simulations demonstrated the predominant influence of the exterior opening areas on diminishing cooling loads on all models. Results for cooling and heating are quite similar because of the little sensitivity of the model. The opaque surface of the faades interior skin, as well as the double glazing are very well insulated, diminishing heat exchanges between offices and cavity space. If a poorer insulated opaque wall were used instead, higher fluctuations on cooling and heating annual and peak loads would be obtained. DSF offers a lot of construction possibilities such the adding of controlled shading devices inside cavity and controlled external and internal openings for night ventilation of the cavity or even natural ventilation of the offices. This is a simulation study for some particular simplified DSF models under particular climate conditions. The results of this study can not be adopted as reference for others climates. This study is not sufficient for technically designing a DSF, as

there are many other DSF components and parameters to be taken into account.

ACKNOWLEDGEMENT
The authors wish to thank the Ctedra UPC Grup JG for its financial support to the project.

REFERENCES
Cdigo Tcnico de la Edificacin 2006. Seccin HE 1 Limitacin de la demanda energtica. Espaa. Gratia E, Herde D, 2003. Optimal operation of a south double-skin faade. Energy and Buildings 36 pp. 41-60. Hien W, Liping, Chandra A, Pandey A, Xiaolin W. 2004. Effects of double glazed faade on energy consumption, thermal comfort and condensation for a typical office building in Singapore. Energy and Buildings 37 563-572. Manz H. 2003. Total solar energy transmittance of glass double faades with free convection. Energy and Buildings 36 pp. 127-136. Oesterle, Lieb, Lutz, Heusler, Double-Skin Faades, Integrated Planning. 2001. Munich. Prestel Verlag. Saelens D. 2002. Energy performance assessment of single storey multiple-skin facades Ph.D. Thesis, Katholieke Universiteit Leuven. (Belgium), 272 pages. Stec W. 2006. Symbiosis of Double Skin Faade and indoor climate installation PhD Thesis. TU Delft (The Netherlands), 251 pages. Zllner A, Winter E, Viskanta R. 2002. Experimental studies of combined heat transfer in turbulent mixed convection fluid flows in double-skin-faades. International Journal of Heat and Mass Transfer 45 pp 4401-4408.

- 205 -