F                 Daytime Ventilation (Comfort Ventilation) is the most common passive cooling system. The system uses outdoor air during daytime to remove the heat gains in the room air in a way shown in Figure 1(a). The system increases the occupant’s thermal comfort by increasing convective and evaporative heat transfer between the occupants and the room air. The maximum indoor air velocity is approximately 2 m/s (Givoni 1998). If the outside temperature is high, the indoor air temperature will then be too high to be accepted by the occupants. This system works better in climates with a mild summer.

Research Approach

The performance study of the two passive cooling systems requires accurate modeling of the building thermal response to outdoor climate conditions and internal heat gains. The thermal response, in conjunction with ventilation and thermal comfort models, can determine whether there is a need for additional mechanical cooling systems. An effective passive cooling system requires little or no mechanical cooling.

Kammerud (1984) used a building thermal analysis program (BLAST) to study the effect on the cooling load of a night cooling strategy in a typical house for several locations in the U.S. He cited two main problems encountered in his work: difficulty in modeling natural ventilation in the building and incorrect estimation of the convective heat transfer coefficients (these coefficients depend on the air velocity). The present investigation uses a model that addresses these two problems. A computational fluid dynamics (CFD) program, PHOENICS (CHAM 1998) is used to calculate natural ventilation in the apartment. PHOENICS calculates detailed airflow in and around the building. An experimental correlation for naturally ventilated buildings (Chandra, 1984) is used to calculate the convective heat transfer coefficients as a function of the air velocity near the walls. The detailed CFD results provide air velocity distributions near the walls, allowing for the accurate determination of the convective heat transfer coefficients. These coefficients are used as boundary conditions in the building thermal analysis. In this way, the approach used couples the thermal analysis with CFD.

Figure 2 shows the coupled program structure. The starting point is the weather data (on the left). The data is required by both the ventilation calculation (upper box) and the thermal analysis (lower box). The CFD program predicts the air velocities near the walls in order to determine the convective heat transfer coefficients for the thermal analysis. Finally, the results of the airflow simulation and thermal analysis are part of the inputs required by the thermal comfort model (ISO 1993).

The thermal response of the building walls and floors was calculated with an explicit finite difference method (Mills 1995) using a one-dimensional heat transfer approximation. The radiative heat transfer is calculated using the radiosity method (Mills 1995). The program calculates the short wave radiation and infrared radiation separately. Since each airflow simulation can take approximately 18 hours on a 450 MHz Pentium II PC, the study used a fixed airflow pattern for each outside wind direction and velocity, independent of the thermal conditions. This approximation is acceptable because the work done by the thermal buoyancy forces  is smaller than the momentum in cross ventilated single story building air flows. The approximation can be validated by the Archimedes number, Ar, for the flow in the apartment:

where   g = the gravity acceleration

L = the typical length scale, such as room height

            DT = the typical average temperature difference between the surface and the air

V = the typical velocity of the external flow

Ar is a ratio of the buoyancy forces (the numerator) over the inertial forces (the denominator). In the present study, Ar varies from 0.05 to 0.5 when the wind changes from 1 to 3 m/s with DT = 5 K and L = 2.7 m. Therefore, the effect of buoyancy is considered negligible.

CASES STUDIED

Figure 3 shows the south-north oriented building and apartment studied. It is a six-story apartment building with ten units (two units per floor) in which the ground floor is used for storage spaces. This building is isolated and located in an open suburban area.The right part of Figure 3 shows the internal layout of the apartment units. Each unit has three bedrooms and a total floor area of 115 m². The figure in dark gray presents the external walls and floors, in medium gray the doors, and in light gray the internal partitions. Access to the apartments is through an external corridor on the north side of the units (upper part in the drawing). The occupants enter each apartment through the door that is visible on the upper part of the living room. On the south side of the living room, there is an external balcony (in the figure the balcony starts where the living room internal wall turns dark gray).

            Several features in the design make this building suitable for cross ventilation. The separation of the living room from the three bedrooms allows for high ventilation rates in the living room during the night when the occupants are in the bedrooms. This arrangement greatly enhances nighttime heat release from the living room surfaces. Another feature to increase cross-ventilation is the large apertures above the internal doors. These apertures allow for air to flow through the bedrooms even when the doors are closed. External shading is used in all the south windows to block the summer sun. The balcony of the living room upstairs provides shading for the living room window in the south. When the apartment is in maximum ventilation mode, both the balcony window and the north windows in the living room are fully open. During the day, ceiling fans are used in the apartment to enhance convective heat transfer between the thermal mass and the occupants.

The windows are double-glazing with aluminum frame without thermal break (DSA, 2.4mm glass). The flooring material is ceramic tile. The living room windows are partially (50%) covered with light venetian blinds with a shading coefficient of 0.33. Table 1 defines further the cases studied. The internal gains, as shown in Table 2, were estimated according to ASHRAE recommendations (ASHRAE 1997).

            In order to better evaluate the performance of the two passive cooling systems, a reference case is used. The reference case, as shown in Table 1, has a reasonable amount of thermal mass. This agrees well with most current buildings in Beijing and Shanghai. Since most of the existing buildings and current designs do not use passive cooling strategies, this study estimates that in conventional buildings (represented by the reference case) the flow rate due to natural ventilation is 30% of the one in the normal case. In the reference case, no ceiling fan is used. The new design optimizes natural ventilation, referred as daytime ventilation ( see Figure 3). The results are also compared with the thermal comfort obtainable with ambient outside air temperature and humidity(referred as outside in Figure 4). The daytime ventilation case only has 20% of open window area during the night (as opposed to 100% for the night cooling case). During daytime, cross ventilation is used whenever the indoor air temperature is higher than the outside air temperature. The daytime ventilation case studied here uses a lighter structure than the other cases.

           In a passive cooling system, the outside air temperature and humidity are crucial to the system performance. Table 3 shows a few important climatic parameters for Beijing and Shanghai in the warmer months of the year (from May to September). The first column shows the number of days with maximum outside air temperature above 30°C. These are the days when, for a properly designed building, cooling may be needed. The second column shows the average temperature variation between day and night in these days (this parameter is important for the night cooling system). A large temperature variation will make night cooling more effective, because the low temperature air flow during the night will remove more heat from the building thermal mass. The last two columns present the average wind speed, during the day (important for the daytime ventilation system) and during the night (important for night cooling). Compared to Beijing, Shanghai has a warmer climate with higher humidity and lower temperature variation. It is more challenging to successfully apply passive cooling systems in Shanghai.

The present analysis uses two criteria to evaluate the passive cooling systems:

F The number of hours of thermal discomfort during the warm season (from May 1st to September 30th), calculated using Fanger’s thermal comfort model (ISO 1993).

F The maximum indoor air temperature for each day in the warm season.

A supplementary criterion, used only for the warmest days of the season (maximum T_out>30°C), is the average difference between the maximum indoor and outdoor air temperatures

The following sections discuss the performance of the two passive cooling systems in Beijing and Shanghai and compare them with the two reference cases.

The Performance of the Passive Cooling Systems in Beijing

 

            Figure 4 shows the percentage of discomfort hours in the warm season. The left figure is for whole day and the right figure from 7 am to 12 pm. The results show that about 95% of the discomfort hours occur in the period between 7 am and 12 pm. The dark gray bars are the results for the two passive ventilation strategies. The results show that night cooling is very effective for Beijing. There are only 330 discomfort hours (9% of the total number of hours in the warm season). Compared to the reference case, there is a 57% reduction in the number of discomfort hours (a reduction of 437 hours of thermal discomfort). Daytime ventilation improves the thermal comfort slightly. The number of discomfort hours when using  daytime ventilation is close to that with ambient air temperature. This is a consequence of the fact that as air moves through the apartment during the day it not only removes the heat gains (positive process) but it also heats the walls of the apartment (negative process). The case labeled
NC+DV is a combination of the other two ventilation strategies. It uses the night cooling control strategy during the night and the daytime ventilation control strategy during the day. For this case the results are much better than the reference case but still not as good as the night cooling case.

Figures 5 and 6 further show the maximum indoor and outdoor air temperatures in the warm season with the two cooling strategies. The air temperature in the living room with night cooling is noticeably lower than the outside air temperature. The maximum air temperature in the living room in the season is 31.6^oC. On average, the indoor air temperature is 3.9 K cooler than outdoor. With daytime ventilation, the windows are open whenever the air temperature in the living room is higher than outside air temperature. Therefore, the maximum indoor temperature is very similar to outside. It can even be higher than outside. The calculation shows that the maximum indoor air temperature can be higher than 34°C in eight days. The results show that night cooling is superior to daytime ventilation in Beijing.

The Performance of the Passive Cooling Systems in Shanghai

Shanghai is warm and humid and has a very small daily-temperature variation as shown in Table 3. The climate conditions indicate the challenges to apply the two passive cooling systems there. Nevertheless, Figure 7 shows that the two passive cooling systems can help to achieve better comfort level (compared with the two reference cases). Night cooling is the best among the four cases but the improvement is not substantial when compared with daytime ventilation or the reference case. This is due to the small daily-temperature variation, although the maximum outdoor air temperature is almost the same as that in Beijing. The small temperature variation outdoors does not allow for significant cooling of the building thermal mass.

Even with night cooling, the maximum indoor air temperature can be as high as 34.9^oC. On average, night cooling can only lower the room air temperature by 0.9 K, as shown in Figure 8. There are 29 days in the season with a day maximum temperature above 32^oC and 32 days with a day maximum temperature between 30 and 32^oC. The indoor air is too hot to be accepted by the occupants.

 

The results show that using the two passive cooling systems alone in the warm season in Shanghai is not enough to provide comfortable indoor thermal conditions.

Discussion

Table 3 shows that the Shanghai climate is hot and humid. In this type of climate, the risk of condensation is high. This risk can be severe when using night cooling, because the high thermal mass surface temperatures are low, increasing the condensation risk. Although the highest relative humidity occurs in the night hours, the lowest dewpoint air temperature occurs in the daytime. During the day, air that infiltrates from the outside (this study uses 1.5 ACH/hour) contacts with the cool surfaces of the thermal mass elements and condensation can occur. This is because, during the day, the temperature difference between inside and outside can be as high as 6 K.

Our calculation shows that in Beijing condensation can occur in the surfaces of the high thermal mass elements in 60 hours during the warm season (when using night cooling). With daytime ventilation, the condensation risk is much lower. This is because night cooling leads to much lower surface temperatures. Surprisingly, condensation is not very significant (it might occur in only 8 hours) in Shanghai, although the relative humidity of the outside air is much higher than that in Beijing. The reason is that the air temperature in the night is high in Shanghai. The high temperature does not cool the building thermal mass very much. Therefore, the surface temperature is higher than the outside air dewpoint temperature.

CONCLUSIONS

This paper evaluates the performance of two passive cooling systems: daytime ventilation and night cooling, for a generic, six-story apartment building in Beijing and Shanghai. The investigation introduced a new detailed simulation method to simulate cross natural ventilation and to analyze thermal response of the apartment building.

The results show that night cooling is superior to daytime ventilation in both Beijing and Shanghai. Night cooling may replace air conditioning systems for a major part of the cooling season in Beijing (90% of the time). On average, the maximum indoor air temperature is 3.9 K cooler than the maximum outdoor air temperature in Beijing with night cooling. Daytime ventilation helps to provide a better comfort level indoors, but the improvement is not very significant. The two passive cooling systems cannot work effectively in Shanghai since comfort conditions are only assured in 66% of the warm period. This is due to the small daily-temperature variation and high air temperature and humidity in Shanghai.

Condensation can be a problem in Beijing when using night cooling. The study found that condensation may occur for 60 hours in the warm period. Condensation is negligible in Shanghai, although it is a hot and humid climate.

Acknowledgments

The authors would like to thank the V. Kahn-Rasmussen Foundation (through the Alliance for Global Sustainability) and the Fundaão para a Ciência e Tecnologia for supporting the research.

REFERENCES

ASHRAE 1997 ASHRAE Handbook Fundamentals SI Edition. ASHRAE Atlanta.

CHAM 1998. PHOENICS Version 3.1. CHAM Ltd., London, UK.

Chandra, S. and Kerestecioglu, A.A. 1984. “Heat transfer in naturally ventilated rooms data from full-scale measurements,” ASHRAE Transactions, 211-224.

Etheridge D.W. and Sandberg M. 1996. Building Ventilation, Theory and Measurement. Wiley & Sons, UK.

Givoni B. 1998. Climate Considerations in Building and Urban Design. Van Nostrand Reinhold.

Givoni B. 1994. Passive and Low Energy Cooling of Buildings. Van Nostrand Reinhold.

ISO 1993. Moderate thermal environments – determination of the PMV and PPD indices and specifications for thermal comfort. International Standard 7730.

Kammerud R., Ceballos E., Curtis B., Place W., and Anderson B. 1984. “Ventilation cooling of residential buildings,” ASHRAE Transactions, 95(2), 226-251.

Mills A.F. 1995. Basic Heat and Mass Transfer. Irwin, USA.

Patankar S.V. 1980. Numerical Heat Transfer and Fluid Flow. Hemisphere.

Shaw, C. T. 1992. Using Computational Fluid Dynamics, Prentice Hall, UK.

Siegel, R. and Howell, J.R. 1981. Thermal Radiation Heat Transfer, 2nd ed.. McGraw-Hill, New York.

back to top