Application of displacement ventilation system to buildings in the tropics

... simplifies the computation of the supply air flow rate that is needed in the design of the displacement ventilation system for buildings with ceiling heights of about 2.6m in the tropics vii List of. .. 2.1 Introduction Since displacement ventilation was first applied to the welding industry in 1978, it has been gaining popularity in Scandinavia as a means of ventilation to provide good indoor... influence on the quality of the inhaled air Conversely, the rising stream had a negative influence on the quality of the inhaled air when the contaminants were generated in the lower part of the room

ACKNOWLEDGEMENT I would like to express my deepest gratitude to A/P David Cheong for his support, guidance and valuable advice throughout this academic exercise. I would like to thank Dr. Risto Kosonen for his support, valuable advice, expertise and knowledge in this project. I would like to express my appreciation to Mr. Yu Weijiang for the assistance, advice and friendship. My appreciation and thanks to the following people who have been very helpful and for their great contributions in the accomplishment of the study: z Mr. Tan Chow Beng, Mr. Zuraimi bin Mohd. Sultan, and Mr. Siew Hock Meng for their assistance in the preparation of the experiments. z Ms Leow Hwee Ching, Ms Lai Yinghui, Mr Lee Chong Leong and Ms Ng Yi Yu for their assistance in the data process and analysis. z All the lecturers who have shared their invaluable knowledge and experience in the Building Science program. Finally, thanks to my friends whom I have been working with throughout the period of study in the National University of Singapore. TABLE OF CONTENTS SUMMARY................................................................................................................... vi List of Tables ...............................................................................................................viii List of Figures ................................................................................................................ ix CHAPTER 1 INTRODUCTION .............................................................................. 1 1.1 Background ................................................................................................ 1 1.2 Objectives and scope of research............................................................... 3 1.3 Outline of dissertation................................................................................ 4 CHAPTER 2 LITERATURE REVIEW ................................................................... 5 2.1 Introduction................................................................................................ 5 2.2 Gradients in occupied space....................................................................... 6 2.2.1 Temperature gradient .......................................................................... 6 2.2.2 Concentration gradient........................................................................ 7 2.2.3 Humidity gradient ............................................................................... 9 2.3 Thermal comfort ...................................................................................... 10 2.3.1 Standards on thermal comfort........................................................... 11 2.3.3.1 Indices (ISO 7730, 1994) .................................................... 11 2.3.3.1.1 Predicted mean vote (PMV)......................................... 11 2.3.3.1.2 Predicted percentage of dissatisfied (PPD).................. 11 2.3.3.1.3 Draft rating (DR).......................................................... 12 2.3.3.2 Criteria in various standards ............................................... 12 2.3.2 Findings of previous research ........................................................... 13 2.3.2.1 Thermal comfort studies on displacement ventilation system ............................................................................................. 13 2.3.2.2 Tropical area thermal comfort studies ................................ 14 2.4 Indoor air quality...................................................................................... 16 2.4.1 Concentration distribution ......................................................... 16 2.4.2 Age of air ................................................................................... 17 2.4.3 Ventilation effectiveness............................................................ 18 2.5 Energy ...................................................................................................... 19 2.6 Conclusions and hypotheses .................................................................... 21 2.6.1 Conclusions....................................................................................... 21 2.6.2 Hypotheses........................................................................................ 22 CHAPTER 3 PRELIMINARY STUDY ................................................................. 23 3.1 Methodology ............................................................................................ 23 3.1.1 Research design ................................................................................ 23 3.1.1.1 Group 1 ............................................................................... 24 3.1.1.2 Group 2 ............................................................................... 25 3.1.1.3 Group 3 ............................................................................... 26 3.1.2 Methods of data collection................................................................ 27 3.1.2.1 Subjective assessment......................................................... 27 3.1.2.1.1 Subjects ........................................................................ 27 3.1.2.1.2 Subjective assessment protocol.................................... 28 3.1.2.1.3 Questionnaire ............................................................... 29 3.1.2.2 Objective measurement protocol ........................................ 29 3.1.2.2.1 Objective parameters ................................................... 29 3.1.2.2.2 Instrumentation ............................................................ 30 3.1.2.2.2.1 Thermal chamber .................................................. 30 3.1.2.2.2.2 Instruments............................................................ 32 ii 3.1.2.2.3 Measuring locations ..................................................... 34 3.1.2.2.4 Measurement procedure............................................... 35 3.1.3 Data collection and processing ......................................................... 36 3.1.3.1 Objective data ..................................................................... 36 3.1.3.2 Subjective data .................................................................... 37 3.1.4 Method of data analysis .................................................................... 37 3.2 Results and discussion ............................................................................. 39 3.2.1 Gradients ........................................................................................... 39 3.2.1.1 Temperature gradient .......................................................... 39 3.2.1.1.1 General description ...................................................... 39 3.2.1.1.2 Effect of different supply air temperature and flow rate with the same room air temperature (Group 1)................................. 40 3.2.1.1.3 Effect of different supply air temperature and flow rate with different room air temperature (Group 2) ................................. 41 3.2.1.2 Humidity gradient ............................................................... 42 3.2.1.2.1 Effect of different supply air temperature and flow rate with the same room air temperature (Group 1)................................. 42 3.2.1.2.2 Effect of different supply air temperature and flow rate with different room air temperature (Group 2) ................................. 43 3.2.1.3 Concentration gradient of carbon dioxide........................... 44 3.2.1.3.1 Effect of flow rate with the same room air temperature (Group 1) ...................................................................................... 44 3.2.1.3.2 Effect of flow rate with different room air temperature (Group 2) ...................................................................................... 45 3.2.2 Thermal comfort ............................................................................... 45 3.2.2.1 Effect of different supply air temperature & relative humidity on thermal comfort (Group 1) .................................................. 45 3.2.2.1.1 Overall Actual Mean Vote (AMV) and comfort acceptability ...................................................................................... 45 3.2.2.1.2 Effect of different supply air temperature on thermal comfort ...................................................................................... 47 3.2.2.1.3 Effect of distance from supply unit on thermal comfort.. ...................................................................................... 48 3.2.2.1.4 Effect of vertical temperature difference on local thermal sensation .............................................................................. 48 3.2.2.2 Effect of different room air temperature on thermal comfort (Group 2) ............................................................................................. 49 3.2.2.2.1 Overall AMV and comfort acceptability ..................... 49 3.2.2.2.2 Effect of vertical temperature gradient on local thermal comfort ...................................................................................... 51 3.2.2.2.3 Effect of distance from supply unit on thermal comfort.. ...................................................................................... 51 3.2.2.3 Comparison of DV with MV (Group 3) ............................. 52 3.2.2.3.1 AMV and comfort acceptability .................................. 52 3.2.2.3.2 Neutral temperature ..................................................... 54 3.2.2.3.3 Local thermal comfort.................................................. 55 3.2.2.4 Application of ISO 7730 ..................................................... 55 CHAPTER 4 CONFIRMATION STUDY.............................................................. 57 4.1. Methodology ............................................................................................ 57 4.1.1 Research design ................................................................................ 57 iii 4.1.1.1 Group 1 ............................................................................... 57 4.1.1.2 Group 2 ............................................................................... 58 4.1.1.3 Group 3 ............................................................................... 59 4.1.2 Methods of data collection................................................................ 59 4.1.2.1 Subjective assessment......................................................... 59 4.1.2.1.1 Subjects ........................................................................ 59 4.1.2.1.2 Subjective assessment protocol.................................... 60 4.1.2.1.3 Questionnaire ............................................................... 60 4.1.2.2 Objective measurement protocol ........................................ 60 4.1.2.2.1 Objective parameters ................................................... 60 4.1.2.2.2 Instrumentation ............................................................ 61 4.1.2.2.2.1 Thermal chamber .................................................. 61 4.1.2.2.2.2 Instruments............................................................ 62 4.1.2.2.3 Measuring locations ..................................................... 62 4.1.3 Data collection and analysis.............................................................. 64 4.1.3.1 Objective data ..................................................................... 64 4.1.3.2 Subjective data .................................................................... 64 4.1.4 Method of data analysis .................................................................... 64 4.2 Results and discussion ............................................................................. 65 4.2.1 Gradients ........................................................................................... 65 4.2.1.1 Temperature gradient .......................................................... 65 4.2.1.1.1 Effect of different supply air temperature and flow rate with the same room air temperature and humidity (Group 1) .......... 65 4.2.1.1.2 Comparison between DV and MV (Group 3).............. 66 4.2.1.2 Humidity gradient ............................................................... 67 4.2.1.2.1 Effect of different supply humidity ratio with the same room air temperature (Group 2) ........................................................ 67 4.2.1.2.2 Comparison between DV and MV (Group 3).............. 68 4.2.1.3 Concentration gradient........................................................ 70 4.2.1.3.1 Effect of different supply air temperature and flow rate with the same room air temperature and humidity (Group 1) .......... 70 4.2.1.3.2 Comparison between DV and MV (Group 3).............. 71 4.2.2 Thermal comfort ............................................................................... 72 4.2.2.1 Effect of different supply air temperature & flow rate on thermal comfort (Group 1)....................................................................... 72 4.2.2.1.1 Overall Actual Mean Vote (AMV) and comfort acceptability ...................................................................................... 72 4.2.2.1.2 Local thermal comfort.................................................. 74 4.2.2.1.3 Draft ............................................................................. 75 4.2.2.1.4 Effect of distance from supply unit on thermal comfort.. ...................................................................................... 77 4.2.2.2 Effect of different humidity levels on thermal comfort (Group 2) ............................................................................................. 78 4.2.2.2.1 Overall Actual Mean Vote (AMV) and comfort acceptability ...................................................................................... 78 4.2.2.2.2 Local thermal comfort.................................................. 79 4.2.2.3 Comparison of DV with MV (Group 3) ............................. 80 4.2.2.3.1 Overall AMV and comfort acceptability ..................... 80 4.2.2.3.2 Local thermal comfort.................................................. 81 4.2.2.3.3 Draft risk ...................................................................... 82 iv 4.2.3 Energy ............................................................................................... 85 4.2.3.1 Comparison between DV and MV...................................... 85 4.2.3.1.1 Experiment conditions, system and conditioning process ...................................................................................... 85 4.2.3.1.2 Energy consumption analysis ...................................... 86 4.2.3.2 Effect of different supply air temperature on energy consumption ............................................................................................. 88 4.2.3.2.1 Experimental conditions, system and conditioning process ...................................................................................... 88 4.2.3.2.2 Energy consumption analysis ...................................... 88 4.2.4 Preliminary design guide .................................................................. 89 4.2.4.1 Model .................................................................................. 89 4.2.4.2 Application of the model .................................................... 91 4.2.4.3 Design procedure: ............................................................... 92 CHAPTER 5 CONCLUSIONS............................................................................... 93 5.1 Review and achievement of research objective ....................................... 93 5.2 Recommendation ..................................................................................... 98 BIBLIOGRAPHY....................................................................................................... 101 v SUMMARY Conventional mixing ventilation (MV) system is used widely in the hot and humid country like Singapore. The nature of this system is to create a good mixture of the room and supply air in order to provide a uniform thermal and air quality environment. Hence every occupant in the same space will be exposed to similar level of pollutants even though one is far away from the source of pollutant. However, if occupants are exposed to harmful and/or excessively high concentration pollutants, they may fall ill and this will lead to a decrease in productivity. Displacement ventilation (DV) system can resolve this problem in a more energyefficient manner. However, research on displacement ventilation has been mainly conducted in the Scandinavian countries. To date there is limited research done in the tropics. Furthermore, as the climate and building loads in this area are different from those in the Scandinavian countries, the results of such research may not be applicable in Singapore. It is therefore of great importance to conduct research to assess its viability in the Tropics. The results and findings would be valuable to local practitioners when adopting the DV system in Singapore. The results and findings would be valuable to researchers who are interested to carry out this area of study in the tropics. It is found that temperature gradient, humidity gradient and CO2 concentration gradient exist in all experiments of this study. The profile of the gradient depends on the supply air flow rate and/or outdoor air flow rate. Results show that supply air temperature, room air temperature and relative humidity have significant influence on subjects’ thermal sensation, but not on subjects’ acceptability ratings. vi Generally subjects have cooler thermal sensation and lower acceptability with DV system than with MV system. Occupants in an environment served by DV system have 1 ºC higher neutral temperature as compared to the MV system. With proper design, DV system may have lower draft risk as compared to the MV system. CO2 concentration gradient can be found in space served by DV system. The concentration of CO2 at the height of 1.1m and below is always lower than in a space served by the MV system, under the same condition. Hence, the ventilation efficiency is always higher for DV system than for MV system. When conditions for both systems are the same, the cooling capacity for DV system is 5% lower than that for MV system. In addition, with the presence of temperature and concentration gradients, the energy consumption of the DV system could be further reduced. A model, which computes the supply-return air temperature difference is developed based on the experimental results. This model simplifies the computation of the supply air flow rate that is needed in the design of the displacement ventilation system for buildings with ceiling heights of about 2.6m in the tropics. vii List of Tables Table 2.1 The criteria stated in different standards................................................ 12 Table 3.1 Test conditions for Group 1. .................................................................. 25 Table 3.2 Test conditions for Group 2. .................................................................. 26 Table 3.3 Test conditions for Group 3. .................................................................. 27 Table 3.4 Instrumentations..................................................................................... 32 Table 3.5 Average clo value, overall AMV and average comfort acceptability.... 46 (Group 1 DV)......................................................................................................... 46 Table 3.6 Mean thermal sensation and comfort acceptability at the workstation closest to the supply unit (Group 1 DV) ......................................................... 47 Table 3.7 Mean thermal sensation and comfort acceptability between workstations nearest and furthest from the supply unit (Group 1 DV) ................................ 48 Table 3.8 Average clo value, overall AMV and average comfort acceptability (Group 2 DV).................................................................................................. 50 Table 3.9 Mean thermal sensation and comfort acceptability between workstations nearest and furthest from the supply unit (Group 2 DV) ................................ 51 Table 3.10 Average clo value and AMV (Group 3) .............................................. 53 Table 4.1 Test conditions for Group 1. .................................................................. 58 Table 4.2 Test conditions for Group 2. .................................................................. 58 Table 4.3 Test conditions for Group 3. .................................................................. 59 Table 4.4 Average clo value, overall AMV and average comfort acceptability (Group 1)......................................................................................................... 73 Table 4.5 Mean thermal sensation and comfort acceptability between workstations closer and further from the supply unit (Group 1 DV) ................................... 77 Table 4.6 Average clo value, overall AMV and average comfort acceptability (Group 2)......................................................................................................... 78 Table 4.7 Average clo value, overall AMV and average comfort acceptability (Group 3)......................................................................................................... 81 Table 4.8 Experiment conditions for Group 3. ...................................................... 85 Table 4.9 Experiment conditions. .......................................................................... 88 viii List of Figures Figure 2.1 Sketch of displacement ventilation......................................................... 6 Figure 2.2 Temperature gradients in a thermal chamber with different cooling loads. ................................................................................................................. 7 Figure 2.3 The CO2 concentration gradients at various locations in a room. .......... 8 Figure 2.4 Relative humidity gradient and moisture ratio gradient with displacement ventilation. .................................................................................. 9 Figure 3.1 Total approach employed in this research study. ................................. 24 Figure 3.2 Continuous scale used in the questionnaire.......................................... 29 Figure 3.3 Chamber layout. ................................................................................... 31 Figure 3.4 Floor-standing, low velocity, semi-circular supply unit....................... 31 Figure 3.5 Return grille.......................................................................................... 32 Figure 3.6 Supply diffuser. .................................................................................... 32 Figure 3.7 Portable RH sensor. .............................................................................. 33 Figure 3.8 Dew-point hygrometer.......................................................................... 33 Figure 3.9 Hood. .................................................................................................... 33 Figure 3.10 Photoacoustic spectrometer multi-gas analyser.................................. 33 Figure 3.12 Plan view of measuring points. .......................................................... 35 Figure 3.13 Typical temperature gradient (1). ....................................................... 40 Figure 3.14 Typical temperature gradient (2). ....................................................... 40 Figure 3.15 Temperature gradients (Group 1). ...................................................... 41 Figure 3.16 Temperature gradients (Group 2). ...................................................... 41 Figure 3.17 Humidity gradients (Group 1). ........................................................... 42 Figure 3.18 Humidity gradients (Group 2). ........................................................... 44 Figure 3.19 Carbon dioxide gradients (Group 1)................................................... 44 Figure 3.20 Carbon dioxide gradients (Group 2)................................................... 45 Figure 3.21 Body parts’ thermal sensation (Group 1 DV)..................................... 49 Figure 3.22 Body parts’ thermal sensation (Group 2 DV)..................................... 51 Figure 3.23 Thermal comfort unacceptability (Group 3)....................................... 54 Figure 3.24 Neutral temperatures for DV and MV system (Group 3)................... 54 Figure 3.25 Body parts’ thermal sensation for DV and MV cases (Group 3) ....... 55 Figure 4.1 Chamber layout .................................................................................... 61 Figure 4.2 Plan view of measuring points. ............................................................ 63 Figure 4.3 Temperature gradients (Group 1). ........................................................ 65 Figure 4.4 Temperature profiles (Group 3)............................................................ 67 Figure 4.5 Humidity gradients (Group 2). ............................................................. 68 Figure 4.6 Humidity gradients (Group 3). ............................................................. 69 Figure 4.7 CO2 gradients (Group 1)....................................................................... 70 Figure 4.8 CO2 gradients (Group 3)....................................................................... 71 Figure 4.9 Body parts’ thermal sensation (Group 1 DV)....................................... 74 Figure 4.10 Draft at different heights (Group 1).................................................... 76 Figure 4.11 Body parts’ thermal sensation (Group 2 DV)..................................... 79 Figure 4.12 Body Parts’ Thermal Sensation (Group 3) ......................................... 81 Figure 4.13 Draft at different heights (Group 3).................................................... 83 Figure 4.14 System for both ventilation modes. .................................................... 85 Figure 4.15 Conditioning process on psychrometric chart. ................................... 87 Figure 4.16 Relationship between measured temperature (symbols) and predicted temperature (lines). ......................................................................................... 91 ix CCCHHHAAAPPPTTTEEERRR 111::: IIINNNTTTRRROOODDDUUUCCCTTTIIIOOONNN CHAPTER 1 INTRODUCTION 1.1 Background In a hot and humid country like Singapore, most of the commercial buildings are air conditioned and mechanically ventilated. The mixing ventilation (MV) mode is the most common air distribution technique used in Singapore. The nature of this conventional system is to supply cool air, up to 12 ºC lower than the room air, at a strong momentum from the ceiling-mounted diffusers above the occupied zone to create a uniform thermal and pollutant’s concentration environment throughout the space. This phenomenon is also called “dilution” whereby the heat or pollutants generated by one person will be carried by air movement to the entire space. This could cause problems to the other occupants in the same space when the pollutantmixed air is inhaled by them. This can lead to a decrease in productivity. If the pollutant is harmful, the occupant can even fall ill. According to U.S. Environmental Protection Agency’s report (1990) the cost to industry has been estimated to be in the magnitude of tens of billions of dollars per year. There is another type of system, based on displacement ventilation (DV) strategy, to provide better air quality in a more energy-efficient way. This system was first applied to the welding industry in the Scandinavian countries in 1978. Since then it has been increasingly used. Recently, this system has become popular not only in industrial facilities, but also in offices and other commercial spaces. In 1989 in the Nordic countries, it was estimated that displacement ventilation accounted for a 50% market 1 CCCHHHAAAPPPTTTEEERRR 111::: IIINNNTTTRRROOODDDUUUCCCTTTIIIOOONNN share in industrial application and 25% in office application. This system can provide better air quality as it supplies cool air with low velocity near the floor level and exhausts at the ceiling level. The air is then transported within the room by the rising convection flows from the heat sources, e.g. human, PCs etc, which take the warm contaminated air from the lower parts of the room to the upper parts. In this way, the person who has not created any pollutants will not be bothered by the pollutants created by others. In comparison with the conventional mixing ventilation, displacement ventilation can achieve considerably higher ventilation efficiency and is more energy-efficient. However, research on displacement ventilation has been mainly conducted in the Scandinavian countries. There is limited research done in the tropics. As the ethnic groups and building loads in the tropics are not the same as those in Scandinavian countries, the results of such research may not be directly applicable. Moreover, in the tropics, due to the all-year-round hot and humid climate, there is high recirculation of air for most of the Air conditioning Mechanical Ventilation (ACMV) systems in order to save energy. The ratio of the return air to the total supply air could range between 70% and 90%, depending on the types of building. DV system has to comply with the energy-saving rule, i.e. 70% to 90% of the exhaust air needs to be recirculated, if it is to be used in Singapore. However, 100% outside air can be used in the Scandinavian countries for DV system due to their climatic condition. Hence, the results of the research conducted in Scandinavian countries may not be applicable in Singapore. There is a growing trend of new buildings exploring the possibility of adopting new air-conditioning technology in Singapore. The acceptance of new system can be 2 CCCHHHAAAPPPTTTEEERRR 111::: IIINNNTTTRRROOODDDUUUCCCTTTIIIOOONNN difficult in the construction industry since the suitability of such systems is not extensively investigated in the tropics. This can be detrimental during the operation of building if it cannot perform or under-perform. Therefore, there is a need to assess the feasibility and viability of displacement ventilation system in the tropics. It is therefore of great importance to conduct research to assess its applicability in the Tropics. The results and findings of this study will help local practitioners to create a comfortable environment with suitable room air temperature, supply air temperature, relative humidity and ∆T (temperature difference between head and feet level) for the occupants. The findings of this study will serve as preliminary directions for future research on DV system in the tropics. This will in turn help local practitioners and ACMV system operators in creating and/or maintaining a comfortable indoor environment economically. 1.2 Objectives and scope of research The main objectives of the research study are: a. To investigate the stratification effect of the wall supply displacement ventilation system; b. To investigate the thermal comfort and energy performances for the wall supply displacement ventilation system; c. To compare performance of the wall supply displacement ventilation system against the conventional ceiling supply mixing ventilation system based on thermal comfort, indoor air quality and energy; and d. To develop a preliminary design guide that can be used for offices in the tropics. 3 CCCHHHAAAPPPTTTEEERRR 111::: IIINNNTTTRRROOODDDUUUCCCTTTIIIOOONNN 1.3 Outline of dissertation This chapter states the background, the objectives and scope of work of this study. It is intended to point out that, though indoor air quality issues have a large scope of interest, this study would only focus on the basic aspects of indoor air quality, i.e. concentration stratification and ventilation efficiency. Chapter two introduces the fundamentals of displacement ventilation. This is substantiated by past research findings. It is arranged in the following order: temperature, pollutant’s concentration and humidity gradients; thermal comfort fundamentals and research findings; and indoor air quality basics and research findings. Chapter three presents the methodology and results of the preliminary study. The methodology includes research design, detailed experimental conditions description, methods of data collection and methods of data analysis. The results are presented in two categories, namely gradients and thermal comfort. Chapter four presents the methodology and results of the second study with a larger sample size. This study is to confirm the results and findings of the preliminary study. Therefore, the basic methodology adopted is the same as the preliminary study. Chapter five summaries the concluding remarks following the data analysis and discussion. A list of recommendations is ascertained from this study. 4 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW CHAPTER 2 LITERATURE REVIEW 2.1 Introduction Since displacement ventilation was first applied to the welding industry in 1978, it has been gaining popularity in Scandinavia as a means of ventilation to provide good indoor air quality (IAQ) and improve energy-efficiency not only for large spaces with high ceilings, such as assembly halls, but also for offices and other commercial spaces use. In 1989 in Nordic countries, it was estimated that displacement ventilation accounted for a 50% market share in industrial applications and 25% in office applications (Yuan, 1999b). A typical displacement ventilation system provides cool air at a temperature several degrees below room air temperature and at a very low velocity of less than 0.5 m/s through large-area supply devices near the floor level and extracts air at the ceiling level. The supply air spreads over the floor and then rises as it comes into contact with heat sources, e.g. persons, computers, in the occupied space. The rising air above the heat source is called a plume. Plumes carry heat and contaminants and entrain the ambient air to the upper part of the room space. Thus the airflow rate of plumes increases with height. The flow rate in the convection flow equals the supply air flow rate at a certain height above the heat source. In order to feed the convection flow above that height, the air in the upper part of the room is naturally recirculated. In this way the air will be stratified with a lower zone of fresh cool air and an upper zone of mixed and contaminated warm air. A schematic flow pattern is shown in Figure 2.1 (Yuan et al, 1998). 5 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW This chapter will present the characteristics and mechanism of displacement ventilation in detail, and some findings of previous research studies. They are categorized into: 1) gradients; 2) thermal comfort; 3) indoor air quality; and 4) energy. Figure 2.1 Sketch of displacement ventilation. (Source: Yuan et al, 1998) 2.2 Gradients in occupied space 2.2.1 Temperature gradient As mentioned earlier, the principle of displacement ventilation is a supply of cool air at low velocity near the floor level and an exhaust at the ceiling level. The air is transported within the room by the rising convection flow from the heat sources, which take the heated air from the lower parts of the room into the upper parts. At a certain height, the air stratifies, thus forming two parts: one with warm and less dense air in the upper space and the other with cool and denser air in the lower space. Figure 2.2 shows an example of the vertical temperature profile in a thermal chamber with different cooling loads (Xu et al, 2001). It is observed that the temperature profile could be separated into two regions: (1) steep temperature gradient (floor level to 1.0 - 6 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW 1.2m high), and (2) gentle temperature gradient (1.0-1.2m high to ceiling level) when the indoor heat load exists. These findings were consistent with those of the other researchers too. For example, Yuan et al (1999) conducted both measurements and computational fluid dynamics (CFD) modelling and found that the temperature gradient at the lower part would be larger than at the upper part when most of the heat sources were in the lower part of the room. Murakami et al (1998) analyzed both flow and temperature fields around a modelled standing human body using CFD program and found that the gradient became very steep between the feet and waist level. Figure 2.2 Temperature gradients in a thermal chamber with different cooling loads. (Source: Xu et al, 2001) 2.2.2 Concentration gradient When the contaminant source is combined with the heat source (this is the usual case, for example, human being generates not only heat, but CO2 and bioeffluent), the plume will carry the contaminants over the heat source to the upper zone of the room. The result is that the air in the upper zone will be polluted while the air in the lower zone is as clean as the supply air. It is necessary to note that the important characteristic of displacement ventilation system is the temperature stratification. It suppresses the 7 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW vertical mixing of air and therefore, vertical concentration can be maintained. It should also be noticed that, normally, the concentration in the upper part of the room is larger than the average concentration, i.e. it is larger than the concentration that occurs in a well mixed room. Figure 2.3 The CO2 concentration gradients at various locations in a room. (a) concentration gradients (upper); (b) plan view of measuring locations (lower). (Source: Xu et al, 2001) An example of the concentration gradient is shown in Figure 2.3 (Xu et al, 2001) where the non-dimensional concentration C* is plotted against the room height. The C* is defined as C* = (Cp-Cs)/(Ce-Cs) where Cp is the CO2 concentration at point p inside the room, Cs is CO2 concentration of supply air, and Ce is CO2 concentration of 8 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW exhaust air. It shows that the steep concentration gradient occurs between 0.6~1.2m and below 0.6m, the concentration is almost the same and much lower than that at the upper zone (above 1.2m). The results of Xu’s work were consistent with other studies, such as Yuan et al (1999a) and Murakami et al (1998) where CFD modelling was used, and Yuan et al (1999c) where measurements and CFD modelling were both employed. In all these studies, it was found that the CO2 concentration in the lower zone was lower than that in the upper zone. 2.2.3 Humidity gradient Plumes not only carry heat and contaminants, they also carry moisture. It is a common perception that relative humidity is constant throughout the whole space with conventional mixing ventilation. Applying this assumption to a space with displacement ventilation where air stratifies, having higher temperature and contaminant concentration in the upper part of the room, one may conclude that humidity also stratifies. relative humidity gradient 20.0 20.0 15.0 15.0 Height, ft Height, ft moisture gradient 10.0 5.0 0.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 m oisture ratio of room air, lb/lb*1000 1/1 load 1/2 load 11.0 10.0 5.0 0.0 20% 30% 40% 50% 60% room air relative hum idity % 1/1 load 1/2 load Figure 2.4 Relative humidity gradient and moisture ratio gradient with displacement ventilation. (Source: Kosonen et al, 2001) 9 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW Figure 2.4 shows measurement data from field studies conducted in 1988 in a foodprocessing facility located in Finland (Kosonen et al, 2001). The measurement data is presented for full load and half load. It was observed that both humidity ratio gradient and relative humidity gradient existed. 2.3 Thermal comfort Thermal comfort has been defined as "the condition of mind that expresses satisfaction with the thermal environment" (ISO 7730, 1994). The reference to "mind" emphasized that comfort is a psychological phenomenon. It is therefore often "measured" using subjective methods—survey into man’s thermal sensation votes. Man’s thermal sensation is mainly related to the thermal balance of their body as a whole. This balance is influenced by his physical activity and clothing, as well as the environmental parameters: air temperature, mean radiant temperature, air velocity and air humidity. Moreover, man’s thermal sensation can also be influenced by factors such as age, sex, body build, etc (Fanger, 1970). Over dozens of years, research on man’s thermal comfort has been carried out throughout the world, but mainly in mixing ventilation system. Recently, studies in displacement ventilation system have been gaining popularity and large amounts of beneficial results have thus been obtained. This section presents the criteria stipulated in thermal comfort standards. Some findings from past research studies will also be presented in the latter portion of this chapter. 10 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW 2.3.1 Standards on thermal comfort 2.3.3.1 Indices (ISO 7730, 1994) 2.3.3.1.1 Predicted mean vote (PMV) The PMV is an index that predicts the mean value of the votes of a large group of persons on the following 7-point thermal sensation scale: +3 +2 +1 0 -1 -2 -3 Hot Warm Slightly Warm Neutral Slightly Cool Cool Cold The PMV index can be predicted when the activity (metabolic rate) and the clothing (thermal resistance) are estimated, and the following environmental parameters are measured: air temperature, mean radiant temperature, relative humidity and partial water vapour pressure. 2.3.3.1.2 Predicted percentage of dissatisfied (PPD) The PMV index predicts the mean value of the thermal votes of a large group of people exposed to the same environment. However individual votes are scattered around this mean value and it is useful to predict the number of people likely to feel uncomfortably warm or cool. The PPD index establishes a quantitative prediction of the number of thermally dissatisfied people. The PPD predicts the percentage of a large group of people likely to feel too warm or cool, i.e. voting hot (+3), warm (+2), cool (-2), or cold (-3), on the 7-point thermal sensation scale. 11 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW The PPD-index predicts the number of thermally dissatisfied person among a large group of people. The rest of the group will feel thermally neutral, slightly warm, or slightly cool. 2.3.3.1.3 Draft rating (DR) Draft is an unwanted local cooling of the body caused by air movement. The draft rating may be expressed as the percentage of people predicted to be bothered by draft. The model of draft applies to people at light activity (mainly sedentary activity), with a thermal sensation for the whole body close to neutral. The draft rating is also called the percentage of dissatisfied due to draft, or PD. 2.3.3.2 Criteria in various standards Table 2.1 The criteria stated in different standards. Criteria ASHRAE 55-1992 CP13&Guidelines (Singapore) 80% 90% (-0.53 ºC for 40% of the locations. For 18% of the measured locations within the occupied zone, PD>15% and △t1.1-0.1>3 ºC was registered. However, the risk of discomfort due to draft and vertical temperature difference was low in some of the investigated rooms. Hence, they concluded that when displacement ventilation system is well designed, it is feasible to create good thermal comfort in rooms. Yuan et al (1999a) evaluated the performance of traditional displacement ventilation systems for small offices, large offices with partitions, classrooms, and industrial workshops under U.S. thermal and flow boundary conditions using CFD program. It was found that generally, the air velocity was less than 0.2m/s, the temperature difference between the head and foot level of a sedentary occupant was less than 2 ºC, and draft rating (PD), predicted percentage of dissatisfied (PPD) were less than 15% in the occupied zone, if the design used the guidelines shown in their paper. The PD and 13 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW PPD were high in region very close to the diffuser (0.5m). Hence, it was concluded that displacement ventilation could maintain a thermal comfortable environment. Wyon and Sandberg (1990) used thermal manikin to predict discomfort due to DV system. Serious local discomfort was identified, usually “too cold”, and most of it was due to cold legs, ankles and feet. The results indicated Equivalent Homogeneous Temperature (EHT) (WB) = 25.1°C for preferred whole-body condition. An optimum sectional air temperature of 24.4°C was suggested for mean thermal sensation to be ‘neutral’ and a range 20.9°C < T < 28.0°C based on 80% acceptability criterion was proposed. Akimoto et al (1999) evaluated the performance of a floor-supply displacement airconditioning system in comparison to a DV system with a sidewall-mounted supply unit and a ceiling-based distribution system. Thermal stratification was observed, as there was a greater vertical air temperature difference in both of the displacement system than in the ceiling-based system. A large vertical temperature difference that may cause local thermal discomfort was observed in several cases for both of the displacement systems. It was observed that the measured skin temperatures of the thermal manikin with both of the displacement systems were slightly lower than those of the ceiling-based system. However, this is not considered too low to cause local thermal discomfort. 2.3.2.2 Tropical area thermal comfort studies De Dear et al (1991) performed thermal comfort field experiments in Singapore. Results of the air-conditioned sample indicated that office buildings were overcooled, and one-third of their occupants experienced cool thermal comfort sensation. The 14 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW observation in these air-conditioned buildings was broadly consistent with the ISO, ASHRAE and local standards. PMV model’s predicted neutralities were all slightly warmer than the empirically observed neutralities by approximately 1K. Busch (1992) conducted a field study in Thailand to explore whether there was justification for adopting a comfort standard that differs from those developed for office workers accustomed to temperate climates. The neutral temperature was found to be 24.7°C EHT for air-conditioned buildings. The author determined the temperature limits of comfort zone for air-conditioned buildings – lower limit of the comfort zone was about 22°C and the upper limit reached about 28°C. These limits were broader than that stipulated by the standards. Tan (1995) carried out field and chamber study to determine whether Singaporeans’ perception of thermal comfort differ from existing literature. The neutral temperature was found to be 24.7°C which is slightly lower than 25.6°C that was found by Fanger. The author also derived PMV-PPD characterization of Singaporeans and it was found to be similar to Fanger’s PMV-PPD curve. Cheong et al (2003) performed a thermal comfort study of an air-conditioned lecture theatre in Singapore using CFD, objective and subjective measurements. It was found that thermal conditions were within limits of thermal comfort standards but the subjective responses were slightly biased towards the ‘cold’ section of the 7-point thermal sensation scale and the occupants were slightly uncomfortable at a 23°C environment. The calculated PMV and PPD were close to the subjective result. 15 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW 2.4 Indoor air quality The concern about energy efficiency has increased since the 1970’s oil crisis. This growing concern has led to many changes in the way buildings are constructed and operated. The most conspicuous ones are the reduction in ventilation rate and increase in air-tightness of buildings. In the age of rapid technology development, more and more synthetic building materials are used to create a comfortable indoor environment. These materials emit pollutants, such as formaldehyde, volatile organic compounds (VOCs) etc. The reduction of the ventilation rate, coupled with these pollutants, can accumulate gradually and finally reach a level at which they can have adverse effects on the occupants’ health. 2.4.1 Concentration distribution Concentration gradient does exist with displacement ventilation system as described in Section 2.2.2 Concentration gradient. In the studies by Yuan et al (1999a), where CFD modelling was used, and Yuan et al (1999c), where measurements and CFD modelling were both employed, it was found that the CO2 concentration in the lower zone was less than that in the upper zone. It was found that as the convective flow around a human body brings the air at a lower zone to the breathing zone, the occupant actually breathes air with lower concentration of contaminant than those at the nose level in the middle of the room. In Murakami’s study (1998) where CFD program was used, three cases of concentration distribution prediction were carried out at different locations of contaminant generation. It was found that the rising stream around the body surface was not broken by the surrounding airflow. The air quality at the breathing zone 16 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW depended on the location of the contaminant generation. When contaminants were generated in the upper part of the room above the breathing height, and the air in the lower part of the room was relatively clean, the rising stream of air had a positive influence on the quality of the inhaled air. Conversely, the rising stream had a negative influence on the quality of the inhaled air when the contaminants were generated in the lower part of the room below the breathing height, and the air in the lower part of the room was relatively dirty. 2.4.2 Age of air The age of air is defined as the time for all air molecules to travel from the air supply device to a point in the space. It can be derived from the measured transient history of the tracer gas concentration. Several field measurements with displacement ventilation system (Yuan et al, 1999; Murakami et al, 1998; Xing et al, 2001; Awbi, 1998; Seppanen et al, 1989) have been performed using the age of air concept. It was found that the mean age of air in the lower part of the room was much younger than in the upper part of the room. Furthermore, Awbi (1998) reported from the measurement data that the age of air at the breathing zone is about 40% lower than the mean value of the occupied zone with displacement ventilation. Similar result was also found by Xing et al (2001) in their studies in which three types of supply units were used in a series of tests: flat-faced wall unit, semi-circular wall unit, and floor swirl unit in a displacement ventilated environmental chamber. It was found that the local mean age of air at the breathing zone of a seated mannequin was 35% and 50% lower than that in the occupied zone. While for the standing mannequin 17 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW the percentage was between 10% and 20%. The difference depended on the type of supply unit. 2.4.3 Ventilation effectiveness Different researchers may have different definitions of the ventilation effectiveness although they are describing the same mechanism. According to Sandberg (1985) the Local Ventilation Effectiveness was defined as: ε= C e (∞ ) − C s (∞ ) C p (∞ ) − C s (∞ ) While in Yuan’s study (1999a) ventilation effectiveness was defined as: η= Ch − Cs . Ce − C s In the above formulas, C is the contaminant concentration, subscripts e, s, p and h refer to point in exhaust, point in supply, point in a room and point at the head level of a sedentary person respectively. Xing et al (2001) carried out measurements with the presence of a heated mannequin and other heat sources and found that the ventilation effectiveness at the breathing zone for both the seated and standing mannequins were greater than for a point at the same height in the chamber for the tests with all DV units, because the mannequin entrained fresh air from the fresh air layer on the floor into the breathing zone. Yuan et al (1999) studied 56 cases using CFD simulation program and found that the ventilation effectiveness of these cases varied between 1.2 and 2. Since the ventilation 18 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW effectiveness for perfect mixing ventilation is 1.0, they concluded that displacement ventilation did provide better indoor air quality. Awbi (1998) reported that the ventilation effectiveness for contaminant distribution (εc) was almost the same for mixing and displacement ventilation. However, the ventilation effectiveness for heat distribution ( ε t = To − Ti T − Ti , T is temperature, (°C), the subscripts i and o refer to inlet and outlet respectively and (¯) represents the mean value for the occupied zone) with displacement ventilation was almost twice the value with mixing ventilation. 2.5 Energy Annual energy consumption over a life-cycle is an important criterion for the evaluation of a ventilation system. Almost all the energy analyses in the literature were done by numerical simulation because it is too expensive and time consuming to conduct hour-by-hour measurements for a building based on a yearly basis. Seppanen et al (1989) evaluated the energy performance of displacement and mixing ventilation systems in a high-rise office building in the United States. The study analyzed the north, south, and core zones of the buildings in four representative U.S. climates and found that the energy consumed by displacement ventilation systems with heat recovery and variable-air-volume (VAV) flow control were similar to the energy consumption of conventional air distribution systems operated with recirculation. Chen et al (1990) used the cooling load program ACCURACY and energy analysis program ENERK to calculate the space load and the annual energy consumption of a room based on the weather data of the Dutch short reference year. It was found that in 19 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW the variable-air-volume system with the same lowest supply air temperature (16 ºC), the DV system would save 26% energy on the chiller and the ventilator and 3% on the boiler, compared with the MV system. The cost of annual energy consumption for the displacement ventilation system was 16% smaller than that for the well-mixed system. However, if the lowest supply air temperature for the DV system remained at 16 ºC and that for the MV system was controlled at 12.5 ºC, the cost of annual energy consumption was nearly the same. Hensen et al (1995) carried out simulations using a computer model of a typical office module located in The Netherlands (temperature sea climate) and found that applying displacement ventilation in a case of relative low casual gains (30 W/m2) resulted in energy savings of up to 14% for cooling during the summer months. During the rest of the year hardly any saving was to be expected. The overall annual energy consumption for cooling could be up to 10% lower. At causal gains above about 35 W/m2 the energy consumption for cooling would be considerably higher than in the case of mixing system only. Hu et al (1999) used a detailed computer simulation method to study the energy consumption of displacement and mixing ventilation systems for an individual office, a classroom, and a workshop for five U.S. climatic regions. The study showed that when free cooling was used for both ventilation systems in the shoulder seasons, when the supply air temperature for displacement ventilation system was 20 ºC while that for the mixing ventilation system was 12.8 ºC, the displacement ventilation system might use more fan energy and less chiller and boiler energy than the mixing ventilation system. The total energy used was slightly less with displacement ventilation. 20 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW Livchak (2001) used displacement ventilation system selection software to compare the energy consumption for a computer room. It was found that if 100% outside air was used during the free cooling season, displacement ventilation system with the supply air temperature of 18 ºC allowed a reduction of the chiller cooling capacity by 1.4 kW and its annual energy consumption by 33%, as compared to mixing ventilation system. 2.6 Conclusions and hypotheses 2.6.1 Conclusions Through the literature review, the following conclusions were drawn: i. Displacement ventilation can create a thermally comfortable environment that has low air velocity, a small temperature difference between the head and foot level, and a low percentage of dissatisfied people if the system is well designed; ii. Displacement ventilation can provide better indoor air quality with lower pollutant concentration and higher ventilation effectiveness in the occupied zone, if the system is well designed. However, the improvement of IAQ could remain small at large recirculation ratios. iii. Displacement ventilation can use less energy while providing thermally comfortable environment and better indoor air quality, if the system is well designed and controlled. iv. Though there are large numbers of research studies done on displacement ventilation, research in the tropics is rather limited. It is therefore of great importance to conduct research to assess its viability in the tropics. 21 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW 2.6.2 Hypotheses The following hypotheses can be drawn after the literature review. i. Displacement ventilation system is applicable to the tropics, although there are geographical, climatic, and racial differences. ii. Displacement ventilation can create a thermally comfortable environment and provide better indoor air quality as compared to mixing ventilation in the tropics. iii. Displacement ventilation can use less energy while providing a thermally comfortable and better indoor air quality environment in the tropical conditions. 22 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY CHAPTER 3 PRELIMINARY STUDY 3.1 Methodology 3.1.1 Research design Experiments and surveys were employed in this research to investigate the thermal and indoor air quality performances of displacement ventilation and assess the acceptability of DV system by the local subjects. In these experiments, possible cause and effect relationships were investigated by exposing experimental groups to certain treatments and comparing the results between groups. In the questionnaire survey, primary data based on a sample was collected, and inferences were made on the population. The approach of the research design is shown in Figure 3.1. In the data collection stage, objective data such as temperature and velocity were measured using instruments in the experiments, and subjective data were collected by means of questionnaire survey. In data analysis stage, data from the objective measurements would be analyzed using normal methods such as normalization and tabulation. Data from survey would be analyzed by statistical tools such as T test and ANOVA. To investigate the thermal and indoor air quality performances of displacement ventilation system, a total of 8 displacement ventilation cases and 4 mixing ventilation cases were formulated. These cases are assigned to different groups depending on the supply air temperature, room air temperature and relative humidity. 23 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Literature Review z Dry bulb Temperature Relative humidity / Dew point temperature Velocity Flow rate CO2 concentration z Subjects survey z z Research Hypothesis z z Research Design Data Collection Data Analysis Conclusions z Objective measurement z Subjective assessment z Normal Analysis z Statistical Analysis Figure 3.1 Total approach employed in this research study. 3.1.1.1 Group 1 In this group, room air temperature was kept constant with variation in the supply air temperature. This led to a variation in the supply air flow rate as the heat gain in the room remained unchanged. The second parameter that was varied is the room air humidity (both absolute and relative humidity). The ratio of outdoor air flow rate to the total supply air flow rate was kept constant at about 36%. Table 3.1 shows the various conditions for Group 1. This group of cases was formulated based on the following hypotheses: i. Subjects would have different thermal sensation when they are exposed to different room humidity levels; ii. Subjects would have different thermal sensation when they are exposed to different supply air temperatures; and 24 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY iii. Subjects would have different thermal sensation when they are seated near or far away from the supply unit. A total of 12 subjects were subjected to the condition in each case. In each experiment, 3 subjects were seated in workstations (WS 1-3 in Figure 3.3) with each one having a computer to allow them to do their own work. The duration of each experiment was 2 hours. Table 3.1 Test conditions for Group 1. Case Ts (°C) Tr (°C) RH (%) 1 16.2 23.1 55.6 2 17.7 23.1 62.0 3 18.9 23.1 68.1 4 20.3 23.2 77.0 3.1.1.2 Group 2 In this group, both the supply air temperature and room air temperature were allowed to vary while the room relative humidity at 1.3m height remained unchanged. The ratio of outdoor air flow rate to the total supply air flow rate was also kept constant at about 36%. Table 3.2 shows the various conditions for Group 2. This group of cases was formulated based on the following hypotheses: i. Subjects would have different thermal sensation when they are exposed to different room air temperatures; and ii. Subjects would have different thermal sensation when they are seated near or far away from the supply unit. 25 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY A total of 12 subjects were subjected to the condition in each case. In each experiment, 6 subjects were involved with each workstation having a computer to allow them to do their own work. The duration of each experiment was 2 hours. Table 3.2 Test conditions for Group 2. Case Ts (°C) Tr (°C) RH (%) 5 16.8 22.2 65.0 6 18.7 24.2 65.5 7 20.4 26.2 64.7 3.1.1.3 Group 3 The objective of the test for this group is to compare the displacement ventilation system with the conventional mixing ventilation system. Cases in each sub-group have similar room air temperature, relative humidity level and outdoor air ratio (36%). Table 3.3 shows the various conditions for Group 3. This group of cases was formulated based on the following hypothesis: Subjects would have different thermal sensation with DV and MV systems even though the room air temperature and relative humidity are similar. A total of 12 subjects were subjected to the condition in each case. In each experiment, 6 subjects were involved with each workstation having a computer. Cases in Group 3 with initial “D” refer to displacement cases while initial “M” denotes mixing cases. The duration of each experiment was 2 hours 26 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Table 3.3 Test conditions for Group 3. Case Ts (°C) Tr (°C) RH (%) D1 16.8 22.2 65.0 M1 15.7 22.0 65.9 D2 18.5 23.3 67.8 M2 17.8 22.9 70.4 D3 18.7 24.2 65.5 M3 17.3 24.0 66.5 D4 20.4 26.2 64.7 M4 17.9 26.1 64.7 A B C D 3.1.2 Methods of data collection 3.1.2.1 Subjective assessment 3.1.2.1.1 Subjects Twelve (five male and seven female) college-age students participated in the experiments as subjects. The subjects were recruited based on the following criteria: exposed to local tropical climate for more than 6 months, familiarity with a PC, impartiality to the chamber in which the study was carried out, and absence of chronic diseases, asthma, allergy and hey-fever etc. The statistical summary of these subjects is shown in Appendix A1. All subjects were volunteers who were paid for taking part in these experiments. Subjects were instructed to eat normally before arrival at the thermal chamber. No intakes of alcohol or drugs were allowed 24 hours prior to each experiment. During the experiments, subjects were asked to be dressed in typical office 27 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY attire to simulate an office environment. They were allowed to put on additional clothing during the experiment to maintain thermal neutrality. Subjects were restricted to only deskbound activities. During the experiments, they were not allowed to eat anything. They could drink only plain water. Subjects were randomly exposed to different test conditions on different days and were kept “blind” to the test conditions to avoid biased results. They were exposed to the test conditions in groups of three or six as stated in Section 3.1.1. 3.1.2.1.2 Subjective assessment protocol Each experiment proceeded as follows: i. Subjects arrived at the chamber 15 min prior to the commencement of the experiment. They were seated in the control room and briefed about the procedure. During this period, as they acclimatized to the environment, they started to answer Section 1 of the questionnaire which inquired about their personal particulars, thermal sensation for the whole body and thermal comfort acceptability. ii. After the acclimatization period, these subjects entered the chamber and started to answer Section 2 of the questionnaire on their thermal sensation for the whole body and thermal comfort acceptability. iii. For every ten minutes thereafter, subjects would complete a questionnaire on their thermal sensation for different parts of the body, thermal comfort acceptability and air movement detection and acceptability. iv. At the end of 60 minutes, subjects would complete the last questionnaire inquiring about their thermal sensation for different parts of the body, thermal comfort acceptability, air movement detection and acceptability, and the type of clothing the subjects were wearing. 28 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY 3.1.2.1.3 Questionnaire A set of the questionnaire is given in Appendix B1. A simple Yes/No category scale was used to ascertain the air movement sensation. If the subjects indicated “Yes”, they would go on to the next question and mark on the related body part(s) where they felt the air movement. Divided continuous scale was used to determine the acceptability for thermal comfort and air movement. This scale is divided into two parts with “Just Acceptable” and “Just Unacceptable” in the middle. A sample of the scale is shown in Figure 3.2. This is to allow the subjects to make a definite choice and grade the degree of acceptability or unacceptability clearly. Undivided continuous scale was used for the remaining parts of the questionnaire, as shown in Figure 3.2. Just Acceptable Very Unacceptable Hot Very Acceptable Just Unacceptable |_______________________________________________| Cold Figure 3.2 Continuous scale used in the questionnaire. (a) Divided continuous scale (upper); (b) Undivided continuous scale (lower). 3.1.2.2 Objective measurement protocol 3.1.2.2.1 Objective parameters Objective measurement started 5~15 minutes prior to the experiments and ended 5~15 minutes after the experiment. During each experiment the following parameters were 29 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY measured: i. Temperature: supply air temperature (Ts), return air temperature (Te), room air temperature (Tr), wall surface temperature (Tw), ceiling surface temperature (Tc), floor surface temperature (Tf); ii. Room air relative humidity and dew point temperature; iii. Air velocity at the area near the supply unit; iv. Supply and return air flow rates; and v. Concentration of CO2, TVOC and formaldehyde. 3.1.2.2.2 Instrumentation 3.1.2.2.2.1 Thermal chamber The experiments were carried out in the thermal chamber, 6.6m (L) x 3.7m (W) x 2.6m (H), at the School of Design and Environment, National University of Singapore. The Air-Conditioning and Mechanical Ventilation (ACMV) system is capable of controlling the air temperature and airflow rates by adjusting the off coil temperature and fan speed using the computer controller to achieve the required room conditions. It can be operated in either mixing (MV) or displacement (DV) modes. The room is illuminated by 6 sets of twin double-battens fluorescent lights. The power consumption of each fluorescent tube is 36W. There are six workstations inside the chamber with two large fixed glass windows (W×H=1.47×1.17 m) on one side of the wall to simulate a typical office environment. Each workstation consists of a table, a chair, and a personal computer (PC). Beside each workstation there is a partition panel, measuring 1.5m (Height), 1.0m (Wide) and 5.5cm (Thick). Figure 3.3 shows the layout of the chamber. 30 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY E1 E3 T1 MV diffuser T2 DV unit E2 Figure 3.3 Chamber layout. Note: WS denotes workstation number and E denotes extract grille number. In the DV mode, the air is supplied from a floor-standing, low velocity, semi-circular unit at one end of the chamber and extracted from two ceiling grilles, E1 and E2. In the MV mode, the air is supplied from two square ceiling diffusers, T1 and T2, and extracted from two ceiling grilles, E2 and E3. The floor-standing, low velocity, semi-circular supply unit, return grille and square ceiling diffuser are shown in Figures 3.4~3.6, respectively. Figure 3.4 Floor-standing, low velocity, semi-circular supply unit. 31 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Figure 3.5 Return grille. 3.1.2.2.2.2 Figure 3.6 Supply diffuser. Instruments Table 3.4 shows the list of instruments employed to measure all the essential parameters. Figures 3.7-3.11 show the photographs of some of the instruments. Table 3.4 Instrumentations. Parameter Instrument Accuracy Temperature Type T thermocouple wire ±0.2°C Room air relative humidity Portable RH sensor ±5% RH Dew point temperature Dew-point hygrometer ±0.15°C Supply and return air flow rate Hood and vane anemometer ±1% Concentration of CO2, Photoacoustic spectrometer formaldehyde, TVOC multi-gas analyser Air flow rate in duct PRA damper Air velocity Omni-directional velocity sensor ±2% ±5% ±0.02m/s ±1% of readings 32 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Figure 3.7 Portable RH sensor. Figure 3.8 Dew-point hygrometer. Figure 3.9 Hood. Figure 3.10 Photoacoustic spectrometer multi-gas analyser. Figure 3.11 PRA damper (left) and Omni-directional velocity sensors (right). 33 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY 3.1.2.2.3 Measuring locations Figure 3.12 shows the measuring locations in the chamber. A detailed description of each measuring location is stated as follows: i. Supply and return air temperature were measured by thermocouple wires placed inside the supply unit and return grilles, respectively. ii. Room air temperature measuring locations: Points 1 to 8. Thermocouple wires were placed at three heights namely 0.1m, 1.3m and 2.5m at Points 3, 4, 5, 7, 8; and six heights namely 0.1m, 0.3m 0.6m, 1.3m, 1.7m and 2.5m at Points 1, 2, and 6 (the reason for this is to monitor the temperature profile along and across the supply air flow more precisely). iii. Wall surface temperature: points marked as 8. At these points thermocouple wires were also placed at three heights namely 0.1m, 1.3m and 2.5m. iv. Ceiling and floor surface temperature: points marked as 7. v. Room air relative humidity: Points 2, 4, 6 and 8. HOBO meters were placed at heights of 0.1m, 1.3m and 2.5m. vi. Dew point temperature: Points 1, 3, 5 and 7. Dew-point hygrometer probes were also placed at heights of 0.1m, 1.3m and 2.5m. vii. Air velocity in the region around the supply unit: Points A to L. Populex sensors were placed at four heights: 0.1m, 0.6m, 1.1m and 1.7m at these points. viii. Supply and return air flow rate: measured at the PRA damper directly above the supply unit or return grille on the ductwork. ix. Concentration of CO2, formaldehyde and TVOC: Point 1, at three heights namely 0.1m, 1.3m and 2.5m. 34 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Figure 3.12 Plan view of measuring points. 3.1.2.2.4 Measurement procedure The above mentioned parameters were measured according to the following principles: i. For DV cases, during the first hour of the experiment, all the instruments were measuring and logging at their original places. This is to ensure that the subjects would not be disturbed by the measuring instruments and their subjective response would be unbiased. During the second hour, the dew point hygrometer sampling tubes were moved from Point 7 to Point 1 at an interval of 15 minutes, and the Omni-directional velocity sensors were moved from Points A, B, C, D to Points E, F, G, H respectively and then to Points I, J, K, L respectively at an interval of 20 minutes. During the entire experiment period (from Aug 7 to Sep 7, 2002) there were only 3 Dew-point hygrometers and 16 Populex sensors. To monitor the stratification of the dew-point temperature, Dew-point hygrometers were placed at three heights at the same location. In order to obtain real-time velocities of the DV 35 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY supply air profile, Populex sensors were placed at four heights at the same location. ii. For MV cases, Dew-point hygrometers were placed at Point 7 and Omnidirectional velocity sensors were placed at Points 1, 3, 5 and 7. The other instruments were kept at the same place as in the DV cases. 3.1.3 Data collection and processing 3.1.3.1 Objective data Objective data were measured and logged simultaneously during the experiment. The logging intervals of various parameters are stated as follow: i. Air temperature: 5 minutes. ii. Room air relative humidity: 5 minutes. iii. Room air dew point temperature: 30 seconds. iv. Air velocity: 10 seconds. v. Concentration of gases: 12 minutes. The interval for concentration of gases is quite long (12’). This is because the instrument has 6 channels and the time taken for each channel is about 2 minutes. The interval for air velocity is quite short (10”). The objective of this is to obtain the realtime velocity profile of the DV supply air. The reason for the short interval for dew point temperature is: humidity was a bit difficult to control during the experiment due to the limitation of the system. To obtain the required and stable condition for the experiments, the real-time condition inside the experimental chamber needed to be known and the system needed to be tuned accordingly. 36 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY 3.1.3.2 Subjective data Subjective data were acquired from various scales on the questionnaires and processed according to the following principles: i. Results from undivided continuous scale for thermal sensation were categorized using ASHRAE’s seven point scale: (-3) cold, (-2) cool, (-1) slightly cool, (0) neutral, (+1) slightly warm, (+2) warm and (+3) hot. ii. The divided continuous scales for acceptability of thermal comfort and air movement were classified as (-1) Very Unacceptable, (0) Just Unacceptable/Just Acceptable and (+1) Very Acceptable. iii. Unacceptable air movement is considered as draft. 3.1.4 Method of data analysis Only data for the last 10 minutes of each experiment were used for analysis since it is believed that subjects would have acclimatized after they were exposed to the environment for about 60 minutes. Microsoft Excel’s Analysis ToolPak and statistical software SPSS (Version 11) were used to analyze the results obtained from the questionnaires. Shapiro-Wilk Test was performed to test whether the samples came from normal population. This test will indicate whether parametric or non-parametric tests would be appropriate for statistical analysis of the data. When the test showed that the samples were normally distributed, parametric tests such as Paired T test and Analysis of Variance (ANOVA) were used to determine whether there were any significant 37 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY differences between different cases. If not, non-parametric tests such as Wilcoxon test and Friedman test were used. Simple linear regression is a method for analyzing the relation between one independent variable and one dependent variable. It was used to determine whether there was any relationship between PMV and AMV, PPD and APD. 38 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY 3.2 Results and discussion 3.2.1 Gradients 3.2.1.1 Temperature gradient 3.2.1.1.1 General description Figure 3.13 shows a typical temperature gradient of one of the cases. It is observed that at 0.1m height, the air temperature rises steadily from Point 1 to Point 7. This is because of the heat transfer from the floor to the air. The conditioned air leaving the DV terminal becomes warmer as it moves into the space. There is continuous heat transfer in the form of conduction and convention between the floor and the air. The temperature gradient, as shown in Figure 3.13, shows two distinct portions: a steep gradient up to a level of 1.3m and a gentler gradient from 1.3m to the ceiling level. This concurs with those studies found in past research works. Figure 3.14 shows the temperature gradient obtained from the other measuring points under the same condition as in Figure 3.13. This is a general phenomenon of displacement ventilation in all the DV cases studied in this research. The temperature profile shows that the maximum temperature difference between various measuring points at the same horizontal plane up to 1.3m is about 1.5 ºC. However, beyond 1.3m level, there is almost no difference among the measuring points. This is due to the existence of a layer of air called “stratification front” in displacement ventilation. The air above this layer is re-circulated and mixed. Hence, a more uniform air temperature is found beyond 1.3m level. 39 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Height (m) Temperature gradient at different locations 3 2.5 2 1.5 1 0.5 0 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 Temp (ºC) Loc-3 Loc-5 Loc-1 22.5 23.0 23.5 24.0 23.0 23.5 24.0 Loc-7 Figure 3.13 Typical temperature gradient (1). Temperature gradient at different locations Height (m) 3 2.5 2 1.5 1 0.5 0 18.0 18.5 19.0 19.5 20.0 20.5 21.0 21.5 22.0 22.5 Temp (ºC) Loc-1 Loc-2 Loc-6 Figure 3.14 Typical temperature gradient (2). 3.2.1.1.2 Effect of different supply air temperature and flow rate with the same room air temperature (Group 1) Figure 3.15 shows the temperature gradients of Group 1. The variation in the supply air temperature led to different gradients. It is observed that, with the supply air temperature rising from Case 1 to Case 4, the temperature difference between various measuring points at the same horizontal plane decreases until it reaches 1.3m. 40 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Height (m) Temperature gradients at different locations 3 2.5 2 1.5 1 0.5 0 18 18.5 19 19.5 20 20.5 21 21.5 22 22.5 23 23.5 24 24.5 Temp (ºC) Case-1, Loc-1 Case-1, Loc-2 Case-1, Loc-6 Case-2, Loc-1 Case-2, Loc-2 Case-2, Loc-6 Case-3, Loc-1 Case-3, Loc-2 Case-3, Loc-6 Case-4, Loc-1 Case-4, Loc-2 Case-4, Loc-6 Figure 3.15 Temperature gradients (Group 1). Temperature gradients at different locations 3 Height (m) 2.5 2 1.5 1 0.5 0 18 18. 19 19. 20 5 5 20. 21 21. 22 22. 23 5 5 5 Temp (ºC) 23. 24 24. 25 25. 26 5 5 5 Case-5, Loc-1 Case-5, Loc-2 Case-5, Loc-6 Case-6, Loc-1 Case-6, Loc-2 Case-6, Loc-6 Case-7, Loc-1 Case-7, Loc-2 Case-7, Loc-6 26. 27 5 Figure 3.16 Temperature gradients (Group 2). 3.2.1.1.3 Effect of different supply air temperature and flow rate with different room air temperature (Group 2) Figure 3.16 shows the temperature gradients for Group 2. With the same temperature difference between supply air and room air, these gradients have the same profiles. In addition it is observed that this group of temperature profiles has steep gradient up to 1.7m. However, steep gradients occur up to 1.3m for Group 1. This could be due to the 41 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY different supply air flow rates, as shown in Appendix C1. With higher supply air flow rate, the convection flow could have more air fed from the lower zone and less air fed from the upper zone. Therefore, the lower zone could expand and the stratification layer could be pushed upwards. 3.2.1.2 Humidity gradient 3.2.1.2.1 Effect of different supply air temperature and flow rate with the same room air temperature (Group 1) Figure 3.17 shows humidity gradients of Group 1. These gradients are significant with displacement ventilation. This is consistent with the field study by Kosonen (2001). This group of gradients has different moisture levels. This is due to different supply air temperatures for different cases. The moisture ratios of the supply air differ with varying supply air temperatures. Hence the humidity level of various cases should be different with the same humidity load inside the room. The higher supply air temperature will contain more moisture and this leads to higher spatial humidity level. Humidity gradients at different locations Height (m) 3 2 1 0 8 8.5 9 9.5 10 10.5 11 11.5 12 12.5 Moisture ratio (g/Kg) Case-1, Loc-7 Case-2, Loc-7 Case-3, Loc-7 Case-4, Loc-7 Case-1, Loc-5 Case-2, Loc-5 Case-3, Loc-5 Case-4, Loc-5 Case-1, Loc-3 Case-2, Loc-3 Case-3, Loc-3 Case-4, Loc-3 Case-1, Loc-1 Case-2, Loc-1 Case-3, Loc-1 Case-4, Loc-1 Figure 3.17 Humidity gradients (Group 1). 42 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY It is observed that Case-1 has the most significant gradient while Case-4 has the least. This is due to the lowest supply air temperature in Case-1 and the highest in Case-4. With the same room air temperature, Case-1 could have larger difference in the supply-return air temperature and thus lower supply air flow rate while Case-4 has a smaller difference in supply-return air temperature and higher air flow rate. In other words, with higher air flow rate, the moisture concentration inside the room will be lower, due to the principle of “dilution”, for a given internal humidity load. 3.2.1.2.2 Effect of different supply air temperature and flow rate with different room air temperature (Group 2) Figure 3.18 shows the humidity gradients of Group 2. These cases have different humidity levels. The reason is the same as that of Group 1. Case-6 has the same humidity level as Case-7 at 0.1m height, although their supply air temperature was different. These two cases have the same off-coil temperature, in other words, the same amount of moisture removed from the supply air. Hence their moisture content was the same. In addition, it is observed that Case 7 has the most significant humidity gradient. This is partly because at a room air temperature of 26 º C, the subjects’ perspiration rate increased considerably and therefore the humidity load increased accordingly. The significant humidity gradient could also be due to the low supply air flow rate. With low air flow rate, the dilution effect of the supply air is small. Hence, the moisture concentration inside the room would be at a higher level as compared to the supply air. 43 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Humidity gradients at different locations Height (m) 3 2 1 0 9 9.5 10 10.5 11 11.5 12 12.5 13 13.5 14 Moisture ratio (g/Kg) Case-5, Loc-7 Case-5, Loc-5 Case-5, Loc-3 Case-5, Loc-1 Case-6, Loc-7 Case-6, Loc-5 Case-6, Loc-3 Case-6, Loc-1 Case-7, Loc-7 Case-7, Loc-5 Case-7, Loc-3 Case-7, Loc-1 Figure 3.18 Humidity gradients (Group 2). 3.2.1.3 Concentration gradient of carbon dioxide 3.2.1.3.1 Effect of flow rate with the same room air temperature (Group 1) Figure 3.19 shows the CO2 concentration gradients of Group 1. It is observed that stratification occurs at 1.3m height. Each case has a different CO2 gradient due to the variation in the outdoor air flow rates. Cases 1 and 2 have higher CO2 concentration at 1.3m and 2.5m as compared to Cases 3 and 4 due to lower outdoor air flow rate. Heights (m) CO2 gradients of different cases 3 2 1 0 450 500 Case-1 550 600 CO2 (ppm) Case-2 Case-3 650 700 Case-4 Figure 3.19 Carbon dioxide gradients (Group 1). 44 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY 3.2.1.3.2 Effect of flow rate with different room air temperature (Group 2) CO2 gradients of different cases Heights (m) 3 2 1 0 500 550 600 650 700 750 800 850 900 CO2 (ppm) Case-5 Case-6 Case-7 Figure 3.20 Carbon dioxide gradients (Group 2). Figure 3.20 shows the CO2 concentration gradients of Group 2. Case 5 has higher outside air flow rate of about 65 l/s and therefore has lowest CO2 concentration at 1.3m and 2.5m height. On the other hand, Case 7 has lower outside air flow rate of about 26 l/s and hence has highest CO2 concentration at 1.3m and 2.5m height. 3.2.2 Thermal comfort 3.2.2.1 Effect of different supply air temperature & relative humidity on thermal comfort (Group 1) 3.2.2.1.1 Overall Actual Mean Vote (AMV) and comfort acceptability Table 3.5 shows average clo value, AMV and comfort (un)acceptability values of Group 1. The experimental conditions are well controlled with small standard deviation of supply air temperature, room air temperature and RH. Table 3.5 shows that the Overall AMV values for all cases are between ‘slightly cool’ and ‘slightly warm’. Statistical analysis shows that the differences among these four cases are significant (p=0.0145). The average comfort acceptability votes are similar 45 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY for all cases and fall between 0.3 and 0.4 which are on the acceptable side. There are no significant differences among these four cases on acceptability votes. The Overall AMV values show that at 23°C all the cases’ room conditions are in the comfort zone and Case 2’s test condition is closest to the “neutral” point. Table 3.5 shows that all cases have fulfilled the “less than 20%” unacceptability criterion (ASHRAE Standard 55-1992, ISO 7730) with Case 4 having the most number of unacceptability votes and Case 3 having the least number of unacceptability votes. Case 4 has higher supply air flow rate in Group 1 as shown in Appendix C2. Hence the higher mean velocity is the reason for the higher unacceptability votes. Table 3.5 Average clo value, overall AMV and average comfort acceptability (Group 1 DV) Case 1 Case 2 Case 3 Case 4 Tr (°C) 23.1 ± 0.1 23.1 ± 0.1 23.1 ± 0.1 23.2 ± 0.09 Ts (°C) 16.2 ± 0.3 17.7 ± 0.2 18.9 ± 0.2 20.3 ± 0.2 RH (%) 55.6 ± 1.2 61.9 ± 0.9 68.1 ± 1.0 77.0 ± 1.4 Average Clo Value 0.53 ± 0.17 0.59 ± 0.20 0.49 ± 0.15 0.53 ± 0.15 Overall AMV -0.38 ± 0.6 -0.06 ± 0.88 -0.32 ± 0.61 0.19 ± 1.01 Acceptability 0.32 ± 0.36 0.36 ± 0.38 0.37 ± 0.35 0.36 ± 0.45 12.50 8.33 4.76 16.67 Unacceptability (%) Note: the figures following “±” denote standard deviation. Out of all these unacceptability votes, 60% are from the people seated at workstation closest to the supply unit and 30% at workstation furthest away from the supply unit. The unacceptability votes at the workstation closest to the supply unit may be due to the lower temperature and/or higher air velocity. The workstation furthest away from 46 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY the supply unit is directly under the exhaust grille, thus the unacceptability votes may be due to higher temperature and/or lower air velocity. 3.2.2.1.2 Effect of different supply air temperature on thermal comfort Subjects at workstation closest to the supply unit are analyzed since they are most susceptible to the supply air temperature. Table 3.6 shows that Cases 1 and 4 have significant differences on thermal sensation for body (p = 0.036) and arms (p = 0.009), with mean value of ‘slightly cool’ for Case 1 and ‘slightly warm’ for Case 4. However, no significant difference exists between these two cases on Overall AMV and comfort acceptability. Hence, supply air temperature can go below 18˚C as recommended by other researchers (Jackman, 1990b and Yuan et al, 1998) without significantly affecting thermal comfort. Most importantly, with the decrease in supply air temperature to about 16 °C, a larger supply-return temperature difference can be obtained and therefore, the supply air flow rate can be reduced and fan energy can be saved. Table 3.6 Mean thermal sensation and comfort acceptability at the workstation closest to the supply unit (Group 1 DV) Ts (°C) Overall AMV Case 1 Case 2 Case 3 Case 4 16.2 ± 0.3 17.7 ± 0.2 18.9 ± 0.2 20.3 ± 0.2 -0.41 ± 0.71 0.11 ± 0.91 -0.42 ± 0.71 0.38 ± 1.15 Comfort Acceptability 0.30 ± 0.44 0.35 ± 0.40 0.28 ± 0.41 0.30 ± 0.46 Body AMV -0.43 ± 0.63 0.11 ± 0.88 -0.25 ± 0.80 0.39 ± 1.15 Arms AMV -0.63 ± 0.96 0.01 ± 0.95 -0.52 ± 0.74 0.26 ± 1.07 47 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY 3.2.2.1.3 Effect of distance from supply unit on thermal comfort Table 3.7 shows the comparison results between the subjects seated at the workstations closest to the supply unit and those seated furthest away from the supply unit. It is observed that subjects seated closest to the supply unit (distance 0.4m) have lower Overall AMV and lower comfort acceptability than those seated furthest away from the supply unit. However, these differences are not statistically significant and the mean thermal sensation votes still lie within the comfort zone. This result suggests that occupants can be seated close to the supply unit (distance 0.4m) without sacrificing the usable floor area. However, this insignificant result could be due to the small sample size of this study (12 subjects altogether). Therefore, it needs to be further examined in future study. Table 3.7 Mean thermal sensation and comfort acceptability between workstations nearest and furthest from the supply unit (Group 1 DV) Overall AMV Comfort Acceptability 3.2.2.1.4 Workstation Furthest Workstation Nearest to Away From Supply unit Supply unit Case 1 -0.21 ± 0.56 -0.41 ± 0.71 Case 2 0.34 ± 0.84 0.11 ± 0.91 Case 3 0.48 ± 0.39 -0.30 ± 0.84 Case 4 0.66 ± 0.93 0.38 ± 1.15 Case 1 0.41 ± 0.36 0.30 ± 0.44 Case 2 0.38 ± 0.44 0.35 ± 0.40 Case 3 0.41 ± 0.47 0.37 ± 0.46 Case 4 0.28 ± 0.50 0.30 ± 0.46 Effect of vertical temperature difference on local thermal sensation Figure 3.21 shows that for all cases, thermal sensation votes for feet and arms are lower than other body parts. During the experiments, subjects normally wore short 48 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY sleeve shirts and this explains the cooler sensation at arms. The cooler sensation at feet may be due to the low temperature of between 18 and 21 °C at the feet level and/or open-toe type shoes worn by the subjects. There is no statistical significant difference between the thermal sensation votes at different body parts. Hence it shows that vertical temperature difference of larger than 3°C between height 0.1m and 1.3m, as illustrated in Figure 3.15, is still tolerable to tropical subjects. It does not have Mean Thermal Sensation Votes significant effect on local thermal sensation and comfort. 0.5 0.3 0.1 -0.1 Feet Calf Thigh Waist Body Arms Case 1 -0.3 Case 2 -0.5 Case 3 -0.7 Case 4 Figure 3.21 Body parts’ thermal sensation (Group 1 DV) 3.2.2.2 Effect of different room air temperature on thermal comfort (Group 2) 3.2.2.2.1 Overall AMV and comfort acceptability Table 3.8 shows average clo value, AMV and comfort (un)acceptability values of Group 2 cases. The experiments were well controlled since the standard deviation values of supply air temperature, room air temperature and room RH are quite small. The table shows that the Overall AMV values for all cases fall between ‘slightly cool’ and ‘slightly warm’. Statistical analysis shows that these values are significantly different (p < 0.01). The average comfort acceptability vote is about 0.2. There is no significant difference between these cases on comfort acceptability. The overall AMV 49 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY values show that all the cases’ room conditions are within the comfort zone and Case 6’s test condition is closest to ‘neutral’. Table 3.8 Average clo value, overall AMV and average comfort acceptability (Group 2 DV) Case 5 Case 6 Case 7 Ts (°C) 16.8 ± 0.09 18.7 ± 0.08 20.4 ± 0.32 Tr (°C) 22.2 ± 0.03 24.2 ± 0.13 26.2 ± 0.18 RH (%) 65.0± 0.59 65.5± 0.70 64.7± 2.91 Average Clo Value 0.56 ± 0.16 0.47 ± 0.13 0.50 ± 0.20 Overall AMV -0.43 ± 0.70 -0.19 ± 0.98 0.79 ± 0.88 Acceptability 0.19 ± 0.38 0.25 ± 0.33 0.14 ± 0.40 Unacceptability (%) 16.67 16.67 33.33 Table 3.8 shows that Cases 5 and 6 comply with the “less than 20%” unacceptability criterion while Case 7 reaches 33.33%. Case 7 has higher room air temperature of 26.2°C and this explains for the high percentage of unacceptability. These results suggest that tropical subjects would prefer cooler room condition than the warmer condition. This finding is consistent with Kwok’s (1998). However, 22°C is slightly lower than the temperature range recommended by ISO Standard 7730, ASHRAE Standard 55-1992 (revised) (summer comfort zone), SS-CP13 and guidelines for good indoor air quality in office premises. Out of all these unacceptability votes, 54% are from the subjects seated at the workstation closest to the supply unit and the rest from the workstation furthest away from the supply unit. The reason for this is the same as those mentioned in Section 3.2.2.1.1. 50 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Mean Thermal Sensation Votes 3.2.2.2.2 1.00 0.80 0.60 0.40 0.20 0.00 -0.20 -0.40 -0.60 -0.80 Effect of vertical temperature gradient on local thermal comfort Case5 Case6 Feet Calf Thigh Waist Body Arms Case7 Figure 3.22 Body parts’ thermal sensation (Group 2 DV) Figure 3.22 shows that for Cases 5 and 6, thermal sensation votes for feet and arms are lower than the other body parts but these differences are not statistically significant. This finding is identical to that of Group 1 and the reason for this is the same. Case 7 does not have this trend because its supply and room air temperatures are quite high. 3.2.2.2.3 Effect of distance from supply unit on thermal comfort Table 3.9 Mean thermal sensation and comfort acceptability between workstations nearest and furthest from the supply unit (Group 2 DV) Workstation Furthest Workstation Closest to Away From Supply unit Supply unit Case 5 0.08 ± 0.62 -0.12 ± 0.66 Case 6 0.65 ± 1.07 -0.45 ± 1.13 Case 7 1.50 ± 0.66 1.43 ± 0.93 Case 5 0.33 ± 0.36 0.14 ± 0.52 Comfort Acceptability Case 6 0.12 ± 0.28 0.10 ± 0.34 Case 7 -0.04 ± 0.36 -0.02 ± 0.21 Overall AMV Table 3.9 shows the comparison of the results between the subjects seated at the 51 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY workstations closest to the supply unit and those furthest away from the supply unit. It is observed that the subjects seated at the workstations closest to the supply unit generally have cooler overall thermal sensation and lower comfort acceptability than those seated furthest away from the supply unit. This result is similar to Group 1’s result. However, when studying thermal sensation for body parts, there are significant differences between these two workstations in Case 5 for feet (p = 0.04) and calf (p = 0.03). This may be due to the lower temperature at the region close to the supply unit. 3.2.2.3 Comparison of DV with MV (Group 3) 3.2.2.3.1 AMV and comfort acceptability Table 3.10 shows that DV cases generally have cooler thermal sensation than MV cases given the same room air temperature and relative humidity except for Cases D4 and M4. The cool sensation is due to the cooler air directly supplied to the occupied zone and the temperature gradient. This is consistent with the findings of Akimoto et al (1999). It is thus possible to save energy with the DV system by supplying less air to the occupied zone, increasing the room air temperature slightly while maintaining the same thermal sensation of the occupants as compared to the mixing ventilation. Case D1 has significantly cooler thermal sensation for overall body (p = 0.047), calf (p = 0.041), thigh (p = 0.027), waist (p = 0.029) and arms (p = 0.001) as compared to Case M1. This is due to the larger than 3 ºC/m temperature gradient in Case D1. Figure 3.23 shows that all cases have fulfilled the ‘less than 20%’ unacceptability criterion except for Cases D4 and M4. This result shows that room air temperature of 26.2°C may be too warm for the subjects. The result shows that DV’s thermal comfort performance may be better than MV at low temperatures of 22.2°C and 23.2°C and 52 15.7 ± 0.02 22.0 ± 0 65.9 ± 3.61 16.8 ± 0.09 22.2 ± 0.03 65.0 ± 0.59 Ts (°C) Tr (°C) RH (%) 0.60 ± 0.16 0.00 ± 0.87 -0.09 ± 1.03 0.08 ± 0.89 0.09 ± 0.83 0.03 ± 0.83 -0.03 ± 0.84 -0.13 ± 0.85 0.54 ± 0.15 -0.39 ± 0.58 -0.13 ± 0.82 -0.33 ± 0.65 -0.38 ± 0.73 -0.36 ± 0.71 -0.42 ± 0.60 -0.74 ± 0.57 Overall AMV Feet AMV Calf AMV Thigh AMV Waist AMV Body AMV Arms AMV Average Clo Value Thermal Sensation Case M1 Case D1 -0.66 ± 0.87 -0.31 ± 0.62 -0.31 ± 0.73 -0.43 ± 0.76 -0.36 ± 0.78 -0.51 ± 0.95 -0.44 ± 0.75 0.53 ± 0.19 67.8 ± 1.26 23.3 ± 0.06 18.5 ± 0.10 Case D2 Table 3.10 Average clo value and AMV (Group 3) -0.32 ± 1.02 -0.18 ± 1.03 -0.11 ± 1.04 -0.12 ± 1.01 -0.08 ± 1.02 -0.23 ± 1.03 -0.17 ± 1.02 0.47 ± 0.14 70.4 ± 0.49 22.9 ± 0 17.8 ± 0.31 Case M2 -0.10 ± 0.89 0.16 ± 0.82 0.18 ± 0.91 0.13 ± 0.91 0.05 ± 0.91 -0.08 ± 0.98 0.05 ± 0.87 0.48 ± 0.13 65.0 ± 0.20 24.2 ± 0.13 18.7 ± 0.08 Case D3 -0.04 ± 0.92 0.17 ± 0.86 0.15 ± 0.86 0.13 ± 0.90 0.16 ± 0.89 0.08 ± 0.95 0.09 ± 0.87 0.53 ± 0.18 66.5 ± 5.09 24.0 ± 0.07 17.3 ± 0.05 CaseM3 0.64 ± 0.80 0.72 ± 0.83 0.78 ± 0.83 0.76 ± 0.88 0.73 ± 0.82 0.73 ± 0.83 0.73 ± 0.84 0.51 ± 0.19 64.7 ± 2.91 26.2 ± 0.18 20.4 ± 0.32 Case D4 0.48 ± 1.11 0.58 ± 1.02 0.58 ± 0.97 0.55 ± 1.02 0.58 ± 0.97 0.55 ± 1.08 0.57 ± 1.03 0.44 ± 0.10 64.7 ± 1.06 26.1 ± 0 17.9 ± 0.11 Case M4 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY poorer than MV at a high temperature of 26.2°C. 53 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY Case M4 has lower percentage of unacceptability than Case D4 even though their room air temperature and relative humidity are the same. The cooler sensation for Case M4 is due to the relatively higher velocity. This suggests that at a warm condition like in Case 4, a relatively higher velocity is preferred. 33.33 % of Unacceptable 40 25.00 30 16.67 16.67 16.67 16.67 20 8.33 8.33 10 0 Case Case Case Case Case Case Case Case D1 M1 D2 M2 D3 M3 D4 M4 Figure 3.23 Thermal comfort unacceptability (Group 3) TS Votes 3.2.2.3.2 Neutral temperature 1.0 0.5 0.0 -0.5 21 -1.0 22 23 24 25 26 27 Room Temperature MV DV Linear (MV) Linear (DV) Figure 3.24 Neutral temperatures for DV and MV system (Group 3) Linear regression is used to explore the relationship between the room air temperature and the AMV, and to determine the neutral temperature of DV and MV cases. Figure 3.24 shows that the DV’s neutral temperature is about 24°C while MV’s is about 23°C. Figure 3.24 also shows that DV system produces cooler thermal sensations than MV 54 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY system at lower temperature range of between 22 and 24°C. 3.2.2.3.3 Local thermal comfort 1.0 0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8 -1.0 Feet Calf Thigh Waist Body Arms Case D1 Case M1 Case D2 Case M2 Case D3 Case M3 Case D4 Case M4 Figure 3.25 Body parts’ thermal sensation for DV and MV cases (Group 3) Figure 3.25 shows that MV cases have cooler thermal sensation for feet and arms although such cooler sensation is not significantly different from the sensation of the other body parts. Hence, this phenomenon is not distinctive to DV cases but is attributable to subjects’ clothing preference of short-sleeve shirts and open-toe shoes. However, the existence of temperature gradient for displacement ventilation may result in subjects being exposed to a lower temperature as compared to mixing ventilation. Therefore, the phenomenon is more significant in DV cases than in MV cases. 3.2.2.4 Application of ISO 7730 It is stated in ISO 7730 that the temperature difference between 0.1m and 1.1m should not exceed 3°C/m while in ASHRAE 55-1992 (revised), the temperature difference between 0.1m and 1.7m should not exceed 3°C/m. However, for Cases 1, 2 and 7 the temperature differences exceeded those criteria and the subjects still felt that it was acceptable. This suggests that tropical subjects could tolerate larger temperature 55 CCCHHHAAAPPPTTTEEERRR 333::: PPPRRREEELLLIIIMMMIIINNNAAARRRYYY SSSTTTUUUDDDYYY difference. This may lead to energy saving by supplying less air at a lower temperature. 56 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY CHAPTER 4 CONFIRMATION STUDY 4.1. Methodology 4.1.1 Research design There are two objectives at this stage of study: firstly, to examine the results of the preliminary study; and secondly, to explore the energy-saving potentials of displacement ventilation system and to develop a preliminary design guide based on the experiment data. In the first objective, the basic methodology of this stage of study is the same as that of the preliminary study. In the second objective, there are some changes on the methodology and they will be presented in details at the later part of this chapter. A total of 5 displacement ventilation cases and 1 mixing ventilation case were formulated in this study. These cases are assigned to different groups depending on the supply air temperature, room relative humidity and ventilation mode. 4.1.1.1 Group 1 In this group, room air temperature was kept constant with variation in the supply air temperature. This led to a variation in the supply air flow rate as the heat gain in the room remained unchanged. The room air humidity (both absolute and relative humidity) were kept constant by fixing the off-coil temperature. The ratio of outdoor air flow rate to the total supply air flow rate was kept constant at about 37%. Table 4.1 shows the various conditions for Group 1. 57 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY This group of cases was formulated based on the following hypotheses: i. Subjects would have different thermal sensation when they are exposed to different vertical temperature differences; ii. Subjects would have different draft sensation when they are exposed to different vertical temperature differences and different supply air flow rates. Table 4.1 Test conditions for Group 1. Case Ts (°C) Tr (°C) RH (%) 1 16.2 23.4 54 2 18.1 23.5 56 3 21.0 23.5 52 4.1.1.2 Group 2 In this group both room air temperature and supply air temperature were kept constant while room air humidity (both absolute and relative humidity) varied. The ratio of outdoor air flow rate to the total supply air flow rate was about 37%. Table 4.2 shows the various conditions for Group 2. Table 4.2 Test conditions for Group 2. Case Ts (°C) Tr (°C) RH (%) 3 21 23.5 52 4 21 23.5 76 This group of cases was formulated based on the following hypothesis: Subjects would have different thermal sensation when they are exposed to different humidity. 58 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 4.1.1.3 Group 3 The objective of the test for this group is to compare displacement ventilation with conventional mixing ventilation. The ratios of outdoor air flow rate to the total supply air flow rate for Case 5 and 6 were 33% and 27% respectively. Table 4.3 shows the various conditions for Group 3. This group of cases was formulated based on the following hypothesis: Subjects would have different thermal sensation with DV and MV even though the room air temperature and relative humidity are similar. Table 4.3 Test conditions for Group 3. Case Ts (°C) Tr (°C) RH (%) Mode 5 17.5 23.4 62 DV 6 17.5 23.4 62 MV 4.1.2 Methods of data collection 4.1.2.1 Subjective assessment 4.1.2.1.1 Subjects Thirty six (10 male and 26 female) college-age students participated in the series of experiments as subjects. The subjects were recruited based on the following criteria: native Singaporean, familiarity with a PC, impartiality to the chamber in which the study was carried out, and absence of chronic diseases, asthma, allergy and hey-fever etc. The statistical summary of these subjects is shown in Appendix A2. These subjects were randomly exposed to different test conditions on different days and were kept 59 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY “blind” to the test conditions to avoid biased results. Requirements for subjects were the same as those for the preliminary study (see Section 3.1.2.1.1). 4.1.2.1.2 Subjective assessment protocol Each experiment lasted for 2 hours and proceeded as follows: i. The subjects arrived at the chamber 15 minutes prior to the commencement of the experiment. They were seated in the control room and briefed on the procedure. ii. Subjects entered the chamber and the experiment commenced. At the 0th and every twentieth minute, the subjects would complete a set of the questionnaire on their whole body thermal sensation, different body parts thermal sensation, thermal comfort acceptability, and air movement detection and acceptability. iii. At 120th minute, the subjects would complete another set of the questionnaire which enquired about the type of clothing the subjects were wearing. 4.1.2.1.3 Questionnaire The questionnaire used in this stage was similar to that used in the preliminary stage. A set of the questionnaire is shown in Appendix B2. 4.1.2.2 Objective measurement protocol 4.1.2.2.1 Objective parameters Objective measurement started 5~15 minutes prior to the experiments and ended 5~15 minutes after the experiments. During each experiment the following parameters were measured: i. Temperature: supply air temperature (Ts), return air temperature (Te), room air 60 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY temperature (Tr), wall surface temperature (Tw), ceiling surface temperature (Tc) and floor surface temperature (Tf); ii. Room air relative humidity and dew point temperature; iii. Air velocity; iv. Supply and return air flow rates; and v. CO2 concentration. 4.1.2.2.2 Instrumentation 4.1.2.2.2.1 Thermal chamber The second stage experiments were carried out in the same chamber as that in the preliminary study. However, at this stage of experiments, a second supply unit was installed at the other end of the chamber. The aim of this is to study the effect of draft with DV system. Figure 4.1 shows the details. E1 E3 New DV unit T1 MV diffuser T2 DV unit E2 Figure 4.1 Chamber layout Note: WS denotes workstation number and E denotes extract grille number. 61 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY In the DV mode, the air is supplied from two floor-standing, low velocity, semicircular supply units. In the MV mode, the air is supplied from two square ceiling diffusers, T1 and T2. In both modes, the air is extracted from two ceiling grilles, E2 and E3. 4.1.2.2.2.2 Instruments The instruments used in this study were the same as those used in the preliminary study. 4.1.2.2.3 Measuring locations Figure 4.2 shows the measuring locations. A detailed description of each measuring location is stated as follows: i. Supply and return air temperatures were measured by thermocouple wires placed inside the supply units and return grilles respectively. ii. Room air temperature measuring locations: Points 1 to 5: Thermocouple wires were placed at six heights namely 0.1m, 0.3m 0.6m, 1.1m, 1.7m and 2.5m. iii. Wall surface temperature: points marked as 8. Thermocouple wires were also placed at these points at three heights namely 0.1m, 1.3m and 2.5m. iv. Ceiling and floor surface temperature: points marked as 7 (in the same row of points 1 to 5). v. Room air relative humidity: Points 3, 6 and 8: HOBO meters were placed at five heights, i.e. 0.1m, 0.6m, 1.1m, 1.7m and 2.5m. 62 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY vi. Dew point temperature: Point 7: Dew-point hygrometer probes were placed at heights of 0.1m, 0.6m, 1.1m, 1.7m and 2.5m. vii. Air velocity: Points 6, 7 and 8: Dantec sensors were placed at five heights: 0.1m, 0.3m, 0.6m, 0.9m and 1.1m. viii. Supply and return air flow rates: Measured at the PRA damper directly above the supply units or return grilles in the ductwork. ix. CO2 concentration: Point 7: At four heights namely 0.1m, 0.6m, 1.1m and 1.7m; and Points 6 and 8, at height of 1.1m. CO2 concentration was also measured in supply and return air duct, and outside the experimental chamber, which were not marked on Figure 4.2. Figure 4.2 Plan view of measuring points. 63 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 4.1.3 Data collection and analysis 4.1.3.1 Objective data Objective data were measured and logged simultaneously during the experiment. The logging intervals of various parameters are stated as follow: i. Temperature: 30 seconds. ii. Room air relative humidity: 2 minutes. iii. Room air dew point temperature: 30 seconds. iv. Air velocity: 30 seconds. v. CO2 concentration: 16 minutes. 4.1.3.2 Subjective data Subjective data were acquired from various scales on the questionnaires. They were processed using the principles employed in the preliminary study as stated in Section 3.1.3.2. 4.1.4 Method of data analysis The data analysis method was the same as that of the preliminary study which was stated in Section 3.1.4. 64 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 4.2 Results and discussion 4.2.1 Gradients 4.2.1.1 Temperature gradient 4.2.1.1.1 Effect of different supply air temperature and flow rate with the same room air temperature and humidity (Group 1) Figure 4.3 shows the temperature gradients of Group 1. It can be seen that the gradients of this group are similar to those of Group 1 in the preliminary study: the gradient can be divided into two parts: one with steep gradient (floor level to 1.7m) and the other with a gentler gradient (1.7m to ceiling level). With the supply air temperature rising from Case 1 (16.2 ºC) to Case 3 (21 ºC), the air temperature near the floor rises from around 19 ºC to 22.5 ºC. The temperature difference between cases decreases until it reaches 1.1m. 25.5 25.0 24.5 24.0 23.5 23.0 22.5 22.0 21.5 21.0 20.5 20.0 19.5 19.0 18.5 3 2.5 2 1.5 1 0.5 0 18.0 Height (m) Temperature gradient Temp (ºC) Case-1, L-1 Case-1, L-2 Case-1, L-3 Case-1, L-4 Case-1, L-5 Case-2, L-1 Case-2, L-2 Case-2, L-3 Case-2, L-4 Case-2, L-5 Case-3, L-1 Case-3, L-2 Case-3, L-3 Case-3, L-4 Case-3, L-5 Figure 4.3 Temperature gradients (Group 1). A more detailed observation shows that there is a layer of air in the lower zone where 65 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY the temperature gradient is not as significant as at the other heights, i.e. from 0.1m to 0.3m and from 0.6m to 1.7m. This layer starts from about 0.3m and ends at about 0.6m. This phenomenon has never been mentioned before in the literature. One possible reason could be: Subjects, PCs and lightings are the heat sources inside the chamber. PCs are placed on the table at about 0.9m height. Subjects are seated throughout the experiment; their bodies are at the height between 0.6m (waist) and 1.1m (head). Between 0.6m and 1.1m, subjects and PCs generate a large amount of heat and this heat could raise the room air temperature. Lightings are also heat sources. However, the increase in air temperature is caused indirectly by radiation—the floor absorbs the radiation and warms up; then the warmed-up floor transfers the heat to the air through conduction and convection. This explains for the rapid rise in the air temperature from floor level to about 0.3m. As for height between 0.3m and 0.6m, there is no significant heat gain. Subjects’ legs would emit some heat. However, this is quite small compared to those from floor, subjects’ bodies and PCs. 4.2.1.1.2 Comparison between DV and MV (Group 3) Figure 4.4 shows the temperature profiles of Group 3. It is observed that significant gradient exists with DV system (Case 5). In MV system (Case 6), the temperature values are almost the same at different measuring points below the height of 1.7m. However, at 2.5m, the differences are quite large. This depends on the location of the measuring points. Points 1 and 5, which are close to the return grilles, will have higher temperature as compared to Point 3, which is close to the supply diffusers, will have lower temperature due to the entrainment of the cooler air. 66 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 25.5 25.0 24.5 24.0 23.5 23.0 22.5 22.0 21.5 21.0 20.5 20.0 19.5 3 2.5 2 1.5 1 0.5 0 19.0 Height (m) Temperature gradient Temp (ºC) Case-5, L-1 Case-5, L-2 Case-5, L-3 Case-5, L-4 Case-5, L-5 Case-6, L-1 Case-6, L-2 Case-6, L-3 Case-6, L-4 Case-6, L-5 Figure 4.4 Temperature profiles (Group 3). 4.2.1.2 Humidity gradient 4.2.1.2.1 Effect of different supply humidity ratio with the same room air temperature (Group 2) Figure 4.5 shows the humidity gradients of Group 2. As the off-coil temperatures for these two cases were controlled at different levels (Case 3, 12.7 ºC; Case 4, 20 ºC), these two humidity gradients are at different levels. The humidity gradient profiles show that the humidity ratio increases while the relative humidity decreases with height. This is due to the different increments of humidity ratio and temperature. Figure 4.5a shows that the humidity ratio increases from 9.5 g/kg (Case 3) and 13.5 g/kg (Case 4), to 9.8 g/kg and 13.7 g/kg, respectively. However, the temperature increases from 22.3 ºC (Case 3) and 22.5 ºC (Case 4), to 24.7 ºC (both cases). This implies that the increase in temperature is larger than that of the humidity ratio. This would cause the RH values to decrease, as the RH is a function of the temperature and humidity ratio. 67 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY On the other hand, it is observed that the humidity difference between lower points and higher points is very small (about 0.2-0.3 g/kg dry air). The humidity load comes solely from the human with no infiltration since the thermal chamber is a room within another room. Heights (m) Humidity ratio at different heights 3 2.5 2 1.5 1 0.5 0 8 9 10 11 12 13 14 15 75 80 Humidity ratio (g/Kg) Case-3 Case-4 Heights (m) Relative humidity at different heights 3 2.5 2 1.5 1 0.5 0 45 50 55 60 65 70 RH (%) Case-3 Case-4 Figure 4.5 Humidity gradients (Group 2). (a) Humidity ratio gradients (upper); (b) Relative humidity gradients (lower). 4.2.1.2.2 Comparison between DV and MV (Group 3) Figure 4.6 shows the humidity gradients of Group 3. As shown in the figure, the humidity ratio gradient also exists with MV system (Case 6) and this gradient shares a similar profile to that of DV case (Case 5). The reason why humidity ratio gradient 68 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY also exists with MV system is very simple. Human subjects generate water vapour through perspiration and breathing. Water vapour and humid air are lighter than dry air at the same temperature (one mole of gas occupies the same volume at a given temperature and pressure; the molecular weight of water (18 g/mole) is smaller than that of the air (about 29 g/mole)). With density difference, the water vapour would “move” to the upper part of the space. This will lead to higher humidity ratio at higher levels in an imperfect mixing environment. Heights (m) Humidity ratio at different heights 3 2.5 2 1.5 1 0.5 0 8 9 10 11 12 13 14 15 75 80 Humidity ratio (g/Kg) Case-5 Case-6 Heights (m) Relative humidity at different heights 3 2.5 2 1.5 1 0.5 0 45 50 55 60 65 70 RH (%) Case-5 Case-6 Figure 4.6 Humidity gradients (Group 3). (a) Humidity ratio gradients (upper); (b) Relative humidity gradients (lower). It is also observed that relative humidity gradient exists with DV system. However, with MV system, there is no such significant gradient, although the humidity ratio 69 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY gradient exists. This is due to the variation in the temperature. With displacement ventilation, the temperature varies from bottom to top of the room, as shown in Figure 4.4. Such variation would result in the observed relative humidity gradient, despite the increasing humidity ratio. 4.2.1.3 Concentration gradient 4.2.1.3.1 Effect of different supply air temperature and flow rate with the same room air temperature and humidity (Group 1) Figure 4.7 shows the CO2 concentration gradients of Group 1. These gradients have similar profiles, with lower concentration at the occupied zone (below 1.1m) and higher concentration at the unoccupied zone (above 1.7m). As shown in the figure, there exists a layer where the concentration increases drastically with height, i.e. from 1.1m to 1.7m. This layer is where the mixing zone (unoccupied zone, with polluted air) meets with the displacement zone (occupied zone, with rather clean air). It shows that the concentration of CO2 increases drastically beyond this layer. Height (m) CO2 concentration gradients 3 2.5 2 1.5 1 0.5 0 450 500 550 600 650 700 Concentration (ppm) Case-1 Case-2 Case-3 Figure 4.7 CO2 gradients (Group 1). In addition, it is observed that at the same height, these cases have different 70 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY concentration levels, with Case 1 having the highest and Case 3 the lowest. This is because these cases have different outdoor air flow rates, as shown in Appendix D1. Higher outdoor air flow rates will lead to a lower CO2 level in a room served by highly recirculated ACMV system. 4.2.1.3.2 Comparison between DV and MV (Group 3) Figure 4.8 shows the CO2 concentration profiles of Group 3. In the DV system, the concentration of CO2 increases with height. There is a significant increase in CO2 level beyond 1.1m height. In the MV system, there is no significant increase in the CO2 level after the 1.1m height. This is due to the mixing nature of the system. Since these CO2 concentration profiles have the same ventilation rate, this proves the fact that displacement ventilation does have higher ventilation efficiency and can provide better indoor air quality than mixing ventilation. CO2 concentration gradients 3 Height (m) 2.5 2 1.5 1 0.5 0 450 500 550 600 650 700 750 Concentration (ppm) Case-5 (DV) Case-6 (MV) Figure 4.8 CO2 gradients (Group 3). The following ventilation effectiveness values obtained from the experimental results for DV system further prove that the DV system has higher ventilation efficiency: ε1.1m, WS-1=2.41; ε 1.1m, WS-2=3.51; ε 1.1m, WS-3=1.70. 71 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY where ε = Ce − C s C1.1 m − C s where C is the contaminant concentration, subscripts e, s, and 1.1m refer to point at exhaust (measured at exhaust grille 2), point at supply (measured in supply duct), and point at the head level (1.1m) of a sedentary person respectively, and WS denotes workstation. Obviously, all the ventilation effectiveness values are larger than 1, a value for perfect mixing ventilation. Therefore, it can be concluded that displacement ventilation system does have higher ventilation efficiency and can provide better indoor air quality than mixing ventilation system. 4.2.2 Thermal comfort 4.2.2.1 Effect of different supply air temperature & flow rate on thermal comfort (Group 1) 4.2.2.1.1 Overall Actual Mean Vote (AMV) and comfort acceptability Table 4.4 shows room conditions, average clo value, AMV and comfort (un)acceptability values of Group 1. It is observed that the experiments are well controlled since the standard deviation values of supply air temperature, room air temperature and RH are very small. Table 4.4 shows that the Overall AMV values for all cases fall on the cooler side of the scale. Subjects feel between ‘cool’ and ‘slightly cool’ for Cases 1 and 2 and ‘slightly cool’ for Case 3. Statistical analysis shows that the difference for Overall AMV among 72 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY these three cases is significant (p=0.011). This agrees with the result of the preliminary study. The average comfort acceptability votes are similar and on the cool side. Statistical analysis shows that the difference for comfort acceptability among these cases is not significant. This also agrees with the result of the preliminary study. The Overall AMV values show that at 23.5°C only Case 3’s room condition is in the comfort region. The significant difference for Overall AMV shows that the supply air temperature (or vertical temperature gradient) has significant influence on subjects’ thermal sensation. However, the influence of supply air temperature (or vertical temperature gradient) on subjects’ comfort acceptability is not significant. Table 4.4 Average clo value, overall AMV and average comfort acceptability (Group 1) Case 1 Case 2 Case 3 Tr (°C) 23.4 ± 0.1 23.5 ± 0.0 23.5 ± 0.1 Ts (°C) 16.2 ± 0.1 18.1 ± 0.2 21.0 ± 0.1 RH (%) 54.3 ± 0.7 55.8 ± 0.7 52.4 ± 0.5 Average Clo Value 0.50 ± 0.13 0.52 ± 0.15 0.54 ± 0.16 Overall AMV -1.31±1.09 -1.11±1.02 -0.83±0.89 Acceptability 0.10±0.34 0.19±0.6 0.21±0.40 Unacceptability (%) 36.1 27.8 33.3 Table 4.4 shows that none of these cases fulfilled the “less than 20%” unacceptability criterion (ASHRAE Standard 55-1992, ISO 7730) and Case 1 has the highest number of unacceptability votes. For Case 1, the high unacceptability votes could be due to the low supply air temperature with large temperature gradient of between 4.0 and 5.5 °C between 0.1m and 1.1m height, as shown in Appendix D2, depending on the distance from the supply unit. 73 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY The above data show that at a room air temperature of 23.5°C and with clo value of about 0.5 (value of typical clothes in tropical area), none of these supply air temperatures are acceptable, given the thermal comfort criteria (ASHRAE Standard 55-1992, ISO 7730). This suggests that the room air temperature of 23.5°C may be too low at the 0.5clo level. Therefore, future study at higher room air temperature of about 25 °C may be recommended. Among these unacceptability votes, 69% are from the subjects seated at the workstation near to the supply unit and 31% at the workstation in the middle of the room. The higher unacceptability votes at the workstation near the supply unit may be due to lower temperature and/or higher air velocity, as compared to the workstation in the middle of the room (see Appendix D2). This suggests that, to obtain the least number of unacceptability votes without sacrificing the region near the supply unit, specially designed supply units which can direct the supply air parallel to the end wall outside the occupied zone should be used. 4.2.2.1.2 Local thermal comfort Mean thermal sensation vote 0.00 -0.20 Feet Calf Thigh Waist Body Arms -0.40 -0.60 Case-1 -0.80 Case-2 -1.00 Case-3 -1.20 -1.40 -1.60 Figure 4.9 Body parts’ thermal sensation (Group 1 DV) 74 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY Figure 4.9 shows that generally for all cases, thermal sensation votes for feet and arms are lower than those of the other body parts. This confirms the results of the preliminary study. Furthermore, statistical analysis shows that significant difference exists between these body parts in terms of their thermal sensation votes for Case 1 (p=0.034). This suggests that given their clothing habits (short sleeves shirt, open toe shoes), 23.5 °C room air temperature and 16 °C supply air temperature is low and unacceptable to local subjects. Statistical analysis also shows that significant differences exist among these three cases for thermal sensation votes on arms (p=0.0014), calf (p=0.0291) and body (p=0.0193). The results show that Case 1 has the lowest thermal sensation votes for these body parts. This implies that 16 °C supply air temperature is too low for these subjects. 4.2.2.1.3 i. Draft Overview There are 14%, 14% and 11% subjects who had draft sensation for Cases 1, 2 and 3 respectively. This is well within the criteria (15%) stated in the standard (ASHRAE Standard 55-1992, ISO 7730). This shows that these supply conditions (both temperature and flow rate) are acceptable to the subjects. Among the draft sensation votes, 71% are cast by subjects seated at the workstations near the supply unit and 29% at the workstations in the middle of the room. This shows that the area near the supply unit has higher risk of discomfort due to the lower temperature and/or higher air movement as shown in Appendix D2. 75 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY ii. Group study Figure 4.10 shows the draft risk and the draft sensation at different heights for different groups. Here, draft is studied according to the subjects’ distances from the supply unit. Those seated at workstations 1, 3, 4 and 6 are classified as “near unit” group and those seated at workstations 2 and 5 are classified as “middle” group. Furthermore, draft is studied according to the data source. “Draft risk” is computed based on measured data while “draft sensation” is determined based on questionnaire survey. Draft risk at different heights Draft risk (%) 30 Case-1, near unit 25 Case-2, near unit 20 Case-3, near unit 15 Case-1, middle 10 Case-2, middle 5 Case-3, middle 0 0.0 0.3 0.6 0.9 1.2 Height (m) Draft sensation at different heights Draft sensation (%) 30 Case-1, near unit 25 Case-2, near unit 20 Case-3, near unit 15 Case-1, middle 10 Case-2, middle 5 Case-3, middle 0 0.0 0.3 0.6 0.9 1.2 Height (m) Figure 4.10 Draft at different heights (Group 1). (a) draft risk (upper); (b) draft sensation (lower). Figures 4.10 (a) and (b) show that the draft risks are higher than relative draft sensations for both groups at 0.1m height. However, at the other heights, i.e. 0.3m, 76 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 0.6m and 1.1m, the draft risks are similar to the draft sensations for both groups. At 0.9m height, the draft risks are lower than the draft sensations. This shows that the draft model has over-predicted the draft problem at the feet level and underestimated the problem at the arms level. Since the differences between the predicted draft risks and the surveyed draft sensations are small at arms level, the draft model can still be used when designing displacement ventilation systems. 4.2.2.1.4 Effect of distance from supply unit on thermal comfort Table 4.5 shows the comparison results between the subjects seated at the workstations close to the supply unit and those seated in the middle of the room. It is observed that subjects seated close to the supply unit (distance 1.0m) have lower Overall AMV and lower comfort acceptability than those seated in the middle of the room except Case 2. However, these differences are not statistically significant. This result is consistent with that of the preliminary study and suggests that occupants can be seated close to the supply unit (distance 1.0m) without sacrificing the usable floor area. Table 4.5 Mean thermal sensation and comfort acceptability between workstations closer and further from the supply unit (Group 1 DV) Overall AMV Workstation Further Workstation Nearer Away From Supply unit to Supply unit Case 1 -1.21 -1.36 Case 2 -1.28 -1.03 Case 3 -0.70 -0.89 Case 1 0.12 0.09 Case 2 0.06 0.25 Case 3 0.23 0.20 Comfort Acceptability 77 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 4.2.2.2 Effect of different humidity levels on thermal comfort (Group 2) 4.2.2.2.1 Overall Actual Mean Vote (AMV) and comfort acceptability Table 4.6 shows the room’s conditions, average clo value, AMV and comfort (un)acceptability values of Group 2. The table shows that subjects feel less cool with higher levels of relative humidity. Statistical analysis shows that the difference is close to significant (p=0.066). This suggests that the RH have some influence on subjects’ thermal sensation. As the p value is very close to the significant level (0.05), further research on this influence is therefore required. The mean acceptability values for these two cases are close. Statistical analysis shows that the difference between these two cases is not significant. Table 4.6 Average clo value, overall AMV and average comfort acceptability (Group 2) Case 3 Case 4 Tr (°C) 23.5 ± 0.1 23.5 ± 0.1 Ts (°C) 21.0 ± 0.1 21.1 ± 0.1 RH (%) 52.4 ± 0.5 76.1 ± 1.5 Average Clo Value 0.54 ± 0.16 0.54 ± 0.17 Overall AMV -0.83±0.89 -0.56±0.85 Acceptability 0.21±0.40 0.20±0.29 Unacceptability (%) 33.3 19.4 Table 4.6 also shows that Case 4 fulfilled the “less than 20%” unacceptability criterion. This shows that at a low temperature of 23.5°C, lower RH level (52%) is not 78 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY welcomed and higher RH level (76%) is more appreciated. It is important to note that the room condition of Case 4, 23.5 °C and 76% RH, exceeded the summer comfort zone proposed by the ASHRAE standard (55-1992 revised) and the local standard (CP13, 1999). The subjects felt comfortable in a relatively “humid” environment. Hence, the RH level can be allowed to go beyond 70%. This will lead to the saving of the energy used to reduce moisture in the supply air and control the room air RH within the 70% limit. This study has only considered thermal sensation. Additional study is needed to study the RH limit with response to condensation and mould growth. 4.2.2.2.2 Local thermal comfort Mean thermal sensation vote 0.00 -0.10 Feet Calf Thigh Waist Body Arms -0.20 -0.30 -0.40 Case-3 -0.50 Case-4 -0.60 -0.70 -0.80 -0.90 Figure 4.11 Body parts’ thermal sensation (Group 2 DV) Figure 4.11 shows that for both cases, thermal sensation votes for feet and arms are lower than those of other body parts. This result is similar to that of Group 1. The chart shows that Case 3 has lower thermal sensation vote than Case 4 for every part of the body. Since the only difference between these two cases is the RH value, it can be concluded that such lower thermal sensation is due to the RH. 79 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY Statistical analysis shows that significant differences exist between these two cases for thermal sensation votes on feet (p=0.0045), calf (p=0.0398), thigh (p=0.0108) and waist (p=0.0426). This reinforces the above conclusion and shows that room RH can have significant influence on subjects’ individual body parts, though it may not have significant influence on subjects’ whole body thermal sensation and comfort acceptability. The results show that 52% RH and 23.5 °C room air temperature may be low to slightly clothed subjects. 4.2.2.3 Comparison of DV with MV (Group 3) 4.2.2.3.1 Overall AMV and comfort acceptability Table 4.7 shows room conditions, average clo value, AMV and comfort (un)acceptability values of Group 3. The Overall AMV values for both cases are similar with subjects perceiving “slightly cool” sensation. The Acceptability vote for Case 6 (MV) is higher than that for Case 5 (DV) and statistical analysis shows the difference is significant (p=0.05). The Unacceptability vote for Case 6 is lower than that for Case 5 and within the ‘less than 20%’ unacceptability criterion. The higher Unacceptability vote for Case 5 is due to the existence of the temperature gradient as shown in Appendix D2. The subjects were exposed to a temperature one or two degrees lower at 0.1m as compared to 1.1m height. Hence, the subjects were exposed to a cooler environment with DV system. This led to relatively lower overall AMV and Acceptability vote for the DV system as compared to the MV system. In order to obtain the same Overall AMV, Acceptability and Unacceptability, the supply air will need to be at a higher temperature. Table 4.7 shows that with clo value of about 0.5, room air temperature of 23.5 °C is on 80 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY the cooler side for both DV and MV systems. Hence, more detailed study of higher room air temperature is required. Table 4.7 Average clo value, overall AMV and average comfort acceptability (Group 3) Case 5 Case 6 Tr (°C) 23.4 ± 0.1 23.4 ± 0.1 Ts (°C) 17.5 ± 0.1 17.5 ± 0.1 RH (%) 62.5 ± 1.4 61.6 ± 0.7 Average Clo Value Overall AMV -0.90±1.01 -0.80±0.89 Acceptability 0.14±0.34 0.30±0.32 Unacceptability (%) 25.0 16.7 Local thermal comfort Mean thermal sensation vote 4.2.2.3.2 0.53 ± 0.11 0.52 ± 0.14 0.00 -0.20 Feet Calf Thigh Waist Body Arms -0.40 Case-5 Case-6 -0.60 -0.80 -1.00 -1.20 Figure 4.12 Body Parts’ Thermal Sensation (Group 3) Figure 4.12 shows that Case 6 (MV) also has cooler sensation for feet and arms. This result is consistent with that of the preliminary one. Furthermore, statistical analysis shows that significant difference exists among these body parts in terms of their thermal sensation votes for Case 5 (p=0.05). This suggests that given the subjects’ clothing habits (short sleeves shirt), and a supply air temperature of 17.5 °C with 23.5 81 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY °C room air temperature, it may be low and unacceptable to local subjects. It is to be noted that Case 5 (DV) has lower thermal sensation for all body parts as compared to Case 6 (MV). This is due to the temperature gradient as stated in the previous section. This explains for the subjects’ lower Overall AMV and Acceptability values, and higher Unacceptability value for Case 5. 4.2.2.3.3 i. Draft risk Overview Altogether, there are 17% and 19% of subjects who had draft sensation for Cases 5 and 6 respectively. These have just slightly exceeded the threshold stated in the standards (ASHRAE Standard 55-1992, ISO 7730). In Case 5, 67% of the draft sensation votes are cast by subjects seated at the workstations near the supply unit and 33% at the workstations in the middle of the room. This is consistent with the result of Group 1 and it shows that the area near the supply unit has higher risk of discomfort due to the lower temperature and/or higher air movement. In Case 6, 43% of the draft sensation votes are cast by subjects seated at the workstations near the supply unit and 57% at the workstations in the middle of the room. ii. Group study Figure 4.13 shows the draft risk and the draft sensation at different heights for different groups. Results show that, at 0.1m height, the draft risks are lower than the draft sensations for both cases for the “middle” group, in particular for Case 6 where the draft sensation reaches 25%. At 0.3m height, the draft risk for Case 6 “middle” group is also lower than the draft sensation, which still exceeds the 15% criteria. The same 82 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY case applies to 0.9m height, when the draft risks for both groups and cases are no more than 10%, the draft sensation for Case 6 “middle” group reaches 25%. The draft sensation for Case 6 “middle” group is always higher than the others. This is because at Case 6, air is supplied from two ceiling diffusers (see Figure 4.1). The right-hand side ceiling diffuser is just above the subjects seated in the “middle”. Hence, the supply air jet from this ceiling diffuser could reach the subjects’ exposed body parts and this could lead to high draft sensation by the subjects (see Appendix D2). Draft risk at different heights Draft Risk(%) 30 25 20 Case-5, middle Case-6, middle Case-5, near unit Case-6, near unit 15 10 5 0 0.0 0.3 0.6 0.9 1.2 Height(m) Draft sensation at diffrent heights Draft sensation (%) 30 25 Case-5, middle Case-6, middle Case-5, near unit Case-6, near unit 20 15 10 5 0 0.0 0.3 0.6 0.9 1.2 Height(m) Figure 4.13 Draft at different heights (Group 3). (a) draft risk (upper); (b) draft sensation (lower). The relatively low draft risk for DV could also be due to the fact that there are two supply units inside the chamber. With two supply units, the velocity around the subjects could be much lower than when there is only one unit in the preliminary study. 83 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY This suggests that in future design, if possible, more supply units should be considered. This will not only lower the velocity around the occupants, but also reduce the “adjacent zone” of each unit with low air temperature and high turbulence intensity. Draft sensation is based not only on the air velocity, but also on low air temperature and high turbulence intensity. Draft problem could be minimized by installing as many supply units as reasonably possible, Figure 4.13 shows that for the same operating conditions, DV system has lower draft sensation votes than MV system. Since this study is conducted in the thermal chamber with a ceiling height of 2.6m, these results may not be applicable to other places with higher ceilings. The reason is that: with high ceiling, there should be sufficient distance for the air supply jet to decrease to a reasonable velocity and temperature level at the occupant’s level. 84 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 4.2.3 Energy 4.2.3.1 Comparison between DV and MV 4.2.3.1.1 Experiment conditions, system and conditioning process Table 4.8 and Figure 4.14 show the experiment conditions for Group 3 and the system used in the experiments for both ventilation modes respectively. Table 4.8 Experiment conditions for Group 3. Supply Air Mode DV MV Return Air Room (1.1m) Total flow Average off-coil Total flow Average RH (%) rate(l/s) temp (ºC) temp (ºC) rate(l/s) temp (ºC) 190 227 17.5 17.5 14.9 14.7 128 157 24.9 23.7 59.3 63 Temp (ºC) 23.4 23.4 Fresh Air Humidity Total flow (%) rate(l/s) 62.5 61.6 66.0 64.0 Figure 4.14 System for both ventilation modes. (a) – DV; (b) - - MV. The entire conditioning process is marked on the psychrometric chart as shown in Figure 4.15 and the details are stated as follow: i. The return air (point A) and outside air (point B) mix at the first section of the 85 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY AHU. The mixed air reaches condition (point C). ii. The mixed air passes through the filter section and the cooling coil. As a result, both the temperature and humidity ratio are decreased. The cooled air reaches condition (point D). iii. The cooled air absorbs the heat as it passes through the fan and enters the room as condition (point E.) iv. The cooled air absorbs the heat and moisture and becomes warmer (point A). It then returns to the AHU via the return duct. 4.2.3.1.2 Energy consumption analysis The energy consumed by the entire air-conditioning system includes that by chiller, cooling tower, pump, fan and their accessories. As the energy consumed can be reflected from the cooling capacity needed for the system (normally, the higher the cooling capacity, the more energy consumption), the cooling capacity is discussed and used as an energy consumption index. It is computed based on the following equation: E = Q × (hC − hD ) where E stands for cooling capacity, in kW; Q is the mixed air flow rate, in kg/s; h stands for enthalpy of air, kJ/kg; C is mixed air condition before the cooling coil and D is the mixed air condition after the cooling coil. Results are as follows: EDV = 4.99 kW EMV = 5.23 kW Obviously, the cooling capacity is lower (5%) for DV system than for MV system. Hence, the energy consumption by DV system would be lower than that by MV system. 86 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY Figure 4.15 Conditioning process on psychrometric chart. Subscripts 1 and 2 stand for DV and MV respectively. B C2 C1 D1/D2 E1/E2 A2 A1 On the other hand, with the DV system, due to the temperature gradient, sedentary occupants are exposed to a condition actually one to two degrees lower than that at 1.1m height. Hence, occupants will receive cooler sensation with the DV system (see Section 4.2.2.3). In order to achieve the same thermal sensation as that with MV system, the room air temperature has to be raised by about 1.0ºC as discussed in Section 3.2.2.3.2 Neutral temperature. The increase in room air temperature will lead to the increase in the return air temperature. There are two conditions accompanied by the increase in the return air temperature. Firstly, if the supply air temperature remains unchanged, this will lead to the increase in the supply-return temperature difference. With larger supply-return temperature difference, the supply air flow rate could be reduced and therefore, a reduction in the fan energy consumption. Secondly, if the supply air temperature increases accordingly and the supply air flow rate remains unchanged, this will lead to the decrease in chilled water flow rate and in turn will lead 87 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY to the decrease in the energy consumption by the pump, chiller, and cooling tower. In both ways, the energy consumption by the DV system would be lower than that of the MV system. Furthermore, as discussed in Section 4.2.1.3.2, because of the concentration gradient, subjects are exposed to lower CO2 level with DV. To obtain the same CO2 level as that with MV, the outdoor air flow rate could be reduced. This would further lower the energy consumption by DV system in treating the outdoor air. 4.2.3.2 Effect of different supply air temperature on energy consumption 4.2.3.2.1 Experimental conditions, system and conditioning process Table 4.9 shows the experimental conditions for the two cases with different supply air temperatures. The system and conditioning process of these two cases are similar to those of Group 3, therefore they will not be presented here. Table 4.9 Experiment conditions. Mode DV 4.2.3.2.2 Supply Air Return Air Case No. Total flow Average off-coil Total flow Average RH (%) rate(l/s) temp (ºC) temp (ºC) rate(l/s) temp (ºC) 1 153 16.2 12.6 107 24.9 50.8 4 373 21.1 19.8 268 24.4 75.1 Room (1.1m) Fresh Air Humidity Total flow Temp (ºC) (%) rate(l/s) 23.4 54.3 58 23.5 76.1 134 Energy consumption analysis The cooling capacity is used here to discuss the energy consumption and the reason is the same as mentioned in the previous section. EDV-1 = 4.97 kW 88 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY EDV-4 = 2.99 kW (if fresh air flow rate is 58 l/s) 5.11kW (if fresh air flow rate is 134 l/s). Clearly, if we can control the fresh air flow rate at the same level for both cases, the cooling capacity of DV-4 would be much lower (34%) than that of DV-1. This suggests that with higher supply air temperature and room RH level, the energy consumption by the chiller, pump and cooling tower could be reduced by a substantial amount. 4.2.4 Preliminary design guide To design a suitable DV system, it is critical to compute the supply air flow rate. Once the supply flow rate is determined, one can calculate the cooling capacity and then choose the system accordingly. The selection of the size, number of the supply units and their location will depend on the room layout. One must first determine the difference between the supply-return temperature. The supply air flow rate can be determined with knowledge of the surplus heat inside a room and the supply-return temperature difference, using the following formula Q = c × m × ∆T where Q stands for surplus heat, m is the supply air flow rate, c is the air specific heat and △T is the supply-return temperature difference. The next section discusses the model used to calculate the supply-return temperature difference. 4.2.4.1 Model It was shown in Section 4.2.2.1.1 that with the DV system, the temperature profiles 89 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY below 1.1m height are close to linear except at the region from 0.3m to 0.6m where steps profile could be observed. Result of the regression analysis confirms a linear relationship. The linear relationship between temperature and height is: T = ah + b where T is the room air temperature at a certain height, in ºC; h is the height of the measuring point, in m; a = −0.435 × Ts + 9.9454 (R2=0.9916) b = 0.5274 × Ts + 11.483 (R2=0.9957) where Ts is the supply air temperature, in ºC. Figure 4.16 shows the measured temperature and the linear relationship between temperature and height. The R2 values for different regression lines are: z 16.2 ºC supply, R2=0.938; z 17.5 ºC supply, R2=0.9903; z 18.1ºC supply, R2=0.971; z 21.0ºC supply, R2=0.9998. 90 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY Temperature at different heights Temperature (ºC) 24 23 22 21 20 0 0.3 0.6 0.9 Heights (m) 1.2 16.2 17.5 18.1 21 Linear (16.2) Linear (17.5) Linear (18.1) Linear (21) Figure 4.16 Relationship between measured temperature (symbols) and predicted temperature (lines). Note: symbols--measured temperature; lines--predicted temperature. Numbers in legend are supply air temperatures. 4.2.4.2 Application of the model This model can be used to design the displacement ventilation system for buildings when combined with the “50%-rule” (Skistad et al, 2002). The “50%-rule” states that the air temperature at floor level is half-way between the supply air temperature and the extract air temperature. This model is developed based on the temperature values from the middle point inside the chamber (Point 3 in Figure 4.2) where the temperature at floor level (0.1m height) is mid-way between the supply air temperature and the extract air temperature. “50%-rule” will always exist at a point inside the room. The temperature at 1.1m height vary very little and this model can be used on that point and the 1.1m height temperature value can be computed from this model and applied to the other points inside the room. 91 CCCHHHAAAPPPTTTEEERRR 444::: CCCOOONNNFFFIIIRRRMMMAAATTTIIIOOONNN SSSTTTUUUDDDYYY 4.2.4.3 Design procedure: a) Select a suitable room air temperature (1.1m height) and supply air temperature, according to the thermal comfort standard (ASHRAE standard 55-1992 revised) or the local standard (CP13, 1999). b) Calculate the air temperature at the floor level (0.1m) using the model developed in the previous section (Section 4.2.4.1). c) Calculate the extract air temperature value based on the temperature difference between supply air and the air near the floor, using the “50%-rule”. d) Calculate the supply air flow rate based on the surplus heat and supply-return temperature difference. e) Find the required fresh air flow rate for acceptable Indoor Air Quality. f) Determine the supply air flow rate by choosing the greater from d) and e). g) Select and position supply units, based on manufacturer’s technical information and the room layout. h) Size the Air Handling Unit (AHU), based on the air flow rates (supply, return, and fresh air flow rates), the surplus heat and the room humidity requirement, and ductwork design. 92 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS CHAPTER 5 CONCLUSIONS Mixing ventilation (MV) system is used widely in a hot and humid country like Singapore. This system may lead to problems like poor air quality and hence decrease in productivity because of the mixing nature. Displacement ventilation (DV) system can resolve this problem in a more energy-efficient way. However, to date, there is limited research done on DV system in the tropics. The purpose of this research is to assess the viability of DV system in the tropics. Tropical subjects were used and thermal comfort, indoor air quality and energy performances were studied. 5.1 Review and achievement of research objective First objective To investigate the stratification effect of the wall supply displacement ventilation system. The first objective is to investigate the stratification effect of the DV system. Three kinds of stratification effects are investigated and discussed. Temperature gradients were observed in all cases at both stages of the experiments. The gradients were found to have two parts: one with steep gradient (from 1.3-1.7m to the floor level) and the other with a gentler gradient [1.3-1.7m to 2.62m (ceiling)]. The height of the stratification plane depends on the supply air flow rate. The higher the supply air flow rate, the higher the stratification plane. This is consistent with previous studies mentioned in the literature review. It was observed that under the stratification plane, the temperature difference between different measuring points at the same horizontal plane could have very large variation, depending on the distance from the 93 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS supply unit. However, above this plane, there was almost no difference between the measuring points. The reason is, above this plane, the air is re-circulated and mixed well. Humidity gradient did exist in the space with displacement ventilation system as shown by past researches. The profile of the humidity gradient depends on the supply air flow rate. The lower the supply air flow rate, the more significant the gradient. The difference in humidity between lower points and higher points is very small. This is due to the low humidity load inside the experimental chamber with the source from the human subjects. CO2 concentration gradients were observed in all cases at both stages of experiments. The concentration of points above the stratification plane depends on the outside air flow rate. The higher the outside air flow rate, the lower the concentration and hence the gradient becomes less significant. Second objective To investigate the thermal comfort and energy performances of the wall supply displacement ventilation system. i. Thermal comfort performance It is found that supply air temperature has significant influence on subjects’ thermal sensation. The higher the supply air temperature, the higher the subjects’ thermal sensation votes for overall body and different body parts. For all supply air temperatures, thermal sensation votes for feet and arms are lower than those of the other body parts. At 16 °C supply, the lower thermal sensation votes for feet and arms reach significant levels. The increase in supply air temperature has no 94 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS significant effect on subjects’ acceptability ratings. In a room with 23.5°C room air temperature and clo value of about 0.5, none of the supply air temperatures met the “less than 20%” unacceptability criterion. Out of the subjects’ unacceptability votes, more than half were cast by subjects seated at the workstation near the supply unit. Generally, draft sensation votes for all cases are lower than 15%. Although more than half of the draft sensation votes were cast at the workstation near the supply unit, the percentage at that workstation does not exceed the criteria (15%). Subjects seated near the supply unit have cooler sensation for all body parts, lower Overall AMV and lower comfort acceptability. However, these differences are not statistically significant. Therefore, occupants can still be seated near (1m) to the supply unit. Relative humidity (54-76% RH at 23.5 ºC) may have significant influence on subjects’ thermal sensation and unacceptability. However, it does not have significant influence on subjects’ thermal comfort acceptability ratings. Room air temperature (22-26ºC at 65% RH) has significant influence on subjects’ thermal sensation and unacceptability votes. The higher the room air temperature, the higher the subjects’ thermal sensation votes for overall body and different body parts. However, room air temperature has no significant effect on subjects’ acceptability ratings. ii. Energy performance To investigate the energy performance of DV system, the cooling capacity is used as energy consumption index and two conditions are compared: (1) 16 ºC supply/54% room RH, (2) 21ºC supply/76% room RH. If the fresh air flow rate 95 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS can be controlled at the same level for both conditions, the cooling capacity for (2) would be 34% lower than that for (1). As this study has only considered thermal sensation, additional study is needed to study the RH limit with response to condensation and mould growth. On the whole, 16 °C supply air temperature and 26 °C room air temperature are not recommended. These are the two extreme conditions with very high unacceptability votes. Third objective To compare the performance of wall supply displacement ventilation system against the conventional ceiling supply mixing ventilation system based on thermal comfort, indoor air quality and energy performances. i. Thermal comfort performance Results of surveys of tropical subjects show that generally subjects have cooler thermal sensation (lower AMV, for overall and various body parts) and lower acceptability with DV system than with MV system given the same room air temperature and relative humidity. The perceived cooler thermal sensation provided by the DV system allows 1 ºC higher neutral temperature as compared to the MV system. At 16 ºC supply/22 ºC spatial/65% RH, subjects have perceived significantly cooler sensation for overall body, calf, thigh, waist and arms with the DV system than with the MV system. At 26 ºC room air temperature and 65% RH the highest unacceptability (>20%) votes for both modes were found. In general, with proper design, DV system can have low draft sensation, which can be even lower than that of MV system. 96 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS ii. Indoor air quality performance DV system creates CO2 concentration gradient with the CO2 concentration of below 1.1m height always lower than that of the MV system. The ventilation efficiency ranges between 1.70 and 3.51 with DV system as compared to MV system with ventilation efficiency of 1. The CO2 concentration profile is uniform throughout the space with MV system. This is due to the mixing characteristics of the MV system iii. Energy performance When the conditions for both the systems are the same, i.e. 17.5 ºC supply air temperature, 62% RH and the same fresh air flow rate, the cooling capacity for DV system is 5% lower than that for the MV system. In addition, due to the temperature gradient, sedentary occupants will receive cooler sensation with the DV system as compared to the MV system. In order to achieve the same thermal sensation as that with MV system, the room air temperature has to be raised by about 1.0ºC. This will lead to the increase in the return air temperature and at last the decrease in the energy consumption by the DV system. Furthermore, because of the concentration gradient, subjects are exposed to lower CO2 level with DV. To obtain the same CO2 level as that with MV, the outdoor air flow rate could be reduced. This would further lower the energy consumption by DV system. On the whole, DV system can provide better thermal, indoor air quality and energy performances than MV system. 16 ºC supply/22 ºC spatial/65% RH is not recommended for DV system. 26 ºC room air temperature and 65% RH is not recommended for both systems. 97 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS Fourth objective T o develop a preliminary design guide that could be used for low ceiling offices in the tropics. To develop a design guide, the first task is to develop a model with which the supply air flow rate can be calculated. Based on the experiment results, a model is developed for 23.5 ºC room air temperature. This model applies only to low ceiling office and with room air temperature of 23.5 ºC with displacement ventilation system. 5.2 Recommendation Based on the findings of this study, some future works are recommended as follow: i. More room air temperatures The experiments at the second stage were all conducted at 23.5 ºC room air temperature. The results have shown that 23.5 ºC may be low for local subjects with 0.5 clo office attire. Hence, to comprehensively investigate the thermal performance of DV system, higher room air temperature, for example 25 ºC, may be appropriate. On the other hand, one of the objectives of the research project is to develop a design guide that could be used in the tropics. As such a design guide needs to embrace and recommend an operating temperature range to the practitioners, therefore higher room air temperature is also necessary. ii. Variation of activity level To simulate an office environment, subjects were required to do some light office work, including data processing and typing, the metabolic rate of which was deemed to be 1.0 or 1.2 met. As activity level is a very important factor influencing thermal comfort, 98 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS in order to comprehensively investigate the thermal performance of DV system, the variation of activity level is therefore necessary. Future study can include activities like quiet reading, walking and medium level office work. iii. Variation of mean radiant temperature The experiments were conducted in the thermal chamber which is located inside another larger chamber. As this larger chamber is well shaded from the sun, the influence of solar radiation on thermal chamber is therefore reduced to the least. With real office buildings that have external walls and/or windows, the effect of solar radiation on the indoor thermal environment could be significant. Hence, to comprehensively investigate the thermal performance of DV system in the tropical area which is exposed to abundant sunshine, the variation of mean radiant temperature is therefore necessary. iv. Variation of heat loads The experiments were conducted in the thermal chamber designed for six subjects. Inside the chamber, there were six lighting fixtures and six PCs. It should be deemed as a “simplified” office environment. In a real office environment, the heat load could be very complicated. Future study can include photocopiers, printers, projectors, fax machines, and other “high-heat-generating” machines. Furthermore, some office buildings may have external walls and/or windows. The heat from all these would add up to a great amount. To develop a complete design guide, these should also be taken into account. Therefore, future study with various heat loads will be necessary. v. Comprehensive energy study 99 CCCHHHAAAPPPTTTEEERRR 555::: CCCOOONNNCCCLLLUUUSSSIIIOOONNNSSS This study used cooling capacity as the energy consumption index to investigate the energy performance of the DV system. The energy consumed by the air-conditioning system can be reflected based on the cooling capacity needed for the system (normally, the higher the cooling capacity, the more energy consumption). However a more comprehensive energy study which details the breakdown of energy consumed by for example, chillers, cooling towers, pumps, fans and its accessories, would be more meaningful. 100 BBBIIIBBBLLLIIIOOOGGGRRRAAAPPPHHHYYY BIBLIOGRAPHY Akimoto, Takashi; Nobe, Tatsuo; Tanabe, Shin-ichi; Kimura, Ken-ichi. (1999). Floorsupply displacement air-conditioning: laboratory experiments. ASHRAE Transactions, Volume 105, Pages 739-748. ASHRAE. (1992). Standard 55 - Thermal environmental conditions for human occupancy. Atlanta, Ga: American Society of Heating, Refrigerating and Air-Conditioning Engineers. ASHRAE. (1995). Standard 55a - Addendum to thermal environmental conditions for human occupancy. World Wide Web: (http://www.ashrae.org). Awbi, H. B. (1998). Energy efficient room air distribution. Renewable Energy, Volume 15, Issues 1-4, 12 September, Pages 293-299 Baker, D.B. (1989). Social and organizational factors in office building-associated illness. Occupational Medicine 4 (4), 607-624. Brohus, H., H.N. Knudsen, P.V. Nielsen, G. Clausen and P.O. Fanger. (1996). Proceedings of Indoor Air ´96, The 7th International Conference on Indoor Air Quality and Climate, July 21 - 26, Nagoya, Japan, Vol. 1, pp. 811 - 816,. Busch, J.F. (1995). Thermal comfort in Thai air-conditioned and naturally ventilated offices. In Nicol, F., Humphreys, M., Sykes, O. and Roaf, S. (Ed.), Standards for thermal comfort: indoor air temperature standards for the 21st century (pp114121). New York: Chapman & Hall. Cheong, K.W.D., Djunaedy, E., Chua, Y.L., Tham, K.W., Sekhar, S.C., Wong, N.H. and Ullah, M.B. (2003). Thermal comfort study of an air-conditioned lecture theatre in the tropics. Building and Environment, 38, 63 – 73. De Dear, R.J., Leow, K.G. and Foo, S.C. (1991). Thermal comfort in the humid tropics: field experiments in air conditioned and naturally ventilated buildings in Singapore. International Journal of Biometeorology, 34, 259-265. Engen, T. (1986). Perception of odor and irritation. Environ. Int. 12:177-187. Fang, L., Clausen, G. and Fanger, P.O. (1998a) Impact of temperature and humidity on the perception of indoor air quality. Indoor Air, 8, 80-90. Fang, L., Clausen, G. and Fanger, P.O. (1998b) Impact of temperature and humidity on perception of indoor air quality during immediate and longer whole-body exposures. Indoor Air, 8, 276-284. 101 BBBIIIBBBLLLIIIOOOGGGRRRAAAPPPHHHYYY Fang, L.F., Clausen, G. and Fanger, P.O. (2000). Temperature and humidity: important factors for perception of air quality and for ventilation requirements. ASHRAE Transaction, 106(2), 503-510. Fanger, P.O. (1970). Thermal comfort: analysis and applications in environmental engineering. New York: McGraw-Hill. Fanger, P.O.; Melikov, A.K.; Hanzawa, H.; Ring, J. (1988). Air turbulence and sensation of draught. Energy and Buildings, Volume 12, Issue 1, Pages 21-39 Gunnarsen, Lars. and Fanger, P.O. (1992). Adaptation to indoor air pollution. Environment International, 18, 43-54. Hensen, J.L.M.; Hamelinck, M.J.H. (1995). Energy simulation of displacement ventilation in offices, Building Services Engineering Research & Technology, Volume 16, Issue 2, Pages 77-81. Hu, Shiping; Chen, Qingyan; Glicksman, Leon R. (1999). Comparison of energy consumption between displacement ventilation systems for different U.S. buildings and climates, ASHRAE Transactions, Volume 105, Pages p 453-464. Institute of Environmental Epidemiology, Ministry of the Environment. (1996). Guidelines for good indoor air quality in office premises. Singapore: Institute of Environmental Epidemiology, Ministry of the Environment. ISO. (1985). Standard 7726 - Ergonomics of the thermal environment - Instruments for measuring physical quantities. Geneva: International Organization for Standardization. ISO. (1994). Standard 7730 - Moderate thermal environments-determination of PMV and PPD indices. Geneva: International Organization for Standardization. Kosonen, R, Livchak, A, Heikkinen (2001). Displacement Ventilation – Efficient System for Room Air Moisture Control. ASIA PACIFIC CONFERENCE ON THE BUILT ENVIRONMENT, Singapore, 14-17 November. Kwok, A.G. (1998). Thermal comfort in tropical classrooms. ASHRAE Transactions, Volume 104(1b), Pages p 1031-1046 Lagercrantz, L., Wistrand, M., Willén, U., Wargocki, P., Witterseh, T. and Sundell, J. (2000) Negative impact of air pollution on productivity: previous Danish findings repeated in new Swedish test room. In: Seppänen, O.and Säteri, J., eds. Proceedings of Healthy Buildings 2000, Vol. 1. Espoo, Finland: Helsinki University of Technology, 653–658. Livchak A., Nall, D. (2001). Displacement Ventilation- Application for Hot and Humid Climate. Clima 2000/Napoli 2001 World Congress- Napoli 15-18 September 2001. McIntyre, D.A. (1978). Seven point scales of warmth. Building Services Engineer, 45: 215-226 102 BBBIIIBBBLLLIIIOOOGGGRRRAAAPPPHHHYYY Melikov, A. and Nielsen, J. (1989). Local thermal discomfort due to draft and vertical temperature difference in rooms with displacement ventilation. ASHRAE Transactions, 95(2), 1050-1057. Murakami, Sh.; Kato, Sh.; Zeng, J. (1998). Numerical simulation of contaminant distribution around a modeled human body: CFD study on computational thermal manikin - Part II. ASHRAE Transactions, Volume 104, Issue 2, Pages 226-233 Nevins, R. G., Rohles, F. H., Springer, W., and Feyerherm, A. M. (1966). A temperaturehumidity chart for thermal comfort of seated persons. ASHRAE Trans. 72. I: 283291. Parsons, K. C. (2002). The effects of gender, acclimation state, the opportunity to adjust clothing and physical disability on requirements for thermal comfort. Energy and Buildings, Volume 34, Issue 6, July, Pages 593-599 Pitchurov, G., Naidenov, K., Melikov, A. and Langkilde, G. (2002). Field survey of occupants thermal comfort in rooms with displacement ventilation. In Roomvent’02, Indoor climate performance in buildings. World Wide Web: (http://www.roomvent.dk). Robinson, J., Nelson, W.C. (1995). National Human Activity Pattern Survey Data Base. United States Environmental Protection Agency, Research Triangle Park, NC. Sandberg, M. 1985. Air-exchange efficiency, ventilation effectiveness, temperature efficiency in closed office room systems with air for both heating and cooling. M85:24. Gavle, Sweden: The National Swedish Institute for Building Research. Seppanen, O.A.; Fisk, W.J.; Eto, J.; Grimsrud, D.T. (1989). Comparison of conventional mixing and displacement air-conditioning and ventilating systems in U.S. commercial buildings. ASHRAE Transactions, Volume 95(2), Pages 1028-1040 SISIR. (1999). Singapore Standard CP13 - Code of practice for mechanical ventilation and air-conditioning in building. Singapore: Singapore Institute of Standards and Industrial Research. Skistad, H., Mundt, E., Nielsen, P.V., Hagström,K., Railio, J. (2002). Displacement ventilation in non-industrial premises. Guidebook No 1, REHVA. Tan, S.C. (1995). Thermal comfort perception of Singaporeans in air-conditioned buildings. Unpublished undergraduate dissertation, School of Building and Estate Management, National University of Singapore. U.S. Environmental Protection Agency. (1990). Indoor Air Fact Sheet No.3R: Ventilation and air quality in offices. http://www.epa.gov/iaq Wargocki, P., Lagercrantz, L., Witterseh, T., Sundell, J., Wyon, D. P., Fanger, P. O. (2002). Subjective perceptions, symptom intensity and performance: a comparison of two independent studies, both changing similarly the pollution load in an office. Indoor Air, 12, 74-80. 103 BBBIIIBBBLLLIIIOOOGGGRRRAAAPPPHHHYYY Wargocki, P., Wyon, D.P., Baik, Y.K., Clausen, G.and Fanger, P.O. (1999). Perceived air quality, Sick Building Syndrome (SBS) symptoms and productivity in an office with two different pollution loads. Indoor Air, 9, 165–179. Wargocki, P., Wyon, D.P., Sundell, J., Clausen, G.and Fanger, P.O. (2000). The effects of outdoor air supply rate in an office on perceived air quality, Sick Building Syndrome (SBS) symptoms and productivity. Indoor Air, 10, 222–236. World Health Organisation (WHO). (1983). Indoor Air Pollutants: Exposure and Health Effects. EURO Reports and Studies, No. 78. WHO, Copenhagen. Wyon, D.P.; Sandberg, M. (1990). Thermal manikin prediction of discomfort due to displacement ventilation. ASHRAE Transactions, Volume 96(1), Pages 67-75. Xing, H.; A. Hatton and H. B. Awbi. (2001). A study of the air quality in the breathing zone in a room with displacement ventilation. Building and Environment, Volume 36, Issue 7, August, Pages 809-820. Xu, M; Yamanaka, T; Kotani, H. (2001). Vertical profiles of temperature and contaminant concentration in rooms ventilated by displacement with heat loss through room envelopes. Indoor Air, Volume 11, Issue 2, June, Pages 111-119. Yuan, Xiaoxiong; Chen, Qingyan; Glicksman, Leon R. (1998). Critical review of displacement ventilation. ASHRAE Transactions, Volume 104, Part 1A, Pages 78-89 Yuan, Xiaoxiong; Chen, Qingyan; Glicksman, Leon R. (1999a). Models for prediction of temperature difference and ventilation effectiveness with displacement ventilation. ASHRAE Transactions, Volume 105 (PART 1), Pages 353-367. Yuan, Xiaoxiong; Chen, Qingyan; Glicksman, Leon R. (1999b). Performance evaluation and design guidelines for displacement ventilation. ASHRAE Transactions, Volume 105 (PART 1), Pages 298-309. Yuan, Xiaoxiong; Chen, Qingyan; Glicksman, Leon R.; Hu, Yongqing; Yang, Xudong (1999c). Measurements and computations of room airflow with displacement ventilation. ASHRAE Transactions, Volume 105 (PART 1), Pages 340-352. 104 [...]... the rising stream had a negative influence on the quality of the inhaled air when the contaminants were generated in the lower part of the room below the breathing height, and the air in the lower part of the room was relatively dirty 2.4.2 Age of air The age of air is defined as the time for all air molecules to travel from the air supply device to a point in the space It can be derived from the measured... is not extensively investigated in the tropics This can be detrimental during the operation of building if it cannot perform or under-perform Therefore, there is a need to assess the feasibility and viability of displacement ventilation system in the tropics It is therefore of great importance to conduct research to assess its applicability in the Tropics The results and findings of this study will... Hence, the results of the research conducted in Scandinavian countries may not be applicable in Singapore There is a growing trend of new buildings exploring the possibility of adopting new air-conditioning technology in Singapore The acceptance of new system can be 2 CCCHHHAAAPPPTTTEEERRR 111::: IIINNNTTTRRROOODDDUUUCCCTTTIIIOOONNN difficult in the construction industry since the suitability of such systems... difference in both of the displacement system than in the ceiling-based system A large vertical temperature difference that may cause local thermal discomfort was observed in several cases for both of the displacement systems It was observed that the measured skin temperatures of the thermal manikin with both of the displacement systems were slightly lower than those of the ceiling-based system However,... systems in order to save energy The ratio of the return air to the total supply air could range between 70% and 90%, depending on the types of building DV system has to comply with the energy-saving rule, i.e 70% to 90% of the exhaust air needs to be recirculated, if it is to be used in Singapore However, 100% outside air can be used in the Scandinavian countries for DV system due to their climatic condition... comfortable indoor environment economically 1.2 Objectives and scope of research The main objectives of the research study are: a To investigate the stratification effect of the wall supply displacement ventilation system; b To investigate the thermal comfort and energy performances for the wall supply displacement ventilation system; c To compare performance of the wall supply displacement ventilation system. .. countries There is limited research done in the tropics As the ethnic groups and building loads in the tropics are not the same as those in Scandinavian countries, the results of such research may not be directly applicable Moreover, in the tropics, due to the all-year-round hot and humid climate, there is high recirculation of air for most of the Air conditioning Mechanical Ventilation (ACMV) systems in. .. breathing zone 16 CCCHHHAAAPPPTTTEEERRR 222::: LLLIIITTTEEERRRAAATTTUUURRREEE RRREEEVVVIIIEEEWWW depended on the location of the contaminant generation When contaminants were generated in the upper part of the room above the breathing height, and the air in the lower part of the room was relatively clean, the rising stream of air had a positive influence on the quality of the inhaled air Conversely, the. .. 2.4 Indoor air quality The concern about energy efficiency has increased since the 1970’s oil crisis This growing concern has led to many changes in the way buildings are constructed and operated The most conspicuous ones are the reduction in ventilation rate and increase in air-tightness of buildings In the age of rapid technology development, more and more synthetic building materials are used to. .. point in exhaust, point in supply, point in a room and point at the head level of a sedentary person respectively Xing et al (2001) carried out measurements with the presence of a heated mannequin and other heat sources and found that the ventilation effectiveness at the breathing zone for both the seated and standing mannequins were greater than for a point at the same height in the chamber for the