ASHRAE - SystemPerformanceEvaluation

System Performance Evaluation and Design Guidelines for Displacement Ventilation
This publication was prepared under ASHRAE Research Project 949 in cooperation with TC 5.3, Room Air Distribution, and TC 4.10, Indoor Environment Modeling.
About the Authors

Qingyan (Yan) Chen is a professor of mechanical engineering at Ray W. Herrick Laboratories, Purdue University, West Lafayette, Indiana. He received his B.Sc. degree from Tsinghua University and M.Sc. and Ph.D. degrees from Delft University of Technology. He has published over 80 archival journal papers and more than 60 conference papers. Since 1995, he has been the principal investigator or co-principal investigator of 30 sponsored research projects, including five from ASHRAE. He has been elected to the International Academy of Indoor Air Sciences. Currently, Prof. Chen serves as an associate editor for the International Journal of HVAC&R Research and as an editorial board member for the International Journal of Ventilation and the International Journal on Architectural Science. Leon R. Glicksman is a professor of building technology in the Department of Architecture as well as professor of mechanical engineering at Massachusetts Institute of Technology (MIT). He received his B.Sc. and Ph.D. degrees from MIT and his M.Sc. degree from Stanford. Currently, Prof. Glicksman is leading an MIT effort to develop energy-efficient, sustainable building technologies and compatible designs. He has conducted research sponsored by the EPA, NSF, DOE, ABB, and numerous industrial sponsors. He has written over 180 technical articles and chapters in four books. Currently, Prof. Glicksman serves as an associate editor for ASHRAEs International Journal of HVAC&R Research.
Qingyan Chen Leon Glicksman
American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.
ISBN 1-931862-42-7 2003 American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. 1791 Tullie Circle, N.E. Atlanta, GA 30329 www.ashrae.org All rights reserved. Printed in the United States of America Cover design by Tracy Becker. ASHRAE has compiled this publication with care, but ASHRAE has not investigated, and ASHRAE expressly disclaims any duty to investigate, any product, service, process, procedure, design, or the like that may be described herein. The appearance of any technical data or editorial material in this publication does not constitute endorsement, warranty, or guaranty by ASHRAE of any product, service, process, procedure, design, or the like. ASHRAE does not warrant that the information in the publication is free of errors, and ASHRAE does not necessarily agree with any statement or opinion in this publication. The entire risk of the use of any information in this publication is assumed by the user. No part of this book may be reproduced without permission in writing from ASHRAE, except by a reviewer who may quote brief passages or reproduce illustrations in a review with appropriate credit; nor may any part of this book be reproduced, stored in a retrieval system, or transmitted in any way or by any meanselectronic, photocopying, recording, or otherwithout permission in writing from ASHRAE.
ASHRAE STAFF
SPECIAL PUBLICATIONS Mildred Geshwiler Editor Erin Howard Assistant Editor Christina Helms Assistant Editor Michshell Phillips Secretary PUBLISHING SERVICES Barry Kurian Manager Jayne Jackson Production Assistant
PUBLISHER W. Stephen Comstock
Contents
PREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .vii ACKNOWLEDGMENTS. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix CHAPTER 1INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Displacement Ventilation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2 Special Features in U.S. Buildings . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.3 Objective of This Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 CHAPTER 2LITERATURE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.1 Temperature Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.2 Flow Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.3 Contaminant Distribution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.4 Comfort Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.5 Energy and Cost Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.6 Design Guidelines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 CHAPTER 3EXPERIMENTAL STUDY AND VALIDATION OF CFD PROGRAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.1 Experimental Facility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.2 Test Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.3 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 3.4 Computational Fluid Dynamics Model . . . . . . . . . . . . . . . . . . . . . . . . 42 3.5 Validation of CFD Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 CHAPTER 4MODELS FOR PREDICTION OF TEMPERATURE DIFFERENCE AND VENTILATION EFFECTIVENESS . . . . . . . . . . . . . . . . . 55 4.1 A Database of Displacement Ventilation . . . . . . . . . . . . . . . . . . . . . . . 56 4.2 Model of the Air Temperature Difference Between the Head and Foot Level . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.3 Ventilation Effectiveness Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
CHAPTER 5PERFORMANCES EVALUATION OF DISPLACEMENT VENTILATION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Evaluation Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 Performance Evaluation of Displacement Ventilation . . . . . . . . . . . . 5.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
81 81 84 91 94
CHAPTER 6ENERGY AND COST ANALYSIS . . . . . . . . . . . . . . . . . . . . . 95 6.1 Load Calculations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 6.2 Secondary Systems and Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 6.3 Energy Analysis for U.S. Conditions . . . . . . . . . . . . . . . . . . . . . . . . 100 6.4 First Cost Analysis for U.S. Conditions . . . . . . . . . . . . . . . . . . . . . . 106 6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 CHAPTER 7DESIGN GUIDELINES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 CHAPTER 8CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 NOMENCLATURE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 INDEX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
vi
Preface
This book presents system performance evaluation and design guidelines for displacement ventilation. The authors first reviewed the literature concerning the performance of traditional displacement ventilation. Since U.S. buildings have different layouts and larger internal heat gains than those studied in the literature, it was necessary to develop design guidelines for displacement ventilation for U.S. buildings under different climatic conditions. The design guidelines present two important models that were not available in the literature: a model to calculate the temperature difference between the head and foot level of an occupant and a model to determine the ventilation effectiveness at the breathing level. The investigation developed the models from the results of 56 cases of displacement ventilation obtained by a computational fluid dynamics (CFD) program. Those cases include a wide range of thermal and flow conditions similar to those found in U.S. offices, classrooms, and workshops. The CFD program was validated by six sets of detailed experimental data obtained from a fullscale environmental chamber simulating a small office, a quarter of a large office with partition, and a quarter of a classroom. The data include airflow patterns and distribution of air velocity, temperature, contaminant concentration, and turbulence. The validation also used some data obtained from the literature. The CFD program was also used to assess the performance of displacement ventilation, such as airflow pattern and distributions of air temperature, percentage dissatisfied due to draft, predicted percentage dissatisfied, contaminant concentration, mean age of air, and ventilation effectiveness. The investigation also conducted energy and first costs analysis. The results show that a displacement ventilation system can provide a thermally comfortable indoor environment at a high cooling load through careful design. The indoor air quality in a space with displacement ventilation is better if the contaminant sources are associated with the heat sources. The displacement ventilation system can also save energy but requires a separate heating system if it is applied to building perimeter zones. This book presents a ten-step design guideline to design the displacement ventilation system for U.S. buildings.
vii
Acknowledgments
This book is based on the research performed for ASHRAE Research Project RP-949, Performance Evaluation and Development for Design Guidelines for Displacement Ventilation. The research was sponsored by TC 5.3, Room Air Distribution, and TC 4.10, Indoor Environment Modeling. Throughout the research, the project monitoring committee and the members of the two technical committees made a substantial contribution to the project, including numerous suggestions in the project meetings, critical comments on the final report, and a site visit. The authors are very grateful for their support and help. The authors would also like to thank their former research associates and students, Dr. Xiaoxiong (John) Yuan, Mr. Shiping Hu, Ms. Yongqing Hu, and Prof. Xudong Yang, for their hard work on the project. Without their contributions, such a book would not exist. Last, but not least, the authors are grateful to the ASHRAE Special Publications staff for their careful and beautiful work on the book layout and edit.
ix
CHAPTER 1
Introduction
Since the energy crisis in the 1970s, the insulation of buildings has been improved in order to reduce heat loss in winter, heat gain in summer, and the infiltration of outdoor air. As a consequence, the heat extracted from or supplied to a room for maintaining a comfortable air temperature is reduced and the ventilation rate is also reduced by a corresponding amount, sometimes much more if the building envelope is made tighter. However, such a reduction of air supply causes an increase in the concentration of indoor pollutants and sometimes generates a nonuniform distribution of air temperature and contaminant concentration. Draft (thermal comfort problems) and sick building syndrome (indoor air quality problems) are very familiar ailments today that are the direct results of the poor distribution of airflow, temperature, and contaminant concentrations. Solving these thermal comfort and indoor air quality (IAQ) problems without consuming too much energy is a challenge for both ventilation engineers and architects. Currently, the United States consumes more than one-third of its energy in buildings, and there is a possibility of saving up to 20% of this energy. Saving energy may result in the reduction of the fresh air supply. This may cause poorer IAQ. Since people spend up to 90% of their time indoors, IAQ is increasingly recognized as an essential factor for the prevention of human diseases and the promotion of people's comfort and welfare. In the United States, about 800,000 to 1,200,000 commercial buildings with 30 to 70 million people have problems related to IAQ (Woods 1989). The problems include eye, nose, and throat irritation, headache, recurrent fatigue, drowsiness or dizziness, and reduced powers of concentration (Spengler 1995). Dissatisfaction with the working environment could result in reduced productivity and economic loss. A survey conducted in the New England area of 94 state government office buildings showed an average productivity loss of 3%, which is attributed to poor IAQ (Axelrad 1989). Fisk (2000) estimated that the economic impact related to respiratory illness, allergies and asthma, and sick building syndrome is $20 to $200 billion. Therefore, it is necessary to provide a good ventilation system that can provide good IAQ and save energy.
1.1 DISPLACEMENT VENTILATION Displacement ventilation has been used quite commonly in Scandinavia during the past twenty years. It was first applied to the welding industry in 1978 (Belin 1978) and has since been increasingly used as a means of ventilation in industrial facilities to provide good indoor air quality and save energy. More recently, its use has been extended to ventilation in offices and other commercial spaces where, in addition to air quality, comfort is an important consideration. 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 (Svensson 1989). Displacement ventilation system can be divided into the following three types: Traditional displacement ventilation, as shown in Figure 1.1 Displacement ventilation with a chilled ceiling panel Displacement ventilation with a raised floor
This book focuses on the first type: traditional displacement ventilation. A typical displacement ventilation system for cooling, as shown in Figure 1.1, supplies conditioned air from a low sidewall diffuser. The supply air temperature is slightly lower than the desired room air temperature, and the supply air velocity is low (lower than 100 fpm or 0.5 m/s). Through the diffuser, the conditioned air is directly introduced to the occupied zone, where the occupants stay. Exhausts are located at or close to the ceiling through which the warm room air is exhausted from the room. Because it is cooler than the room air, the supply air is spread over the floor and then rises as it is heated by the heat sources in the occupied zone. These heat
Figure 1.1
Sketch of displacement ventilation.
Introduction
sources (e.g., persons and computers) create upward convective flows in the form of thermal plumes. These plumes remove heat and contaminants that are less dense than air from the surrounding occupied zone. Traditionally, the amount of supply air in a displacement ventilation system has been less than that of mixing-type systems. This necessitates careful design of the system configuration and operation to adequately handle the space cooling loads. The supply temperature, velocity, and vertical temperature gradient in the occupied zone are all very important comfort-related design parameters. Compliance with the specification of ASHRAE Standard 55-1992 (ASHRAE 1992) for acceptable vertical temperature difference in the occupied zone places limitations on the magnitude of supply-room temperature difference and/or space cooling loads for a given supply airflow rate. This is especially important when the system is applied to a U.S. building in which the cooling load can be high and weather can be hot. Previous research (Svensson 1989; Sandberg and Blomqvist 1989; Wyon and Sandberg 1990) has indicated that in office environments with normal room heights of around 9 ft (2.7 m), displacement ventilation cannot maintain acceptable comfort for cooling loads above 8 to 10 Btu/(hft2) (25 to 30 W/m2) unless the air supply volume is increased or additional heat removal capacity is provided through the use of cooled ceiling panels. With higher ceiling heights, displacement ventilation systems are capable of removing larger heat loads. A stable, vertically stratified temperature field is essential for this type of system to function properly. Numerous studies show that, when properly designed, displacement ventilation can take advantage of the naturally occurring thermal stratification in the room and, thus, can increase the ventilation efficiency. 1.2 SPECIAL FEATURES IN U.S. BUILDINGS Research on displacement ventilation has been mainly conducted in Scandinavian countries. Recently, REHVA (2002) published a guidebook on designing displacement ventilation in non-industrial premises. Many U.S. cities have higher temperatures in summer than those in Scandinavian cities, and U.S. offices may have more lighting and equipment that produces more heat. Therefore, the cooling load could be higher in the U.S. than in Scandinavian countries (Chen et al. 1999). In many U.S. offices, there are large core spaces that are completely isolated from the external climate. Cooling is always needed in the core spaces, and there is great potential for the use of displacement ventilation in such spaces. On the other hand, heating and cooling are required in the perimeter spaces. In Scandinavian countries, a radiator is often used to offset heating load in winter and fresh air is supplied by the displacement ventilation system. This implies that the supply air temperature in winter can still be somewhat lower than the room air temperature, and a stratified flow can be maintained. However, in many U.S. office buildings, air-conditioning systems are often used for both heating and cooling and there is no radiator available. If a displacement ventilation system is used in the
perimeter spaces, a separate heating system is needed to maintain the flow pattern. Convectors, baseboard heaters, radiant panels, or resistance wires can be used. However, the first costs and operating costs with two systems would be different. Displacement diffusers can be used for heating as well, but the airflow pattern will be of the mixing type. In addition, many U.S. offices and restaurants are large spaces with many partitions to form individual work stations or dinning areas, while most European offices and restaurants are small spaces. Therefore, direct application of the Scandinavian results for U.S. design is not feasible. 1.3 OBJECTIVE OF THIS BOOK Displacement ventilation may improve indoor air quality and has the potential to save energy. However, the performance of displacement ventilation is still not totally understandable, and the special features of U.S. buildings have not been considered in previous research. The objective of this book is to answer the following two questions: Is displacement ventilation suitable for U.S. buildings? How should displacement ventilation systems be designed?
To evaluate whether a ventilation system is suitable for U.S. buildings, we need to consider, simultaneously, its impact on indoor air quality, comfort, energy consumption, and costs. In order to design such a ventilation system, it is necessary to provide a design guide. This book tries to answer the above questions by providing the following information for displacement ventilation systems: 1. 2. 3. 4. 5. 6. 7. Literature review to identify the existing results and problems Experimental study to get reliable data, including the distribution of velocity, turbulence intensity, temperature, tracer-gas concentration, etc. Validation of a computational fluid dynamics (CFD) program by the experimental data to determine the accuracy of the program Numerical simulation of a large number of cases by the CFD program to establish a database on the performance of displacement ventilation Model development to develop models needed for design guidelines Energy and cost analysis to assess the impact of energy and first costs Guidelines to help designers in the U.S. to design displacement ventilation
Chapter 2 of this book presents a state-of-the-art review on displacement ventilation. Chapter 3 describes the experimental study in a full-size test room simulating a small office, a large office with partitions, and a classroom. The experimental results are used to validate a CFD program. Chapter 4 describes a database of displacement ventilation by CFD computations of numerous cases for different thermal and flow conditions for different types of U.S. buildings. Based on the computed
Introduction
results, two models are developed for prediction of the air temperature difference between head and foot level and the ventilation effectiveness at head level. Chapter 4 also introduces a simplified CFD program for calculating indoor airflow. Chapter 5 discusses the performance of displacement ventilation. Chapter 6 describes energy and cost analysis. Chapter 7 outlines a ten-step design guideline for displacement ventilation. Chapter 8 offers important conclusions about displacement ventilation.
CHAPTER 2
Literature Review
This chapter discusses some of the literature on displacement ventilation, including studies on the distribution of temperature, velocity, and contaminants, comfort, energy, cost analysis, and design guidelines. 2.1 TEMPERATURE DISTRIBUTION Since displacement ventilation systems supply cold fresh air directly to the occupied zone, potential draft exists at floor level. In addition, the large temperature stratification that exists in a space with displacement ventilation may also cause discomfort. Therefore, a designer needs information on the air temperature distribution in spaces with displacement ventilation. Dimensionless Temperature and Vertical Temperature Gradient Researchers found that the air temperature in a space appears to vary linearly with space height in the stratified zone and is nearly constant in horizontal directions except in the region near the supply diffusers. Figure 2.1 presents a typical temperature profile assumed for a room with displacement ventilation, where Ts, Tf, and Te are air temperatures at supply outlets near the floor in the occupied zone and at exhausts. This linear profile is widely used by most investigators, e.g., Mathisen and Skaret (1983), Flatheim (1984), Sandberg (1985), Holmberg et al. (1987), Nielsen (1993), and Skistad (1994). Figure 2.2 plots the vertical dimensionless temperature profiles, = (T Ts / Te Ts), in normal offices obtained from several different investigations. The dimensionless temperature near the floor, f, or the ratio of the temperature difference between the supply air and the air near the floor to the difference between the supply and extract air, varies from 0.2 to 0.7. In addition, the air temperature gradient is not the same for different investigations. The temperature does not vary linearly
Figure 2.1
Simplified vertical temperature profile in a room with displacement ventilation.
Figure 2.2
Temperature profiles in office rooms.
from the floor to the exhaust for most cases. The discrepancies among the profiles in Figure 2.2 could be due to different thermal and flow conditions, such as: ventilation rate cooling load heat source type and position wall temperature and wall radiative characteristics space height diffuser type
Literature Review
Since the temperature difference between head and feet is a critical criterion of thermal comfort, it is desirable to have a model to predict the vertical temperature gradient in the occupied zone. Impact of Ventilation Rate and Cooling Load Sandberg (1985) measured vertical temperature profiles in a test room (14 ft 12 ft 8.3 ft or 4.2 m 3.6 m 2.5 m) with displacement ventilation. The results show that f is between 0.56 and 0.48 when the air change rate is between 2 and 4 ach. The results of Chen et al. (1988) for a test chamber of 18.7 ft 10 ft 10.7 ft (5.6 m 3.0 m 3.2 m) confirmed that f is a function of ventilation flow rate. The f decreases from 0.4 to 0.2 when the air change rate increases from 3 to 5 ach. Similar results can be found from Mundt (1990) and Li et al. (1992). Mundt (1990) developed a formula to calculate f based on the assumption that the convective heat transfer from the floor to air raises the air temperature from Ts to Tf. The radiative heat transfer from the ceiling to the floor keeps the energy balance on the floor surface. The f is a function of ventilation rate given as
1 -, f = ------------------------------------------------V Cp 1 1 ------------- ----- + ------ +1 A r cf
(2.1)
where V = ventilation flow rate, = air density, Cp = specific heat at constant pressure, A = floor area, r = radiative heat transfer coefficient from ceiling to floor, and cf = convective heat transfer coefficient from the floor to room air. As shown in Figure 2.3, Equation 2.1 is in good agreement with most measured data in the literature (the same references cited in Figure 2.2) when r = 0.9 Btu/ (hft2F) (5 W/m2K) and cf = 0.7 Btu/(hft2F) (4 W/m2K). Point 1 in Figure 2.3 is relatively small because the walls were covered with aluminum in the experiment and the radiative heat transfer to the floor was small. Point 2 is relatively large because the cooling load was small and the total temperature difference was only 5F (2.8C) in the experiment. Equation 2.1 accounts for the impact of cooling load on f because ventilation rate and cooling load are interrelated. If the temperature varies linearly with the elevation, the temperature gradient, s, can be estimated as s = (1 f) (Te Ts) / H or (2.2)
Figure 2.3
Dimensionless temperature near the floor vs. supply flow rate.
1 sH -------- = 1 f = 1 ------------------------------------------------ . V cp 1 T e 1 ------------ ----- + ------ +1 A r cf
(2.3)
Unfortunately, Equation 2.3 does not correlate the temperature gradient, as shown by Figure 2.4, where the points represent the measured data (the same references cited in Figure 2.2). Impact of Heat Source Type and Position and Wall Characteristics Nielsen et al. (1988) showed that, in a room with constant load from a concentrated heat source, the gradient of dimensionless temperature decreases slightly as the Archimedes number (ghTe/us2) increases. The gradient is strongly related to the surface temperature of the heat sources (Nielsen 1992). Nielsen (1996) later presented a design chart, shown in Figure 2.5, from experimental results in rooms with heights of 8.3 ft to 15 ft (2.5 m to 4.5 m) to determine dimensionless temperature of air near the floor. It is a common practice to assume adiabatic thermal conditions for internal walls. In many cases, the internal walls are not adiabatic. Since wall area is large, a small temperature difference between the walls and room air could result in a significant downflow (if the walls are colder) or upflow (if the walls are warmer). In addition, the temperature along a vertical line of an internal wall is not a constant, and there is a temperature gradient in room air. Heat transfer occurs between room
10
Literature Review
Figure 2.4
Dimensionless temperature gradient vs. supply flow rate.
Figure 2.5
A design chart for the dimensionless temperature near the floor (Nielsen 1996).
11
Figure 2.6
Comparison of the temperatures between the models and the experiment.
air and internal walls. Most of the investigations neglect the impact of internal wall temperature on the vertical temperature gradient. A study conducted by Jarmyr (1982) showed vertical temperature profiles at five different times of day in a workshop. The temperature gradient increased from 0.23F/ft (0.38 K/m) in the early morning to 0.42F/ft (0.7 K/m) at noon and then decreased slightly to 0.39F/ft (0.65 K/m) in the evening. The nondimensional temperature near the floor, f, varied from 1/3 in the morning to 1/7 in the afternoon. It is clear that heat from external walls and windows contributes to the temperature gradient. Li et al. (1992) showed that heat conduction through walls and radiation between room surfaces, particularly between ceiling and floor, make a significant contribution to the vertical temperature profile. For example, the dimensionless air temperature near the floor was changed from 0.62 to 0.3 when the black walls were covered with aluminum. To include the contribution of radiative heat transfer and conduction through walls, Li et al. (1992) suggested a four-node model and a multinode model. Good agreement was found between the models and their measured data. Mundt (1996) recently extended her early model (Mundt 1990) to consider the influences of heat transfer through the building enclosure and the heat sources on the vertical temperature profile. The new model relates air temperatures near the floor and ceiling to room geometry and the heat transfer among the room air, heat sources, floor, and ceiling. Figure 2.6 shows that the models of both Mundt (1996) and Li et al. (1992) could predict the measured data of Li et al. (1992). However, the models do not closely predict the results of various investigators. Since both models assume the air temperature gradient to be constant, they need improvements. Cooled ceiling panels with displacement ventilation are often used in spaces with a high cooling load. The vertical temperature gradients in the spaces with the
12
Literature Review
cooled ceiling panels are smaller than those without the panels. The temperature distributions are almost uniform in the upper zones, as reported by Skistad (1994) and Taki et al. (1996). If the panel temperature is too low, the displacement ventilation could be transformed into mixing ventilation. Impact of Space Height Displacement ventilation is more suitable for high spaces, such as concert halls and workshops (Skistad 1994). Skistad (1989) studied temperature profiles in a concert hall with supply openings under chairs. The temperature rises rapidly from the supply air temperature at the floor to the elevation where people are located. Above the people, there is only a slight temperature gradient up to the elevation where the lights are located. At that level, another temperature jump occurs, which brings the air temperature up to the exhaust air temperature at the ceiling level. Niemela and Koskela (1996) made measurements in a large industrial hall with a height of 90 ft (27 m). Their results show that the temperature increases with elevation in the zone lower than 23 ft (7 m). In the upper zone, the temperature is almost a constant. These measurements confirm again that the vertical temperature gradient is not a constant. Large spaces may be divided into a few zones for determining the temperature distribution. Impact of Diffuser Type Skaret (1985) and Nielsen et al. (1988) investigated the impact of supply diffusers on the temperature distribution. It is better to increase the entrainment of room air so as to decrease the temperature gradient in the occupied zone. The performance of diffusers is critical to avoid draft near the diffusers. Recently, manufactures have developed new products, such as the aspirating diffuser and the modulating diffuser. The performance data can be found from product catalogs. Conclusions The air temperature near the floor and vertical temperature gradient in the occupied zone are two of the most important parameters to evaluate the performance of displacement ventilation in terms of comfort. The ventilation rate or cooling load has a significant impact on the two parameters. It is possible to determine the air temperature near the floor, but it is difficult to calculate the gradient. The type and position of the heat sources and wall heat transfer also influence the air temperature near the floor and the vertical temperature gradient. Some of the contributions have been estimated in previous investigations. However, the estimation sometimes gives erroneous prediction of the vertical temperature gradients because some of the influential parameters are not accounted for. If displacement ventilation is used for space with a high ceiling, it is desirable to use a more sophisticated model.
13
A good diffuser should mix the supply air with the surrounding air quickly to reduce possible draft. It is necessary to develop a universal but simple equation that can put all the results together for design purpose. 2.2 FLOW DISTRIBUTION One important feature in displacement ventilation is to properly control and design the airflow distribution. Proper distribution will ensure good air quality and comfort level in the space. For example, well-designed displacement ventilation can achieve a one-dimensional displacement flow in the occupied zone and transport the contaminants to the upper zone. Both thermal plumes and supply air from diffusers play an important role in the airflow distribution. Impact of Thermal Plumes For proper design of displacement ventilation, it is important to calculate the entrained flow as a function of height. The thermal plume generated by a heated object will increase its volume with the height above the object, as shown in Figure 2.7.
Figure 2.7
A thermal plume above a heated object.
14
Literature Review
According to Baturin (1972), the flow rate, V, at a height, y, from a concentrated heat source in homogeneous surroundings can be expressed by V = 0.005Qc 1/3 (y + y0) 5/3, where y + y0 Qc = distance from a virtual origin of the flow (y0 can be approximated as 2d and d is the hydraulic diameter of the heat source) and = convective heat emission from the source. (2.4)
This equation is valid for y > 2d (Kofoed and Nielson 1990). Qc can be estimated by Qc = k Qt, where Qt is the total energy consumption of the heat source (including convective and radiative). The coefficient k is 0.7 to 0.9 for pipes, 0.4 to 0.6 for small components and 0.3 to 0.5 for large machines (Nielsen 1993). A line source generates a two-dimensional plume, for which Skaret (1986) suggested using the following formula to calculate the flow rate: V = 0.014(Qc /l)1/3 (y + y0) l (2.5)
where l is the length of the source and y0 is one to two times the heat source width. Stabi (1988) presented a list of volume flow rates above different heat sources (including people, machines, windows, and radiators) in homogeneous surroundings. In an environment with temperature stratification, such as a space with displacement ventilation, the air temperatures in the plume and surrounding are identical at a certain level. Higher than this level, no buoyancy force exists in the plume. Therefore, the thermal plume can only reach a maximum height in an environment with temperature stratification. Mundt (1992) found that the flow rate of a thermal plume in a space with a vertical temperature gradient is a little smaller than that without the gradient. The maximum height of the plume is significantly shorter. If the height of a plume is less than the height of the occupied zone (6 ft or 1.8 m from floor), the contaminants within the plume will spread in the occupied zone and cannot reach the upper zone. Therefore, the maximum height of a plume is an important design parameter. Mundt (1992) presented the following equations to calculate the flow rate and the maximum height of a plume in a space with air temperature gradient: V = 0.00238 Qc3/4 s-5/8 (0.004 + 0.039 y1 +0.380 y12 0.062 y13) ymax = 0.98 Qc1/4 s3/8 y0 where y1= 2.86 (y + y0) Qc1/4 s3/8 (2.8) (2.6) (2.7)
15
= flow rate in a plume, m3/s = maximum height of the plume, m = air temperature gradient, K/m = convective heat emission, W = height above the heated object, m = distance between the virtual origin of the plume and the heated object, m Many researchers investigated the plume generated by a sedentary person. Mierzwinski and Popiolek (1981) reported that the convective airflow is in a range of 1.1 to 2.2 ft3/s (30 to 60 l/s) through a section 2.5 ft (0.75 m) above the persons head. According to Clark and Edholm (1985), the flow rate may increase as the surrounding temperature decreases because of the increase of a persons metabolism. Danielsson (1987) showed that the vertical temperature gradient in the air surrounding a person has a strong impact on the flow rate above the person. He also provided a chart for determination of the flow rate above a person. According to the chart, the flow rate at 6 ft (1.8 m) above the floor may decrease from 1.56 ft3/s to 0.74 ft3/s (42 l/s to 20 l/s) when the room temperature gradient is increased from 0.3 to 0.9F/ft (0.5 to 1.5C/m). Fitzner (1989) confirmed the results. Kofoed and Nielsen (1990) further reported that the flow rate may be influenced by not only the temperature gradient but also the ventilation rate. The maximum height of a plume generated by a person is about 6.6 to 10 ft (2 to 3 m), depending on the vertical temperature gradient. The measured flow rates by Mundt (1992) are about twice as much as those measured by Danielsson (1987) and Fitzner (1989). Figure 2.8 shows the flow rate of the plume calculated by Equations 2.6 and 2.7. The calculated results agree reasonably with the measured data (Mundt 1992; Kofoed and Nielsen 1990). The flow rate in a plume generated by a lamp is smaller than that generated by a person, and the flow rate in the plume created by a desk lamp is much smaller than that created by a computer (Mundt 1992), although the energy consumption is at the same level. Figure 2.9 presents the measured data of Mundt (1992) for the volume flow rate above a desk lamp, a fluorescent lamp, and a personal computer. Equations 2.6 to 2.8 may be applied to fluorescent lamps and personal computers. The results seem logical because a point heat source has a smaller area that would entrain much less air than a heat source with a large area. In many cases, a heated object is placed close to a wall. Due to the Coanda effect, a plume close to a wall can be considered as one half of the flow in a free plume with a double convective heat emission 2 Qc. The flow rate of a plume close to a corner is about one quarter of the rate in a free plume with 4 Qc. The plumes generated by a number of sources near each other may form a large plume with a flow rate of about N1/3 V, where N is the number of the heat sources and V is the flow rate in a free plume (Nielsen 1993). V ymax s Qc y y0
16
Literature Review
Figure 2.8
Volume flow rate around and above a person.
Figure 2.9 Volume flow rate above a personal computer and a lamp (Mundt 1992).
17
Impact of Walls Buoyancy will drive airflow up (or down) along a hot (or cold) vertical surface, such as a wall. The flow rate in the turbulent boundary layer may be calculated from (Nielsen 1993) V = 0.0028 Tw2/5 y6/5 l (2.9)
where V = flow rate in the boundary layer, m3/s; Tw = temperature difference between room air and the wall surface, C; y = length measured from the leading edge, m; l = horizontal width of the surface, m. The up or down airflow along a wall is a typical wall jet. Heiselberg (1993) presented a formula to calculate the maximum velocity in the layer. For a modest temperature difference of a few degrees between the wall and room air, the flow along the wall may be as large as that from several heat sources in the room such as people or equipment. A cold downdraft from vertical cold surfaces may cause a stratified flow with a typical wall jet profile near the floor. Heiselberg (1993) measured the profiles and presented formulae to calculate the maximum velocity in the flow and to evaluate the draft risk. The maximum velocity near the wall and the floor is about 40 fpm (0.21 m/s) for a 5 ft (1.5 m) high cold wall with a difference of 18F (10C) between the cold wall surface and the room air. Impact of Diffusers Since relatively cold air is supplied directly to the occupied zone in displacement ventilation, the velocity has to be well controlled to avoid draft. The velocity near a diffuser depends on the flow rate from the diffuser, the temperature difference between the supply and exhaust Te, and the diffuser type. Figure 2.10 shows a typical velocity distribution near a diffuser (Nielsen 1993). Skistad (1994) divided the flow near the floor into two regions: the primary region (where the flow is dominated by the characteristics of the diffuser) and the secondary region (the part outside the primary region). He presented a formula to calculate the maximum velocity in the primary region through (1) the induced ambient airflow volume caused by the dynamics of the jet discharged from the diffuser, (2) the entrained airflow volume caused by the shear in the boundary layer between the supply air and the ambient air, (3) the thickness of the supply air blanket, and (4) the Archimedes number Arh (ghTe/us2). More research is needed to calculate parameters (1), (2), and (3). In the secondary region, Nielsen (1993) presented the following equation to determine the maximum velocity in the center plane umax, x in a distance x from the wall-mounted diffuser:
18
Literature Review
Figure 2.10 A typical velocity distribution near a diffuser (Nielsen 1993).
umax, x = Kdr(h/x)uf where h uf = diffuser height
(2.10)
= face velocity defined as the volume flow rate divided by the face area of the diffuser
Kdr = a function of the Archimedes number that strongly depends on the structure of the diffuser Nielsen (1993) also provides Kdr data as shown in Figure 2.11. The Kdr depends on diffuser tape. The distance from a wall-mounted diffuser to the 40 fpm (0.2 m/s) velocity contour along the center line, ln, is an important parameter. According to ASHRAE Standard 55-1992, the air velocity should be no higher than 50 fpm (0.25 m/s). Skistad (1994) suggested discharging air obliquely to both sides, with some degree of turbulence to reduce ln, and using perforated panels instead of a filter mat to reduce the draft effect. To make ln smaller is a primary goal for diffuser manufacturers. Normally, the manufacturers provide charts to determine ln and velocity distribution near the diffuser in their product catalogs. Figure 2.12 shows an example. The flow from a number of diffusers placed closed to each other on the wall will merge to a two-dimensional flow, in which the velocity is lower than that in the radial flow near a single diffuser (Nickel 1990). However, if the diffusers with oblique discharges are located too close, the discharged flows meet and turn straight into the room, and ln could be several meters (Skistad 1994).
19
Figure 2.11 Measured value for some diffusers (Nielsen 1993).
Figure 2.12 Design chart for choosing a diffuser.
20
Literature Review
Conclusions Stratification height, which is a function of the flow rates of supply air and thermal plumes, is an important design parameter for displacement ventilation. The flow rate at a certain height in a thermal plume can be determined by the heat source type, location, and geometry. The temperature gradient in a space has an impact on the flow rate and the maximum height of a plume. To avoid draft, velocity in the occupied zone, especially near the diffusers, needs to be well controlled. Previous studies provide sufficient information to develop design guidelines. The design charts provided by diffuser manufacturers are also useful. 2.3 CONTAMINANT DISTRIBUTION The advantage of displacement ventilation is that it may provide better indoor air quality in the occupied zone than mixing ventilation. It is therefore important to study the impact of different parameters, such as contaminant source type and location, human body convection, wall surface temperature, and space height, on the contaminant distribution. Impact of Contaminant Source Type and Location Typically, the occupied zone with displacement ventilation has a lower contaminant concentration level than that in the upper zone, as shown in Figure 2.13 (Heisel-
Figure 2.13 Typical profiles of the contaminant concentration vs. different ventilation rates (Heiselberg and Sandberg 1990).
21
Figure 2.14 Concentration profiles with different types of tracer gases (Olesen et al. 1994).
berg and Sandberg 1990). Chen et al. (1988) showed that both the energy and ventilation efficiencies of displacement ventilation are higher than those of mixing ventilation when the contaminant source is combined with a heat source. The ventilation efficiency increases as the ventilation rate increases or the cooling load decreases. Olesen et al. (1994) reported that the concentration distribution depends on the contaminant density (Figure 2.14). However, the amount of contaminants must be sufficiently large to form the density difference. In most measurements using tracer gas technique, the impact of density is negligible. There are cases when the contaminant concentration is not lower in the occupied zone than that in the upper zone. Figure 2.15 presents a measured concentration profile in a room with displacement ventilation and a pollution source located at a low level and outside the thermal plume (Nielsen 1996). In this case, the lower zone has a high contaminant concentration level. Stymne et al. (1991) showed that the contaminant concentration level varies significantly in both the vertical and horizontal directions, depending on the position of pollutant sources related to the thermal plumes. As illustrated in Figure 2.16, the contaminant concentration in the occupied zone is high when the contaminant is
22
Literature Review
Figure 2.15 Concentration profile with pollutant source located at low level and without heat source (Nielsen 1996).
Figure 2.16 Concentration contours in a room with a tracer gas emitted above a 4 W heat source in a low level (Stymne et al. 1991).
23
Figure 2.17 Inhaled air is located below the breathing level. (The measured concentration of the inhaled air at 1.5 m is 0.58ce, instead of 0.96ce (Brohus and Nielsen 1994).
combined with a weak heat source. The thermal plumes are too weak to reach the upper zone. Mundts (1996) measurements showed that the local air quality is good when a tracer gas source is placed above a heat source that produces a plume that can reach the ceiling (a plume that can reach the upper zone should be able to maintain a good air quality in the low zone). When the tracer gas source is placed outside of the thermal plume, the local air quality depends strongly on whether the tracer gas has a positive or negative buoyancy on the room flow pattern. In this case, the occupied zone might have a high contaminant concentration level. The conclusions are similar to those of Stymne et al. (1991). Impact of Convection from Human Bodies Holmberg et al. (1987) found that a free convection flow around a person may protect the breathing zone from surrounding contaminants at the head level, but it may also bring contaminants from the source below the breathing zone. Saeteri (1992) showed that CO2 concentration in the air inhaled is lower than that at the same elevation some distance from the person because the convection flow around the human body brings fresher air from the floor level directly to the breathing zone. This has been confirmed by Murakami et al. (1997) through a detailed computational fluid dynamics simulation. As indicated in Figure 2.17, Brohus and Nielsen (1994) showed that the concentration in the inhaled air is 0.58cethe same as that at 1.7 ft (0.5 m) below the breathing levelalthough the concentration outside the thermal boundary layer around the person at the breathing level is 0.96ce. They found the concentration of inhaled
24
Literature Review
contaminant Ci may be expressed as a linear function of the stratification height, yst, as follows: Ci = Cy (Cy Cf) yst / yb where Cy = concentration outside the thermal boundary layer around the person at the breathing height, yb Cf = concentration at the floor (yst < yb) (2.11)
Impact of Wall Surface Temperature Nielsen (1993) pointed out that the downdraft caused by a cold wall or window may bring polluted air from the upper zone to the lower zone and reduce ventilation efficiency. Skistads (1994) measurements showed that, in a displacement ventilated room with cooled ceiling panels, the contaminant concentration increased quickly in the region from the floor to the elevation of 3 ft (1 m), and the concentration in the breathing zone was almost the same as that near the ceiling. Kruehne and Fitzner (1993) observed a downfall of polluted air from the upper part into the occupied zone when the cooled ceiling panel temperature was low. However, when Niu (1994) placed contaminant sources within thermal plumes in a space with cooled ceiling panels, he found that the concentration profile would be similar to that without the cooled panel if the panel temperature was kept at 68F (20C). Impact of Space Height Many researchers reported that the benefits of displacement ventilation are more likely to be realized in spaces with high ceilings, such as industrial spaces, than those with low ceilings. Skistad (1989) measured the concentration of carbon monoxide emitted by a silicon carbide furnace in a workshop. A clean occupied zone was found in the measurements. Niemela and Koskelas (1996) measurements in a large industrial hall indicated that the concentration of hexavalent chromium in the occupied zone was two or three times lower than that in the upper zone, whereas opposite results were observed for dust. Impact of Other Parameters The distribution of contaminants is sensitive to disturbances in room airflow, such as those caused by the opening or closing of doors and the movement of people. Mattsson and Sandberg (1994) showed that both air change efficiency and contaminant removal effectiveness increase when a person simulator moves forward and backward at velocities less than 60 fpm (0.3 m/s). However, when the velocity increases beyond this point, the efficiency will decrease and the displacement ventilation may instead take the form of mixing ventilation. Brohus and Nielsen (1996)
25
found that the movement of people causes an increase of the concentration of inhaled contaminants due to the disturbance to the free convection flow around people. This flow transports fresh air from the floor level to the breathing zone. Fukao et al. (1996) conducted measurements in two larger offices with different ventilation systems. The results indicated that the air quality with the floor-mounted displacement system is better than that with a ceiling-mounted mixing system, while the thermal environments are almost the same between the two systems. Tanabe and Kimura (1996) measured the mean age of air in an office room with three different ventilation systems. They concluded that a wall-mounted displacement system provides better air quality than a floor-mounted displacement system, and the floormounted system is better than a ceiling-mounted mixing system. Conclusions Contaminant concentration distribution depends on contaminant source type and location and its associated plume strength, etc. Low contaminant concentration may be obtained in the occupied zone when the contaminant source is associated with a heat source and the thermal plume generated by the heat source is sufficiently strong to reach the upper zone. Because the upward free convection around a person brings the air from a lower level to the breathing zone, the inhaled air is cleaner than the air at the same height. Cold walls or cooled ceiling panels may lead to a higher contaminant concentration in the occupied zone because of possible downflow driven by the walls or panels. It is more beneficial to apply displacement ventilation for spaces with high ceilings, if the contaminants are buoyant gases. Prediction of contaminant distribution is more difficult than air temperature and flow distribution. 2.4 COMFORT ASPECTS The primary reason for using displacement ventilation is to achieve a high IAQ environment. However, the ventilation must maintain an acceptable comfort level. Previous investigations showed that large vertical temperature gradient and draft are the two main causes of discomfort with displacement ventilation. To reduce the temperature gradient, the supply flow rate must be increased. This will lead to a high air velocity at the floor level and to a high draft risk. It is also not feasible to increase ventilation rate because of energy concerns. Draft Risk Assessment In a room with displacement ventilation, Wyon and Sandberg (1990) tested sensitivity of 36 male and 36 female subjects to different velocity and temperatures. The percentage of discomfort is summarized in Figure 2.18. The vertical coordinate is the percentage of dissatisfied people and the horizontal coordinate is air temper-
26
27 Literature Review
Figure 2.18 Predicted percentage of discomfort (a) above chair height and (b) below chair height (Wyon and Sandberg 1990).
ature. The results showed that the ankle and foot (below chair height) are more sensitive to air temperature than the rest of the body. At a velocity of 0.6 ft/s (0.2 m/s), fewer than 20% of people will complain of local discomfort in a temperature range of 72F to 81F (22.1C to 27.0C). Skistad (1994) noted that the air velocities in the range between 0.5 and 0.7 ft/s (0.15 and 0.2 m/s) are acceptable for air temperatures of about 68F (20C), and velocities of up to 0.83 ft/s (0.25 m/s) seem acceptable for higher temperatures. Many researchers (Chen 1988; Sandberg and Blomqvist 1989; Kegel and Schulz 1989; Olesen et al. 1994; Akimoto et al. 1995; and Taki et al. 1996) reported that displacement ventilation may generally provide a good thermal comfort environment in various spaces. However, the draft risk at the floor level seems rather high in spaces with displacement ventilation. Melikov and Nielsen (1989) evaluated the thermal comfort condition in 18 displacement ventilated spaces. Within the occupied zone, they found that 33% of measured locations had higher than 15% of dissatisfied people due to draft. Also, 40% of the locations were found to have a temperature difference between head and foot of larger than 5.4F (3.0C). Some measures are available to reduce discomfort level caused by temperature gradient. Glicksman et al. (1996) used low flow rate fans at the floor level to reduce the temperature difference between the ankle and breathing level of a seated person. The measure does not affect the flow in the upper zone in a room with displacement ventilation if the vertical momentum of the fan exhaust is kept low enough. Impact of Cooling Load and Cooled Ceiling Panel Temperature Figure 2.19 shows the range of cooling load per floor area investigated by some researchers. Most of the studies show that the displacement ventilation system can only provide acceptable comfort if the corresponding cooling load is less than about 13 Btu/(hft2) (40 W/m2). With higher ceiling heights, the displacement system is capable of removing larger cooling loads (Skistad 1994). By increasing the area of the air supply outlet (e.g., supplying air through a perforated floor) or by providing additional heat removal capacity (e.g., using cooled ceiling panels), displacement ventilation may be applied to a space with higher cooling load. Olesen et al. (1994) found that no thermal comfort problems existed under the tested conditions with the cooling loads up to 14 Btu/(hft2) (44 W/m2) in a room with a perforated floor. Niu (1994) showed that displacement ventilation combined with cooled ceiling panels may provide a comfort environment at a cooling load up to 16 Btu/(hft2) (50 W/m2). Taki et al. (1996) measured the vertical temperature profiles for four different cooling loads with and without cooled ceiling panels. The results showed a significant influence of the panel temperature on the air temperature distribution in the room. The cooled ceiling panel may create downdrafts in the occupied zone. To avoid it, the surface temperature should be higher than 59F (15C), and the ratio of panel area to ceiling area should be less than a certain amount. A minimum surface temperature is also required to avoid condensation on the panel surface.
28
Literature Review
Figure 2.19 Ranges of cooling load per floor area for three types of displacement ventilation: side-wall diffuser (system 1), side-wall diffuser with cooled ceiling panel (system 2), and rise floor (system 3).
Conclusions Large vertical temperature gradient and draft are the two main causes of discomfort with displacement ventilation. Previous research shows that displacement ventilation without cooled ceiling panels is suitable for spaces with a cooling load less than 13 Btu/(hft2) (40 W/m2). However, the current study, as shown in Chapter 6, indicates that the upper limit is much higher. With cooled ceiling panels, displacement ventilation can remove a cooling load of 16 Btu/(hft2) (50 W/m2). It is important that the surface temperature of the panels should not be lower than 59F (15C). Low flow rate fans at floor level may reduce the vertical temperature difference and extend the application range. 2.5 ENERGY AND COST ANALYSIS Annual energy consumption, first costs, and operation and maintenance costs over a life cycle are important criteria for evaluation of a ventilation system. Almost all of 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. Energy Analysis Seppanen et al. (1989) evaluated the energy performance of displacement ventilation systems and mixing ventilation systems in U.S. office buildings. The study is
29
for south, north, and core zones with four representative U.S. climates (Minneapolis, Seattle, Atlanta, and EI Paso). They compared different control strategies, such as variable-air-volume system and constant-air-volume system, and systems with different components, such as recirculation, economizer, and heat recovery device. The energy consumption was found to depend very much on the control strategies and air-handling systems, as shown in Figure 2.20. The energy consumed by displacement systems with heat recovery and variable-air-volume flow control is similar to that of mixing systems. Seppanen et al. (1989) used an average cooling load of 4.4 Btu/(hft2) (14 W/m2), and the maximum load is 7.5 Btu/(hft2) (24 W/m2) in the core space of U.S. office buildings (Table 2.1). Since the core region does not need heating, application of displacement ventilation is particularly attractive. However, the load is much higher in the perimeter region. The cooling load in the perimeter seems too high to use a displacement ventilation system. According to the results shown in Table 2.1, the traditional displacement ventilation system can only be used in the north zone of buildings in Seattle. For the rest, the cooling loads are much higher than the traditional displacement ventilation system can handle. Chen and Kooi (1988) pointed out the significant impact of the vertical temperature gradient on energy consumption in a room with displacement ventilation when
Figure 2.20 Comparison of annual energy cost of different systems to the same costs for system 1 (VAV mixing system) in the Minneapolis climate (Seppanen et al. 1989) (Mmissing ventilation, Ddisplacement ventilation).
30
Literature Review
Table 2.1 Heating and Cooling Loads for Each Location and Representative Zone in the U.S.
Atlanta Max. North zones Heating (Btu/hft2) Heating (W/m2) Cooling (Btu/hft2) Cooling (W/m2) Core zones Heating (Btu/hft2) Heating (W/m2) Cooling (Btu/hft2) Cooling (W/m2) South zones Heating (Btu/hft2) Heating (W/m2) Cooling (Btu/hft2) Cooling (W/m2) 19.6 62.0 22.2 70.0 0.0 0.0 7.5 23.6 19.2 60.5 37.6 118.7 Ave. 5.2 16.3 10.1 31.9 0.0 0.0 4.4 14.0 5.0 15.8 14.1 44.5 El Paso Max. 16.8 53.1 24.6 77.5 0.0 0.0 7.5 23.6 16.8 53.0 41.4 130.5 Ave. 4.3 13.6 11.7 36.9 0.0 0.0 4.4 14.0 4.7 14.8 17.6 55.6 Minneapolis Max. 14.1 44.5 17.5 55.3 0.0 0.0 7.5 23.6 16.0 50.3 36.9 116.4 Ave. 4.6 14.4 7.8 24.6 0.0 0.0 4.4 14.0 5.0 15.9 12.4 39.0 Seattle Max. 9.1 28.7 15.7 49.5 0.0 0.0 7.5 23.6 9.1 28.6 35.9 113.2 Ave. 3.2 10.0 7.1 22.3 0.0 0.0 4.4 14.0 3.3 10.3 11.3 35.7
Table 2.2
The Costs of Annual Energy Consumption
Ventilation Supply Air Temperature Energy Consumption Systems Air-Handling Systems F (C) ($/m2) Displacement Mixing Mixing Displacement Mixing Variable air volume Variable air volume Variable air volume Constant air volume Constant air volume 61 (16) 55 (12.5) 61(16) 108 104 126 241 222
they analyzed a Dutch office with different ventilation systems. The energy consumption of displacement ventilation can be either smaller or larger than that of mixing ventilation, as shown in Table 2.2, depending on the control strategies and the HVAC systems. The conclusions are similar to those of Seppanen et al. (1989), although the approaches and weather data are different between the two investigations. Nius (1994) calculation showed that the annual energy consumption of displacement ventilation with a water-cooled ceiling system is almost the same as that of an all-air system. His investigation used a variable-air-volume system.
31
Zhivov and Rymkevich (1998) compared the energy consumption between displacement and mixing ventilation systems for a restaurant in different U.S. climates. They found that the displacement ventilation can save up to 50% of cooling energy but may increase heating energy. Previous studies show that both the supply air temperature and the exhaust temperature in displacement ventilation are higher than those of mixing ventilation. The air temperature difference between the supply and the exhaust is nearly the same between the two ventilation systems. According to Skistad (1994), the temperature difference for displacement ventilation can be larger for high spaces and, therefore, supply airflow rate can be reduced considerably. Note that displacement ventilation may use more natural cooling, since the supply air temperature is 4F to 6F (2C to 3C) higher than that of mixing-type ventilation. First Cost Analysis Seppanen et al. (1989) found that the first cost of a system is difficult to estimate. They compared different air-handling systems, such as a variable-air-volume system and a constant-air-volume system, and systems with different components, such as re-circulation, economizer, and heat recovery device. Figure 2.21 shows that the first costs of displacement systems are substantially higher than those of mixing systems when cooled ceiling panels are required. Without cooled ceiling panels, the costs of the displacement system are similar to those of mixing system. Skistad (1994) also
Figure 2.21 Comparison of the first cost of different systems to the same costs for system 1 (VAV mixing system) in the Minneapolis climate (Seppanen et al. 1989) (Mmixing ventilation, Ddisplacement ventilation).
32
Literature Review
reported that there is no significant first cost difference between the two systems, except that the cost of diffusers in the displacement ventilation is higher than that in mixing ventilation. Conclusions There are not many publications in the literature concerning energy and cost analysis for displacement ventilation. Energy consumption varies significantly with climatic regions. Compared with mixing ventilation, displacement may or may not save energy. The energy consumption depends on the control strategies and airhandling systems. In U.S. offices, displacement ventilation is ideal for core zones. Previous studies show that it may not be appropriate for perimeter zones in most regions of the U.S. because the cooling load is too high. If cooled ceiling panels are used, the first cost of displacement ventilation is much higher than that of mixing ventilation even though the system may not be able to remove high cooling loads found in the south zone of U.S. office buildings. Displacement ventilation combined with radiators is used in Europe for winter heating. However, the U.S. uses different heating methods, and few studies are available. 2.6 DESIGN GUIDELINES According to the analysis in the previous sections, the following parameters are most important in design of the displacement system: Supply airflow rate and temperature Air temperature at floor level Vertical temperature gradient Maximum air velocity at floor level Stratification height (lower zone height) or contaminant concentration gradient Energy consumption First costs and maintenance costs
The most complete design guidelines available are those developed by Skistad (1994). He used a five-step approach: 1. 2. 3. Determine the required airflow rate for removal of surplus heat based on the cooling load and the air temperature difference between supply and exhaust openings. Find the required airflow rate for removal of pollutants according to ventilation standards. Choose the larger of the two flow rates determined at Steps 1 and 2 as the ventilation rate.
33
4. 5.
Determine supply air temperature under assumptions of f = 0.5 and constant vertical temperature gradient. Choose supply diffusers according to the data provided by manufacturers in order to avoid draft.
A more recent version of the deign guide can be found from the REHVA guidebook (REHVA 2002). Despite the simple design guidelines, there are problems. Figure 2.2 shows that f varies from 0.2 to 0.7, and the vertical temperature gradient is not a constant. If the actual Tf < 0.5 Te, the vertical temperature gradient will be larger than the expected gradient. The design guidelines assume Tf = 0.5 Te. If the actual Tf > 0.5 Te, the selected V based on Tf = 0.5 Te will be larger than needed. A different approach proposed by Chen et al. (1991) used a design atlas. The atlas, based on experimental and computational results, contains detailed information on indoor airflow, indoor air quality, and thermal comfort for various configurations of spaces. Unfortunately, the atlas at present does not cover a wide range of spaces and conditions. It seems necessary to improve current available design guidelines for the displacement ventilation system to ensure good indoor air quality and thermal comfort in the space. From the above review and a survey among architects and energy consumption in the U.S. (Chen et al. 1999), we may conclude that design guidelines available in the literature cannot be used with confidence. Many assumptions need further clarification so that designers can use the design guidelines with confidence. U.S. buildings, especially perimeter zones of such buildings, have a high cooling load. Design of displacement ventilation must address these zones. The special features in U.S. buildings considered in this book include offices, classrooms, and industrial workshops. For these types of spaces, this book will present experimental tests used to obtain reliable data on the performances of displacement ventilation. A CFD program was validated against these data. By using the program, this book shows how to conduct numerical simulations of a large number of cases of displacement ventilation in the three types of spaces and how to establish a database on the performances. Based on the database, this book further presents a model for prediction of temperature difference between head and foot levels and a model for ventilation effectiveness for design purpose. The book will also compare the energy and first costs of the displacement ventilation system with a mixing ventilation system. Finally, the book presents design guidelines for the displacement ventilation system developed from the study. These will be discussed in the following chapters.
34
CHAPTER 3
Experimental Study and Validation of CFD Program

Displacement ventilation may provide better indoor air quality and save energy, but there is a question of the usefulness of this technology in U.S. buildings with higher cooling requirements. A first step in preparing a design guideline is careful study of displacement ventilation for several typical U.S. buildings. Two main approaches are available for the study of airflow and pollutant transport in buildings: experimental investigation and computer simulation. Experimental investigation, although it is reliable, is very expensive and time consuming. Computer simulation is inexpensive, but it may not be reliable. For evaluation of the indoor environment provided by displacement ventilation, the computational fluid dynamics (CFD) technique seems most appropriate, if it is validated by experimental data. This combined approach was used to generate the results presented in this book. Many experimental data are available in the literature, but very few of them can be used for the validation. Experimental data for CFD validation must contain detailed information of flow and thermal boundary conditions as well as flow and thermal parameters measured in the space. The data must also include an error analysis. Unfortunately, not many of the experimental data include such detailed information. Popular data for validating room airflow are from Cheesewright et al. (1986) and Nielsen et al. (1978). Cheesewright's data are for natural convection and Nielsen's data are for forced convection. However, it is still not certain that a CFD program validated by their data can be used for normal room airflow with mixed convection (a combination of natural and forced convection). This chapter presents detailed experimental data for displacement ventilation, which is mixed convection and represents ventilation reality in many buildings. The experimental data are used to validate a CFD program with a suitable turbulence model. There are many turbulence models available. The standard k- model (Launder and Spalding 1974) is probably most widely used in engineering calculations due to its relative simplicity. However, the model sometimes provides poor results for indoor airflow. Many modifications have been applied to the standard
35
model. However, the modified models do not have a general applicability for indoor airflow. Chen (1995 and 1996) calculated the various indoor flows with eight different turbulence models. His study concluded that the re-normalization group (RNG) k- model (Yokhot et al. 1992) is the best among the eddy-viscosity models tested. This chapter will also compare this models prediction for displacement ventilation in a room with the experimental data. 3.1 EXPERIMENTAL FACILITY The Chambers and HVAC Systems An environmental test facility was used to study indoor air quality, thermal comfort, energy efficiency, and ventilation systems. The test facility, shown in Figure 3.1, consists of a well-insulated enclosure. Not shown in the figure are two doors at either end. A movable wall divides the enclosure into a test chamber and a climate chamber. We use the larger one as the test chamber and the smaller one as the climate chamber. The lower part of the movable wall is an insulated exterior wall, and the upper part is a double-glazed window extending almost the whole width. Table 3.1 lists the dimensions and thermal resistance of the chambers. Each chamber has a separate HVAC system. The two systems are nearly identical. Table 3.1 also lists the capacities of the HVAC systems. Figure 3.2 illustrates the HVAC system configuration and control interface. Three louvers control outdoor air rate between 0% and 100%. The supply fan and return fan have a variable speed drive. The interface allows an interactive control of the systems. An operator can change any parameter, such as airflow rate, supply and return temperature, and humidity. The air parameters in different sections of the HVAC systems are shown in the monitor and/or are written into a file in the time interval specified by the oper-
Figure 3.1
Sketch of the test facility.
36
Table 3.1 Dimension, Thermal Resistance, and HVAC System Capacity of the Test Facility
Length Width Dimension
Height Length Width Height Partition height window width
Test Chamber 17 ft (5.16 m) 12 ft (3.65 m) 8 ft (2.43 m) 3.9 ft (1.16 m) 11.5 ft (3.45 m) 30 ft2hF/Btu (5.3 Km2/W) 1.5 ft2hF/Btu (0.27 Km2/W) 30 ft2hF/Btu (5.3 Km2/W) 30 ft2hF/Btu (5.3 Km2/W) 30 ft2hF/Btu (5.3 Km2/W) 30 ft2hF/Btu (5.3 Km2/W) 8 kW 560 cfm (930 m3/h) 21 kW 8 kW 11 kgsteam/h 560 cfm (930 m /h) 560 cfm (930 m3/h)
3
Climate Chamber 10 ft (3.08 m)
Thermal resistance
Partition wall Partition window Other walls Ceiling Floor Door
Capacity of Preheater HVAC system Supply fan Chiller Reheater Humidifier Return fan Dampers
none
ator. Nine-probe hot-wire anemometers, arranged in a matrix form, are used to measure the airflow rate. The HVAC and control design allows a variable air supply rate ranging between 1 ach to 20 ach for the test chamber and 2 ach to 40 ach for the climate chamber. Equipment The major measuring equipment of the test facility includes the following: A flow visualization system for observing airflow patterns A hot-sphere anemometer system for air velocity, velocity fluctuation, and temperature measurements
37
Figure 3.2
The control interface of the HVAC system.
A tracer-gas system for measuring contaminant concentrations and humidity A thermocouple system for measuring surface and air temperatures
The test chamber has two long slots and several observation windows on the walls. Light penetrates through the slots and forms a thin light sheet. By injecting smoke into the room, the airflow pattern can be observed through those observation windows normal to the light sheet. The walls of the room were painted black to enhance visualization. The test facility has two types of smoke source: a theater fog generator and an air current kit. The theater fog generator produces smoke with a temperature far above the environmental temperature. The smoke is cooled through a long tube before entering the room to approach neutral buoyancy. This generator produces a large amount of smoke and is good for observing how supply air is distributed in the room. However, it is not ideal for visualizing airflow pattern in a particular area. The air current kit remedies this problem. The kit generates a very small amount of smoke locally. Since there is no effective technique at present to measure low air velocities throughout an indoor space, the experiment used hot-sphere anemometers for the measurements of air velocity, velocity fluctuation, and temperature in the room. An ADD board was used for data acquisition. For velocity, the measurement range of the hot-sphere anemometers is 10 to 1000 fpm (0.05 to 5 m/s); the repeatability is 2 fpm (0.01 m/s) or 2% of the readings. The anemometers cannot reliably measure velocity when the magnitude is lower than 20 fpm (0.10 m/s). The measuring errors
38
for air temperature are 0.8F (0.4 K), including the errors introduced by the data acquisition systems. Since the probe size is large (about 1/8 in. or 0.003 m in diameter), the probes are not sensitive to high-frequency velocity fluctuation. It is difficult to estimate the errors for velocity fluctuation. A multi-gas monitor and analyzer system is used to determine indoor air quality. The tracer-gas system can measure many different types of tracer gases. This investigation used SF6 and CO2. SF6 is better than CO2 because the background concentration of SF6 is almost zero. CO2 is inexpensive and was used to check the results obtained with SF6. In addition, we also measured the water vapor concentration for determining relative humidity in the room. The error for measuring concentration is at an acceptable 10%. We also used thermocouples to measure air temperature and surface temperature of the room enclosures. The error for measuring temperature by the entire system is about 0.8F (0.4 K). 3.2 TEST PROCEDURE We conducted several measurements with different configurations: a small office, a large office with partitions, and a classroom. Figure 3.3 shows the configurations used in the experiment. The internal heat sources and the ventilation rate used in the experiment are listed in Table 3.2. These cases present some typical situations in the U.S. but they do not cover a wide range. The main purpose is to obtain experimental data to validate a CFD program. A perforated displacement diffuser, 1.7 ft (0.5 m) wide and 3.6 ft (1.1 m) high, was placed at the middle of the right side wall near the floor in the small office. The effective area ratio is 10%. The exhaust, 1.4 ft 1.4 ft (0.43 m 0.43 m), was at the center of the ceiling. The occupants in the test room were simulated by boxes 1.3 ft (0.4 m) long, 1.2 ft (0.35 m) wide, and 3.6 ft (1.1 m) high, each heated by three 25 W light bulbs. The measured surface temperature was between 82F and 86F (28C and 30C). Two point sources of SF6 were introduced at the top of the two boxes to simulate contaminants from the occupants, with an initial velocity of 9 fpm (0.045 m/s in the horizontal direction). Two PCs were used to generate heat. One generated 108 W and the other 173 W. Six 34 W fluorescent lamps were used during the experiment as overhead lighting. In addition, two tables and two file cabinets were also in the room. The ventilation rate was 4 ach, corresponding to a face velocity of 18 fpm (0.09 m/s) at the diffuser. The ventilation rate is very high compared to that used in Scandinavia. This is because U.S. buildings have a much higher cooling load based on a survey in Boston (Chen et al. 1999). A lower ventilation rate will require a lower supply air temperature. This will lead to draft at ankle level. The supply air temperature corresponding to 4 ach was controlled at 62.5F (17.0C). The window surface temperature was 81.1F to 82.6F (27.3C to 28.1C), and the surface temperature on the
39
(a)
(b)
(c)
Figure 3.3
Space layout used in the experiment: (a) small office, (b) large office with partitions, and (c) classroom.
40
Table 3.2
Specifications for the Experimental Measurements

I-P Units Internal Persons Equipment Lighting Load (number) (Btu/hft2) (Btu/hft2) (Btu/hft2) 2 2 2 2 6 6 4.7 4.7 4.7 4.7 2.9 0 3.4 3.4 3.4 3.4 3.4 3.4 SI Units Persons Equipment Lighting (number) (W/m2) (W/m2) 2 2 2 2 6 6 14.9 14.9 14.9 14.9 9.2 0 10.8 10.8 10.8 10.8 10.8 10.8 Internal Load (W/m2) 33.7 33.7 33.7 33.7 43.9 34.7 Ventilation Rate (ach) Climate 4 4 8 4 8 4 Summer Winter Summer Winter Summer Winter 10.7 10.7 10.7 10.7 13.9 11.0 Ventilation Rate (ach) Climate 4 4 8 4 8 4 Summer Winter Summer Winter Summer Winter
Case 1 2 3 4 5 6
Type Small office Large office classroom
Case 1 2 3 4 5 6
Type Small office Large office classroom
movable wall was 75.6F to 79.9F (24.2C to 26.6C). The surface temperatures on the other walls were 73.9F to 78.8F (23.3C to 26.0C). Five movable poles were placed in the test room, and each supported six hotsphere anemometers and six air-sampling tubes. Additionally, two thermocouples were also attached on each pole to measure air temperature near the floor and ceiling. A total of 40 thermocouples were used to measure the surface temperatures of the floor, ceiling, window, and walls. Measurements were conducted under steady-state conditions by stabilizing the room thermal and fluid conditions for more than 12 hours before recording the data. Air velocity, air temperature, and SF6 concentration were measured in nine different positions with a total of 54 measuring points for air velocity, 72 for air temperature, and 54 for tracer gas. The measured air velocities can be used for determining turbulence intensity, and the measured concentrations can be used to find the local mean age of air. Since
41
the omnidirectional anemometers have a large uncertainty in measuring low air velocity, we feel the fluctuating velocity, |u|, could provide more accurate information than turbulence intensity. This investigation used the step-up and decay method to determine the mean age of air, :
=
c(t) c(0) - dt 1 ---------------------------c() c(0)
( step-up )
(3.1)
= ------------------------------------------c(0) c()
0 [ c ( t ) c ( ) ] dt
( decay )
(3.2)
where c(t), c(0), and c() are the tracer-gas concentration measured at time = t, 0, and infinity, respectively. 3.3 EXPERIMENTAL RESULTS Table 3.3 lists the measured air temperatures and concentrations at the supply inlet and exhaust and the surface velocity at the supply diffuser. 3.4 COMPUTATIONAL FLUID DYNAMICS MODEL With the detailed experimental data, we can validate a CFD program. A validated CFD program can then be used to establish a database to evaluate the performance of displacement ventilation and to develop design guidelines for displacement ventilation. The CFD program used in the present investigation is a commercial program (CHAM 1996) that uses a renormalization group k- model (Yokhot et al. 1992). Table 3.3
Ts Case 1 2 3 4 5 6 Type Small office Large office classroom (F) 62.6 63.0 62.8 62.2 67.6 66.2 (C) 17.0 17.2 17.1 16.8 19.8 19.0 (F) 80.1 75.4 73.6 75.9 76.3 78.4
The Measured Parameters

Te (C) 26.7 24.1 23.1 24.4 24.6 25.8 cs ce us Tracer Gas (ppm) (ppm) (fpm) (m/s) SF6 CO2 CO2 CO2 CO2 CO2 0 562 520 493 477 432 0.42 725 602 657 723 924 18 18 36 18 37 20 0.09 0.09 0.18 0.09 0.19 0.10
42
The Re-Normalization Group (RNG) k- Model The governing equations for the RNG k- model are
---- ( ) + ------ ( u j ) = ------ ------ + S , xj x j x j t
(3.3)
where t xj S = time; = air density, kg/m3; = 1 for mass continuity; = uj (j = 1, 2, and 3) for three components of momentum; = k for kinetic energy of turbulence; = for dissipation rate of turbulence energy; = T for temperature; = c for contaminant concentration; = coordinate; = source term. The , and S are further listed in Table 3.4. Boundary Conditions Since the RNG k- model is valid for high Reynolds number turbulent flow, wall functions are needed for the near wall region where flow Reynolds number is low. The present investigation uses the following wall functions (Launder and Spalding 1974): For velocity:
U = -
1 21
,eff = effective diffusion coefficient; and
y -E -- log --- y
(3.4)
where U y E y* = velocity parallel to the wall = wall shear stress = von Karman constant (0.41) = distance between the first grid node and the wall = an integration constant (9.0) = a length scale
43
Table 3.4
1 ui 0 + t ( + t) /k ( + t)/ /l + t /t ( + t)/c
Values of , , and S
S 0 p - gi ( T T0 ) ----- xi G +GB (C1G C2 + C3GB) / k + R ST Sc
T C
where is laminar viscosity k - is turbulent viscosity t = C --- ui ui uj G = t ------- ------- + ------- is the turbulent production x j x j x i t T - is the turbulent production due to buoyancy G B = g i ------- -----Pr t x i C ( 1 0 ) 2 - is the source term from renormalization - ---R = ---------------------------------------3 k 1 + k 1 u i u j - , S = (2Sij Sij)1/2, S ij = -- ------- + ------ = S 2 x j x i C =0.0845, C1 = 1.42, C2 = 1.68, C3 = 1.0 are the model constants k = 0.7194, = 0.7194, l =0.71, t = 0.9, c = 1.0 are Prandtl or Schmidt numbers
3 2
For kinetic energy of turbulence:

1 - -k = ----------1 2 C
(3.5)
For dissipation rate of turbulent kinetic energy:

= -
32
1 ----y
(3.6)
For temperature:
q = hc ( Tw T )
(3.7)
44
where q hc Tw = heat flux = convective heat transfer coefficient = wall temperature
We use Equation 3.7 for temperature boundary instead of the standard wall function proposed by Launder and Spalding (1974). This is because the wall function would predict grid dependent heat flux and cause an unacceptable error. The hc used is the one based on our experimental data (Chen et al. 1989). This is undesirable in numerical prediction because the hc is generally unknown. We have developed a new one-equation model for the near wall flow (Xu 1998). The heat transfer can be correctly calculated with the new model. Numerical Technique The governing equations are solved numerically. The whole computational domainthe space of the roomneeds to be divided into a number of finite volumes by a grid system. The flow variables, such as velocity, temperature, and concentration, are solved at the center of each finite volume. The more grids used, the more accurate the results will be. However, a fine grid will cost more computing time and capacity. Table 3.5 shows the computing time for three different grid sizes in an SGI R-5000 workstation. The difference between the results with two finer grids is very small. Therefore, we use a 48 44 24 grid for the comparison with the experimental data. A commercial CFD program (CHAM 1996) was used for the computations. By default, the code uses the finite-volume method and the upwind-difference-scheme for the convection term. The convergence criterion was set such that the respective sum of the absolute residuals of p, ui, T, c, k, and must be less than 10-3. 3.5 VALIDATION OF CFD PROGRAM We validated the CFD program by comparing the flow patterns, vertical profiles of temperature, concentration, velocity, and turbulence intensity between measured data and computed results for a small office, a cubicle office, a quarter of a classroom, and a workshop. Table 3.5
Grid Size 72 66 36 48 44 24 29 30 19
Grid Refinement
CPU Time (hour) 180 45 13
45
Figure 3.4
The airflow pattern observed by using smoke visualization (left figure) and computed by the CFD program (right figure) in the midsection. The length of the arrows is proportional to the velocity magnitude.
Small Office The validation uses Case 1 listed in Table 3.2 as the small office. Figure 3.4 shows the flow pattern, observed by using smoke and computed by the CFD technique in the midsection through the diffuser. The velocity determined from the smoke visualization is rather reliable because the speed is low. The computed flow pattern agrees with the observed one. Due to buoyancy, the cold air from the supply diffuser spreads on the floor level. This cold airflow is like a jet and induces the surrounding air. As a result, the induction causes a reverse flow in the layer between 1.5 to 3 ft (0.5 to 1 m) above the floor. The figure does not show the thermal plumes generated by occupants and computers because they are in a different section. Figures 3.5 through 3.8 present, respectively, the measured and computed temperature, SF6 concentration, velocity, and velocity fluctuation in the office. The measurements were done with nine poles; each pole had ten sensors to measure temperature and six sensors to measure velocity and tracer-gas concentration. The vertical axes are dimensionless elevation normalized by room height (Z = 0 is the floor and Z = 1 is the ceiling). The horizontal axes are dimensionless measured parameters. Figure 3.5 clearly shows that the displacement ventilation system created temperature stratification. The temperature gradient in the lower part of the office is much larger than the one in the upper part because most heat sources (occupants and computers) are located in the lower part of the room. Since occupants stay in the lower part of the room, the temperature stratification represents a potential risk of draft. One important criterion in design of a displacement ventilation system is to ensure that the temperature difference is sufficiently small between the head and foot level. The agreement between the computed temperature and measured data is excellent. Figure 3.6 shows the tracer-gas concentration profiles in the room. The tracergas sources were introduced at the head level of the two occupants (the two small
46
47
Figure 3.5
Comparison of the vertical temperature profiles for a small office; = (T Ts) / (Te Ts), Ts = 62.6F (17.0C), Te = 80.1F (26.7C); Z = z / h, h = 8 ft (2.43 m).
48
Figure 3.6
Comparison of the concentration of SF6 for a small office; C = (c cs) / (ce cs), cs = 0 ppm, ce = 0.42 ppm; Z = z / h, h = 8 ft (2.43 m); symbols: measurement, lines: computation.
49
Figure 3.7
Comparison of the velocity profiles for a small office; U = u / us, us = 18 fpm (0.09 m/s); Z = z / h, h = 8 ft (2.43 m); symbols: measurement, lines: computation.
50
Figure 3.8 Comparison of the fluctuating velocity for a small office; U = u / u s , us = 18 fpm (0.09 m/s); Z = z / h, h = 8 ft (2.43 m); symbols: measurement, lines: computation.
squares in the right bottom figure). The SF6 concentration in the occupied zone is much lower than that in the upper zone. The concentration increases rapidly between Z of 0.4 and 0.5, which can be considered as the stratification height. Since convective flow around the human body may bring the air at the lower level to the breathing level, displacement ventilation provides better indoor air quality than mixing ventilation. There are discrepancies between the computed concentration profile and the measured data. Since the tracer gas is a point source and recirculating flow exists in the upper part of the office, the tracer-gas concentration in the upper part is not uniform and very sensitive to the position and boundary conditions. For example, the results for Pole 9 show the difference of the concentration distributions between two positions that are only 1 ft (0.3 m) apart. Nevertheless, the accuracy of the computation is acceptable. Figure 3.7 shows that the velocity in most of the space, except near the floor, is lower than 10 fpm (0.05 m/s). The magnitude is so low that the hot-sphere anemometers may fail to give accurate results. Nevertheless, the measured velocity is close to that observed through the use of smoke, and the computed results agree well with the data. The velocity near the floor is larger than that in the center of the room because the diffuser is installed on the floor level. Draft risk exists in the near diffuser area, as indicated in pole 1, where the air velocity is high and the air temperature is low. The experiment also measured velocity fluctuation. Figure 3.8 shows the normalized fluctuating velocity by the mean supply air velocity, instead of the local mean velocity, in order to avoid additional uncertainty of the low local mean velocity. The computed profiles in the figure are the quantities of 2 k , where k is the turbulent kinetic energy. The computed values are larger than the measured ones. This might be due to the errors introduced by the anemometers we used for measurement of the fluctuating velocity. Because of the large probe size, 1/8 in. (3 mm) in diameter, the anemometers may not able to measure the high-frequency velocity fluctuations. Furthermore, the turbulence model may not accurately calculate turbulent energy. Therefore, it is not surprising to see the large discrepancies between the computed profiles and measured data. Figure 3.9 shows the measured transient CO2 concentration at the middle of pole 4. From the measured data, we have calculated the mean age of air in the office. Figure 3.10 compares the computed (contours) and measured (values in boxes) mean age of air in the midsection of the office. Table 3.6 further shows that the measured mean age of air is 10% smaller than the computed one. The mean age of air is younger in the lower part of the office with displacement ventilation than that in the upper part. This is why displacement ventilation can provide better indoor air quality in the occupied zone, that is, in the lower part of the space.
51
Figure 3.9
Typical concentration history.
Figure 3.10 The computed (contours) and measured (values in boxes) mean age of air in the midsection of the office (second).
The ASHRAE report for RP-949 (Chen et al. 1999) provides more results of the validation for the office under a winter condition, a cubicle office, a classroom, and an industrial workshop. The results are similar to those shown in this section. 3.6 CONCLUSIONS The experimental data of displacement ventilation for a small office, a large office with partitions, and a classroom are available for validation of CFD programs. The data include detailed information, such as
52
Table 3.6 Comparison of the Mean Age of Air Between CFD Computation and Measurement
Position Point 1 2 3 4 5 6 7 x (ft [m]) 5.8 (1.74) 5.8 (1.74) 5.8 (1.74) 12.2 (3.66) 12.2 (3.66) 12.2 (3.66) 8.6 (2.58) y (ft [m]) 6.3 (1.88) 6.3 (1.88) 6.3 (1.88) 6.3 (1.88) 6.3 (1.88) 6.3 (1.88) 6.3 (1.88) z (ft [m]) 0.5 (0.15) 2.2 (0.65) 5.2 (1.55) 0.5 (0.15) 2.2 (0.65) 5.2 (1.55) 8 (2.43) Mean Age of Air CFD (s) 575 728 1083 600 683 1166 1087 Data (s) 511 609 900 556 621 888 966
thermal and flow boundary conditions; airflow patterns observed by using smoke; the distributions of air temperature, concentration of tracer gas, air velocity, and velocity fluctuation; and the mean age of air.
The experimental data include error analysis for the measuring equipment. A CFD program with the RNG k- model of turbulence has been used to predict the airflow pattern, the distribution of the air temperature, concentration of tracer gas, air velocity and velocity fluctuation, and the mean age of air in a small office, a large office, a classroom, and a workshop. The computed air temperature and velocity agree well with the measured data. However, some discrepancies are found between the computed and measured tracer-gas concentration. The agreement is less satisfactory between the computed and measured distributions of velocity fluctuation. The discrepancies between computed and measured mean age of air are about 10%. Despite the discrepancies, the CFD program can be used for simulation of airflow in a room with displacement ventilation.
53
CHAPTER 4
Models for Prediction of Temperature Difference and Ventilation Effectiveness

As reviewed in Chapter 2, displacement ventilation generally provides an acceptable comfort level in the room. However, a risk of draft exists at the floor level because of the high air velocity and low air temperature. In addition, the temperature difference between the head and foot level may be too large due to the vertical temperature stratification. Melikov and Nielsen (1989) evaluated the thermal comfort condition in 18 displacement ventilated spaces. Within the occupied zone, they found that 33% of measured locations had higher than 15% dissatisfied people due to draft. They found that 40% of the locations had a temperature difference between head and feet larger than 5F (3 K), the limit defined by ASHRAE Standard 55-1992 (ASHRAE 1992). Obviously, these displacement ventilation systems were not properly designed. The computational fluid dynamics (CFD) technique and a full-scale experimental rig can be used to determine the temperature and velocity distribution in a room with displacement ventilation. Nevertheless, the tools are not generally available for most designers. The designers need a simple model to predict the temperature difference between the head and foot level. It seems difficult to predict the nonlinear vertical temperature profile. This is because many parameters, such as ventilation rate, heat source type and position, wall temperature and wall radiative characteristics, space height, and diffuser type, contribute to it (Yuan et al. 1998). Actually, it is not necessary to predict the whole vertical profile. Only the air temperatures between the head and foot level are required in the design of comfort conditions. If the air temperature at the head level of a sedentary person is defined as the room design temperature, it is determined by comfort and is known in the design. If we can establish an accurate model to calculate the air temperature difference between the head and foot level, we can calculate the air temperature near the floor, Tf. With Mundt's model relating Ts and Tf (Equation 2.1) and the steady-state room energy balance equation,
Qt = Cp V ( Te Ts ) ,
(4.1)
55
we can determine the air supply temperature Ts and exhaust temperature Te. Therefore, a model of the air temperature difference between the head and foot level is important to design a displacement ventilation system. On the other hand, the primary purpose of displacement ventilation is to improve indoor air quality. A designer needs a model to estimate the ventilation effectiveness in a room with displacement ventilation. However, the literature review and our studies show that the distribution of contaminant concentration and ventilation effectiveness are strongly influenced by the position of the contaminant sources and the heat sources. Therefore, no general model for the ventilation effectiveness at the breathing level is available at present. Displacement ventilation has higher ventilation effectiveness than mixing ventilation. The required amount of fresh air can be reduced for displacement ventilation to save energy. Most designs use an assumption of complete mixing to estimate the amount of fresh air needed. Such a calculation would lead to a substantial error. Therefore, the objective of this study is to develop simple models to estimate the air temperature difference between the head and foot level and the ventilation effectiveness at the breathing level. 4.1 A DATABASE OF DISPLACEMENT VENTILATION To develop models to estimate the air temperature difference between the head and foot level and the ventilation effectiveness at the breathing level, information about the air temperature and contaminant distributions in rooms with displacement ventilation are needed. The air temperature and contaminant distributions are needed for a large number of cases to develop an accurate simplified model. This requires a database of the air temperature and contaminant distributions for rooms with various kinds of geometrical, thermal, and flow boundary conditions. There are two approaches to establish a database of air temperature and contaminant distributions: direct measurements and numerical simulations. Direct measurements in rooms with different geometric, thermal, and flow boundary conditions give the most realistic information. However, they are very expensive and time consuming for many difficult cases. The control of thermal and flow boundary conditions is also difficult. The use of numerical simulations seems a good choice at present. To establish a large database, the present investigation uses the CFD technique to simulate the air temperature and contaminant concentrations in different rooms with displacement ventilation. The CFD program has been validated by comparison to seven sets of measured data from a small office, a large office with partitions, a classroom, and an industrial workshop, as presented in the previous chapter. In this chapter, the validated CFD program is used to expand the database to 56 cases for four types of indoor spaces in the U.S.: small offices, large offices with partitions, classrooms, and industrial workshops.
56
Figure 4.1 shows the typical configurations of the four types of spaces. The 56 cases break into 18 cases of small offices (SO), 12 cases of large offices with partitions (LO), 14 cases of classrooms (CR), and 12 cases of industrial workshops (WS). The thermal and flow conditions for the cases are summarized in Table 4.1. The cases vary the space height (H), ventilation rate (n), heat generated by occupant (Qo), heat generated by equipment (Qe), heat generated by overhead lighting (Ql), heat from transmitted solar radiation (Qslr), and heat from the building envelope other than the transmitted solar radiation (Qwl). The table also summarizes the total cooling load (Qt) per floor area. These thermal and flow boundary conditions cover a wide range in U.S. buildings: 8 ft room height 18 ft (2.43 m room height 5.5 m) 2 ach ventilation rate 15 ach 6.6 Btu/(hft2) Qt / A 38 Btu/(hft2) (21 W/m2 Qt / A 120 W/m2) 0.08 Qoe / Qt 0.68 0 Ql / Qt 0.43 0 Qex / Qt 0.92
where Qt = A = Qoe = Ql = Qex =
the total cooling load in the room the floor surface area the heat generated by the occupant and equipment the heat generated by the overhead lighting the heat from exterior walls and windows and the transmitted solar radiation.
Since there is temperature stratification in a room with displacement ventilation, the ceiling and floor surface temperatures are unknown. In this investigation, we use the two temperatures in the CFD program to calculate the air temperature and contaminant distributions. The following paragraphs discuss a procedure used to estimate the temperatures. In a space with displacement ventilation, as shown in Figure 4.2, the steadystate heat balance on the surfaces of the floor and the ceiling can be expressed as
Q af = Q sf + Q rf + Q of Q ac = Q sc + Q rc + Q oc
(4.2) (4.3)
where Qaf = the convective heat transfer from the floor to the air, Qsf = the radiative heat transfer from the heat sources to the floor, Qrf = the radiative heat transfer from the ceiling and walls to the floor,
57
(a)
(b)
(c)
(d) Figure 4.1 Typical rooms studied: (a) a small office, (b) a large office with partitions, (c) a classroom, and (d) a workshop.
58
Table 4.1a
SOSmall Offices H ft 9.2 8 11 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 9.2 15 2.52 3.71 9 2.52 3.71 8 2.52 3.71 6 2.52 3.71 6 2.52 3.71 3.17 3.17 3.17 3.17 3.17 4 1.26 0 0 4 2.52 3.71 3.17 4 2.52 3.71 3.17 0 6.34 12.7 6.34 14.1 14.1 14.1 27 4 2.52 0 0 3.17 4 2.52 3.71 0 3.17 4 2.52 1.85 3.17 3.17 4 2.52 0 3.17 3.17 2.23 2.19 2.23 2.33 0 2.04 2.13 2.07 1.91 1.91 1.91 1.63 4 1.26 3.71 3.17 3.17 2.16 6 2.52 3.71 3.17 3.17 2.14 3 2.52 3.71 3.17 3.17 2.14 4 2.52 3.71 3.17 3.17 2.14 14.7 14.7 14.7 13.5 11.1 12.9 11.6 8.02 9.41 17.8 16.1 17.8 25.4 25.4 25.4 38 4 2.52 3.71 3.17 3.17 2.14 14.7 4 2.52 3.71 3.17 3.17 2.14 14.7 76.3 75.2 77.4 75.7 76.8 76.1 76.1 76.1 75.7 75.6 76.3 76.6 76.5 77.4 78.6 79.3 79.5 83.1 n ach Qo/A Btu/hft2 Qe/A Btu/hft2 Ql/A Btu/hft2 Qslr/A Btu/hft2 Qwl/A Btu/hft2 Qt/A Btu/hft2 Tfs F Tcs F 82 81.7 82.4 82.8 81.3 81.5 79.9 81 80.4 78.3 80.1 83.5 82 82.6 85.6 84.9 84.6 88
Case Specifications (I-P Units)
Case
Ts F 60.8 57.7 64.4 55.2 66 62.1 64.4 62.6 64 67.5 66 57.7 59.4 64 59.2 63.5 64.9 66.2
SO1
SO2
SO3
SO4
SO5
SO6
59
SO7
SO8
SO9
SO10
SO11
SO12
SO13
SO14
SO15
SO16
SO17
SO18
Table 4.1a LOLarge Offices H ft Qe/A Btu/hft2 3.04 3.04 3.04 3.04 3.04 0 1.52 6.09 3.04 0 3.04 3.04 3.8 3.8 0 0 3.17 3.17 0 6.34 3.8 3.17 3.8 3.17 3.8 3.17 3.8 3.17 1.19 1.29 1.22 1.1 1.32 1.38 0 1.06 3.8 3.17 1.19 3.8 3.17 1.19 3.8 3.17 1.19 13.3 13.3 13.3 13.3 10.3 11.8 16.2 9.6 6.62 8.91 16.3 3.8 3.17 1.19 13.3 Ql/A Btu/hft2 Qslr/A Btu/hft2 Qwl/A Btu/hft2 Qt/A Btu/hft2 9.8 11 13 11 11 11 11 11 11 11 11 11 4 2.07 4 2.07 4 2.07 4 2.07 4 2.07 4 2.07 4 2.07 6 2.07 3 2.07 4 2.07 4 2.07 4 2.07 n ach Qo/A Btu/hft2 Tfs F 78.1 78.4 79 78.3 78.6 77.9 78.3 79 77.5 77 77.2 79.9 Tcs F 81.7 81.9 81.9 82.4 81.1 80.2 81 83.5 79.7 78.1 79.7 83.5
Case Specifications (I-P Units) (Continued)
Case
Ts F 63 64.6 66.9 60.4 68.5 66.9 65.7 62.4 67.6 69.8 68.4 61.9
LO1
LO2
LO3
LO4
LO5
LO6
60
LO7
LO8
LO9
LO10
LO11
LO12
Table 4.1a CRClassrooms H ft Qe/A Btu/hft2 0 0 0 0 0 0 1.2 0 0 0 0 0 0 0 3.8 0 3.8 3.8 3.8 0 6.34 9.51 9.51 3.17 5.71 3.17 1.9 3.17 3.8 3.17 1.14 1.24 1.11 0 1.05 0.92 1.05 1.17 3.8 3.17 1.14 3.8 3.17 1.2 3.8 3.17 1.2 3.8 3.17 1.17 13.8 13.8 12.5 15.1 15 11.9 15.6 9.45 16.8 19.9 16.2 13.8 3.8 3.17 1.2 13.8 3.8 3.17 1.17 13.8 78.6 77.5 79.3 79 78.3 79 78.8 78.1 79.2 77.4 80.1 81.5 80.4 78.8 Ql/A Btu/hft2 Qslr/A Btu/hft2 Qwl/A Btu/hft2 Qt/A Btu/hft2 Tfs F 11 9 13 11 11 11 11 11 11 11 11 11 11 11 4 5.64 3 5.64 3 5.64 3 5.64 3 5.64 3 5.64 3 5.64 3 5.64 3 7 3 4.28 5 5.64 3 5.64 3 5.64 3 5.64 n ach Qo/A Btu/hft2 Tcs F 82.2 81.7 82.4 81.3 81.5 82.8 82.9 81.1 83.3 79.9 84 85.8 83.7 81.7
Case
Ts F 59.7 55.2 63 66.6 61.2 58.3 58.5 61.9 57.7 64.9 55.9 52.2 56.3 64
CR1
CR2
CR3
CR4
CR5
CR6
61
CR7
CR8
CR9
CR10
CR11
CR12
CR13
CR14
Table 4.1a WSIndustrial Workshops H ft Qe/A Btu/hft2 1.94 1.94 1.94 1.94 1.94 0.97 0 1.94 1.94 1.94 1.94 1.94 0 3.17 3.17 3.17 0 6.34 9.51 9.51 0 3.17 3.17 3.17 3.17 3.17 1.4 1.39 1.49 0 1.3 1.2 1.3 3.17 3.17 1.17 3.17 3.17 1.11 3.17 3.17 1.2 3.17 3.17 1.17 14.3 14.3 14.2 14.3 11.1 12.6 11.4 9.95 17.6 20.7 17.6 3.17 3.17 1.14 14.3 Ql/A Btu/hft2 Qslr/A Btu/hft2 Qwl/A Btu/hft2 Qt/A Btu/hft2 15 10 18 15 15 15 15 15 15 15 15 15 3 4.85 3 4.85 3 4.85 3 4.85 3 4.85 3 4.85 3 2.42 4 4.85 2 4.85 3 4.85 3 4.85 3 4.85 n ach Qo/A Btu/hft2 Tfs F 74.8 72.9 75.2 74.7 74.8 73.9 74.5 73.9 73.9 75.6 76.5 75.6 Tcs F 79.9 79.7 79.7 81.3 79.2 78.1 78.8 77.9 77.9 81.7 83.5 81.3
Case
Ts F 62.2 54.5 64.6 56.5 64.9 64.4 63.3 64.2 64.9 60.1 57.9 60.1
WS1
WS2
WS3
WS4
WS5
WS6
62
WS7
WS8
WS9
WS10
WS11
WS12
Table 4.1b
SOSmall Offices H m Qe/A W/m2 11.7 11.7 11.7 11.7 11.7 11.7 0 5.84 11.7 0 11.7 11.7 0 11.7 11.7 11.7 11.7 11.7 10 10 10 10 10 0 10 20 40 20 44.4 44.4 44.4 85.2 10 0 0 10 0 10 7.04 7.34 0 6.44 6.72 6.54 6.04 6.04 6.04 5.14 10 10 6.9 10 10 7.04 10 10 6.82 42.5 35 40.7 36.7 25.3 29.7 56.1 50.7 56.2 80.1 80.1 80.1 120 10 10 6.74 46.4 10 10 6.74 46.4 10 10 6.74 46.4 25.2 24.3 24.9 24.5 24.5 24.5 24.3 24.2 24.6 24.8 24.7 25.2 25.9 26.3 10 10 6.74 46.4 24 10 10 6.74 46.4 24.6 27.8 27.6 28 28.2 27.4 27.5 26.6 27.2 26.9 25.7 26.7 28.6 27.8 28.1 29.8 29.4 Ql/A W/m2 Qslr/A W/m2 Qwl/A W/m2 Qt/A W/m2 Tfs C Tcs C 2.8 2.4 3.3 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 2.8 15 7.96 9 7.96 8 7.96 6 7.96 6 7.96 4 3.98 4 7.96 4 7.96 4 7.96 4 7.96 4 7.96 4 7.96 4 3.98 6 7.96 3 7.96 4 7.96 4 7.96 4 7.96 n ach Qo/A W/m2
Case Specifications (SI Units)
Case
Ts C 16 14.3 18 12.9 18.9 16.7 18 17 17.8 19.7 18.9 14.3 15.2 17.8 15.1 17.5
SO1
SO2
SO3
SO4
SO5
SO6
63
SO7
SO8
SO9
SO10
SO11
SO12
SO13
SO14
SO15
SO16
SO17
26.4 28.4
29.2 31.1
18.3 19
SO18
Table 4.1b LOLarge Offices H m Qe/A W/m2 9.6 9.6 9.6 9.6 9.6 0 4.8 19.2 9.6 0 9.6 9.6 12 12 0 20 0 10 0 10 12 10 12 10 3.86 3.46 4.16 4.36 0 3.36 12 10 4.06 12 10 3.76 12 10 3.76 41.9 41.9 32.6 37.2 51.2 30.3 20.9 28.1 51.5 12 10 3.76 41.9 12 10 3.76 41.9 12 10 3.76 41.9 25.8 25.6 26.1 25.7 25.9 25.5 25.7 26.1 25.3 25 25.1 26.6 Ql/A W/m2 Qslr/A W/m2 Qwl/A W/m2 Qt/A W/m2 Tfs C 3.3 3 3.9 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 4 6.54 4 6.54 4 6.54 4 6.54 4 6.54 4 6.54 4 6.54 6 6.54 3 6.54 4 6.54 4 6.54 4 6.54 n ach Qo/A W/m2 Tcs C 27.7 27.6 27.7 28 27.3 26.8 27.2 28.6 26.5 25.6 26.5 28.6
Case Specifications (SI Units) (Continued)
Case
Ts C 18.1 17.2 19.4 15.8 20.3 19.4 18.7 16.9 19.8 21 20.2 16.6
LO1
LO2
LO3
LO4
LO5
LO6
64
LO7
LO8
LO9
LO10
LO11
LO12
Table 4.1b CRClassrooms H m Qe/A W/m2 0 0 0 0 0 0 3.8 0 0 0 0 0 0 0 12 0 12 12 12 0 20 30 30 10 18 10 6 10 12 10 3.6 3.9 3.5 0 3.3 2.9 3.3 3.7 12 10 3.6 12 10 3.8 12 10 3.8 43.6 39.3 47.7 47.2 37.7 49.3 29.8 53.1 62.7 51.1 43.5 12 10 3.7 43.5 12 10 3.8 43.6 12 10 3.7 43.5 26 25.3 26.3 26.1 25.7 26.1 26 25.6 26.2 25.2 26.7 27.5 26.9 25.9 Ql/A W/m2 Qslr/A W/m2 Qwl/A W/m2 Qt/A W/m2 Tfs C 3.3 2.7 3.9 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3.3 3 17.8 3 17.8 3 17.8 3 17.8 3 17.8 3 17.8 3 17.8 3 17.8 3 22.1 3 13.5 5 17.8 3 17.8 3 17.8 4 17.8 n ach Qo/A W/m2 Tcs C 27.6 27.6 28 27.4 27.5 28.2 28.3 27.3 28.5 26.6 28.9 29.9 28.7 27.9
Case
Ts C 17.8 12.9 17.2 19.2 16.2 14.6 14.7 16.6 14.3 18.3 13.3 11.2 13.5 15.4
CR1
CR2
CR3
CR4
CR5
CR6
65
CR7
CR8
CR9
CR10
CR11
CR12
CR13
CR14
Table 4.1b WSIndustrial Workshops H m Qe/A W/m2 6.11 6.11 6.11 6.11 6.11 3.05 0 6.11 6.11 6.11 6.11 6.11 0 10 10 10 0 20 30 30 0 10 10 10 10 10 4.42 4.4 4.69 0 4.09 3.79 4.09 10 10 3.69 10 10 3.49 10 10 3.79 45.2 44.9 45.1 35.1 39.7 36.1 31.4 55.5 65.2 55.5 10 10 3.69 45.1 10 10 3.59 45 23.8 22.7 24 23.7 23.8 23.3 23.6 23.3 23.3 24.2 24.7 24.2 Ql/A W/m2 Qslr/A W/m2 Qwl/A W/m2 Qt/A W/m2 Tfs C 4.5 3 5.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 3 15.3 3 15.3 3 15.3 3 15.3 3 15.3 3 15.3 3 7.63 4 15.3 2 15.3 3 15.3 3 15.3 3 15.3 n ach Qo/A W/m2 Tcs C 26.6 26.5 26.5 27.4 26.2 25.6 26 25.5 25.5 27.6 28.6 27.4
Case
Ts C 16.8 12.5 18.1 13.6 18.3 18 17.4 17.9 18.3 15.6 14.4 15.6
WS1
WS2
WS3
WS4
WS5
WS6
66
WS7
WS8
WS9
WS10
WS11
WS12
Figure 4.2
Heat transfer in a space with displacement ventilation.
Qof Qsc Qrc
= the heat transfer from the space under the floor to the floor surface, = the radiative heat transfer from the heat sources to the ceiling, = the radiative heat transfer from the ceiling to the walls and floor, Further, Newtons law reads Qaf = cf (Tfs Tf)A Qac = cc (Tc Tcs)A (4.4) (4.5)
Qac = the convective heat transfer from the air to the ceiling,
Qoc = the heat transfer from the ceiling surface to the space above the ceiling.
where cf = the convective heat transfer coefficient on the floor, Tfs = the floor surface temperature, Tf = the air temperature near the floor, cc = convective heat transfer coefficient on the ceiling, Tcs = the ceiling surface temperature, A = floor/ceiling area. The convective heat transfer on the floor causes an air temperature increase from the supply temperature to the air temperature on the foot level. Therefore,
67
Qaf = Cp V(Tf Ts) .
(4.6)
The radiative heat transfer from the heat sources to the floor and the ceiling, respectively, may be estimated by
Q sf = r fj Q j
j
(4.7) (4.8)
Q sc = r cj Q j
j
where Qj = heat emitted by jth heat source, including transmitted solar radiation; rfj = the fraction of radiative heat transfer from jth heat source to the floor; rcj = the fraction of radiative heat transfer from jth heat source to the ceiling. The rfj and rcj need to be estimated from the room geometry. According to Mundt (1996), the radiative heat transfer from the ceiling and walls to the floor, Qrf, and the radiative heat transfer from the ceiling to the floor and walls, Qrc, can be estimated via Qrf = rA (Tcs Tfs) Y1 Qrc = rA (Tcs Tfs) Y2 (4.9) (4.10)
where = the radiative heat transfer coefficient, r Y1 and Y2 = coefficients. The values of Y1 and Y2 depend on the distribution of surface temperatures and the geometry of the room envelope. Mundt (1996) showed that the Y1 and Y2 are between 0.6 to 0.8 for rooms with displacement ventilation. The lower value corresponds to rooms with a high H/A (the ratio of the room height to floor area). The heat transfer from the space under the floor to the floor surface, Qof, and the heat transfer from the space above the ceiling to the ceiling surface, Qoc, can be expressed as Qof = A(Tof Tfs)/Rf Qoc = A(Tcs Toc)/Rc where Tof = Toc = Rf = Rc = (4.11) (4.12)
the temperature of the space under the floor, the temperature of the space above the ceiling, the thermal resistance from the space under the floor to the floor surface, the thermal resistance from the space above the ceiling to the ceiling surface.
68
The total cooling load is offset by the ventilation system, i.e., Qt = CpV(Te Ts) , where V Te Ts = air density, = volume flow rate from the supply, = air temperature at the exhaust, and = air temperature at the supply. Cp = specific heat of air, (4.13)
From the above equations, Mundt (1996) developed the following equation to calculate f = Tf Ts / Te Ts.
( Q sc Q oc ) Y 1 + ( Q sf + Q of ) Y 2 ( Q sf + Q of ) AH -------------------------------------------------------------------------- + --------------------------- + ------------------- C p VH e Q t Y 1 cc Qt Y1 r f = ----------------------------------------------------------------------------------------------------------------------------------------Y2 1 1 AH ------------- + ----------- + ------ + -------------------Y 1 cc Y 1 r cf C p VH e
(4.14)
With the assumption of a constant vertical air temperature gradient in the space, s, we have
Te Tf -, s = --------------He
(4.15)
where He = the exhaust elevation.

Qt ( 1 f ) -. s = ---------------------- C p VH e
Combination of f with Equations 4.13 and 4.15 leads to (4.16)
Once f and s are obtained, the temperatures can be determined for a given design as follows. The air temperature near the floor is Tf = Th s Hh , where Th = desired design room temperature at the head level of a sedentary person, and The exhaust air temperature is Te = Tf + s He . (4.18) Hh = the head elevation. (4.17)
69
The supply air temperature is Ts = Te Qt/(CpV) . The floor surface temperature is Tfs = Tf + CpV(Tf Ts)/(Acf) . The air temperature near the ceiling is Tc = Te + (H He)s . The ceiling surface temperature is
cc T c + r Y 2 T f + ( Q sc Q oc ) A -. T cs = ------------------------------------------------------------------------------- cc + r Y 2
(4.19)
(4.20)
(4.21)
(4.22)
Since Qof, Qoc, and Qt depend on Tfs and Tcs, iterations are necessary between Equations 4.11 to 4.22. The calculated temperatures are listed in Table 4.1. The above derivation assumes a constant vertical temperature gradient in the room air. This assumption is only used for the estimation of the surface temperatures on the floor and the ceiling and the supply air temperature. The experimental data from the literature show that the gradient is not a constant in many cases, but the average value of the data is close to constant. The CFD program will provide detailed temperature distributions. From the results of the 56 cases, we can obtain the air temperature difference between the head and foot level and the ventilation effectiveness at the breathing level of a sedentary person. 4.2 MODEL OF THE AIR TEMPERATURE DIFFERENCE BETWEEN THE HEAD AND FOOT LEVEL In a room with displacement ventilation, as illustrated in Figure 4.3, the air in the layer between the head and foot level of a sedentary occupant is heated by the occupants, equipment, transmitted solar radiation, overhead lighting, and walls. In other words, the increase in temperature from the foot to the head level results from the convective heat from the occupants and equipment, transmitted solar radiation, overhead lighting, and the heat gain/loss through the exterior walls/windows. Obviously, the heat from the occupants and equipment contributes more significantly to the temperature increase in this layer than that from overhead lighting. This is because the occupants and equipment are located in this layer. We can assume the heat transfer to the air between head and foot level by Thf CpV = aoeQoe + alQl + aexQex , where (4.23)
70
Figure 4.3
Sketch of a room with displacement ventilation.
Qoe Ql Qex
= heat generated by occupants, desk lamps, and equipment (Qo + Qdl + Qe), = heat generated by overhead lighting, = heat from the exterior wall and window surfaces and the transmitted solar radiation, and
aoe, al, and aex = weighting coefficients for the contribution of the convective heat to the air between head and foot level. Since V = nHA , where V n A = = = the supply flow rate, the air change rate, the room height, and the floor area, (4.24)
H =
Equation 4.23 turns to

a oe Q oe + a l Q l + a ex Q ex -. T hf = -------------------------------------------------------- C p nHA
(4.25)
Equation 4.25 is a model to calculate the air temperature difference between the head and foot level of a sedentary person in a room with displacement ventilation. Based on the database for the cases listed in Table 4.1, the coefficients that give the best agreement are
71
a oe = 0.295 , a l = 0.132 , and a ex = 0.185 .
(4.26)
The model should only be applied to cases within the range of the present database. This range covers most U.S. buildings except large spaces, such as theaters and atria. We recommend the use of a validated CFD program or experimental measurements to design displacement ventilation systems in large spaces. For people and unhooded equipment, the cooling load factor is about 0.75. The model (Equation 4.25) indicates that one-third of the cooling load enters the space between foot and head level. The other two-thirds enter the upper space by the thermal plume. The radiative heat from the overhead lighting to the building envelope is about 20% of the total energy input to the lamps. About two-thirds of the radiative heat is projected to the floor and lower part of the wall. This eventually heats the air between foot and head level. Mundt (1996) measured the air temperature profile in a room with displacement ventilation. The only heat source is a simulated person. The temperature difference between the head and foot level of the occupant over the temperature difference between the return and supply air is 0.3, which is in excellent agreement with the aoe value. This suggests the values of the weighting coefficients, aoe, al, and aex, are physically sound. Figure 4.4 compares the air temperature differences obtained from the database and calculated by the model (Equation 4.25). The temperature differences calculated with the assumption of a constant vertical temperature gradient, (Te Tf)/H, are also presented in the figure for comparison. The correlation between the model and data-
Figure 4.4
Correlation of the air temperature difference between the head and foot level.
72
base is very good. This implies the temperature differences calculated with the model are close to those by the CFD simulation. It is not surprising that the average values calculated by the model agree with the simulation since the values of the coefficients were obtained from the same simulations. It is gratifying that the model does accurately capture the influence of individual parameter variations on the foot to head temperature difference. However, the assumption of a constant temperature gradient from floor to ceiling is not very good. Straub (1962) has provided a good explanation of how temperature gradient is formed. Obviously, the constant gradient assumption neglects many factors despite its simple form. Figure 4.5 provides a more detailed case-by-case comparison for the small offices, large offices with partitions, classrooms, and industrial workshops. The results show that the assumption of a constant temperature gradient is more problematic when the ceiling height is high, such as in the workshops shown in Figure 4.5d. Figure 4.6 compares the Thf obtained by the model and the constant temperature gradient assumption with the measured data from the literature. The agreement between the model and the data is less satisfactory in some cases, such as those cases from Holmberg et al. (1987) and Nielsen et al. (1988). No temperature information on the walls is available from Holmberg et al. (1987). The temperature difference from head to foot will be influenced by heat transfer from the vertical walls. There is no indication of wall temperatures to determine if the wall adds or removes heat from the room air. In Nielsens cases, there was a large glass wall in the test room for which no temperature information is available. Among the three cases from Brohus and Nielsen (1996), the calculated values agree with the last two cases but not with the first one. The measured Thf in the first case should be much larger than that in the third one because the heat sources are almost the same between the two cases and the ventilation flow rate in the first case is less than half of that in the third case. Thus, some other unreported changes in room conditions may have occurred. For all other cases, the Thf calculated by the model is close to the measured data. The above comparison shows that the model estimates the temperature difference between head and foot level much better than the assumption of a constant temperature gradient from floor to ceiling. The model (Equation 4.25) shows that a large cooling load can cause a large Thf. The head-to-foot temperature difference, Thf, should be less than 3.6F (2 K) for comfort consideration. Therefore, the cooling load has an upper limit for acceptable comfort with displacement ventilation. However, the model suggests that the air temperature difference between the head and foot level not only depends on the total cooling load but also the type of heat gains. The maximum cooling load is not a fixed value for thermal comfort in displacement ventilation. If a majority of the cooling load is from overhead lighting or other heat sources above the stratification level, displacement ventilation can operate with a much higher cooling load and still provide comfortable conditions.
73
(a)
(b)
Figure 4.5
Comparison of the Thf obtained by the model, the CFD data, and the constant temperature gradient assumption: (a) small offices and (b) large offices with partitions.
74
(c)
(d)
Figure 4.5
(Continued) Comparison of the Thf obtained by the model, the CFD data, and the constant temperature gradient assumption: (c) classrooms and (d) industrial workshops.
75
76
Figure 4.6
Comparison of the Thf obtained by the model, measured data, and constant gradient assumption.
The model indicates that an increase of the ventilation rate, n, may reduce Thf. However, the air speed from the diffuser cannot be too high. Maintaining a thermally comfortable environment requires a large diffuser area when the ventilation rate increases. Therefore, the maximum cooling load depends on the area available for installing diffusers and the distribution of heat sources. Note that a higher ventilation rate will consume more energy from the fan and requires a larger air-handling unit. 4.3 VENTILATION EFFECTIVENESS MODEL It is difficult to derive a general model for ventilation effectiveness with displacement ventilation. We restrict our efforts to a model for rooms where the contaminant sources are associated with the heat sources. The database was used with a technique similar to that for the model of the air temperature difference (Equation 4.25). The model for prediction of ventilation effectiveness, , at the breathing level of a sedentary person in a displacement ventilated room is
= ( 2.83 ( 1 e
n 3
) ( Q oe + 0.45 Q l + 0.63 Q ex ) Q t ) ,
(4.27)
where n = ventilation rate.

ch cs = --------------ce cs
(4.28)
where ch = the mean contaminant concentration at the head level of a sedentary person, cs = the contaminant concentration at the supply air, ce = the contaminant concentration at the exhaust air. Equation 4.27 is purely an empirical best fit with the data. However, the ratios between the coefficients for Qoe, Ql, and Qex are the same as those for the model of the air temperature difference (Equation 4.26). The ventilation effectiveness in the database is with contaminants from the occupantsthe contaminant sources are combined with heat sources. The ventilation effectiveness model (Equation 4.27) is only valid for the same conditions. Additionally, the ventilation effectiveness is not a constant in both vertical and horizontal directions. The model calculates the average ventilation effectiveness throughout the room at the height of the breathing level. One must also consider occupants standing in a space with displacement ventilation. Saeteri (1992) and Brohus and Nielsen (1996) showed that the air inhaled originates at a lower elevation because the convective flow around the human body brings fresher air from the lower level to the breathing level. Therefore, the air quality inhaled by a standing person is probably close to the quality of that at the breath-
77
Figure 4.7
Correlation of ventilation effectiveness at the breathing level between the model and CFD data.
Figure 4.8
Comparison of ventilation effectiveness between the model and CFD data.
ing level of a sedentary person. The model may still be valid for spaces primarily occupied by standing people. Figure 4.7 compares the ventilation effectiveness between the model and database. The correlation is good. Figure 4.8 provides a detailed case-by-case comparison. The values of ventilation effectiveness predicted by the model are generally in good agreement with the CFD results. The ventilation effectiveness for the 56 cases varies between 1.2 and 2. Since the ventilation effectiveness for perfect mixing ventilation is 1.0, displacement ventilation does provide better indoor air quality.
78
The model indicates that the effectiveness increases as the ventilation rate increases. When the ventilation rate is sufficiently low, the increase of the effectiveness with ventilation rate is very pronounced. The model also suggests that the ventilation effectiveness is high when the fraction of the heat sources in the occupied zone (Qoe) over the total heat source is large. This is because a large Qoe generates strong thermal plumes that can transport the contaminants from the occupied zone to the upper zone. 4.4 CONCLUSIONS A model has been developed to estimate the air temperature difference between the head and foot level in a space with displacement ventilation. The model was developed from a database of 56 displacement ventilation conditions by use of a validated CFD program. The 56 cases cover four different types of buildings: small offices, large offices with partitions, classrooms, and industrial workshops under different thermal and flow boundary conditions normally found in U.S. buildings. The model should not be applied to large spaces such as theaters and atria until it is validated for such conditions. This investigation shows that the maximum cooling load in a room with displacement ventilation is not a fixed value. The cooling load depends on the distribution of the heat sources and the ventilation rate of the indoor space. Based on the same database, a model of ventilation effectiveness at the breathing level of a sedentary person has also been developed. The model is applicable to indoor spaces where the contaminant sources are associated with heat sources. The study also shows that ventilation effectiveness is high when a large fraction of the total heat source is in the occupied zone.
79
CHAPTER 5
Performance Evaluation of Displacement Ventilation

At the design stage, a designer needs to predict the performance of displacement ventilation. The performance is often evaluated by the thermal comfort level, indoor air quality, energy consumption of the HVAC system, and the first cost and maintenance costs of the system. The air temperature distribution, the percentage of dissatisfied people due to draft (PD), and the predicted percentage of dissatisfied for the thermal comfort (PPD) are widely used as criteria to evaluate the thermal comfort. The contaminant concentration distributions and the mean age of air are often good indicators for indoor air quality. The energy consumption is related to air temperature distribution and ventilation rate. All of these performance parameters are determined by the thermal and flow boundary conditions, such as the space size and geometry, heat sources, and contaminant sources. 5.1 EVALUATION CRITERIA This chapter will evaluate the performance of displacement ventilation for individual small offices, large offices with partitions, classrooms, and industrial workshops. The investigation uses the CFD program validated in Chapter 3. This program is used to calculate the flow and thermal distribution for a large number of different boundary conditions. However, normal CFD programs do not calculate PD, PPD, and the mean age of air. We have implemented the following models in the CFD program to calculate these parameters. Fanger et al. (1989) developed a model to calculate PD as PD = 0.021 (93.2 T) (u 9.8)0.62(3.14 +0.0019u Tu) [%] PD = (34 T)(u 0.05)0.62(3.14 +0.37u Tu) [%] for PD > 100%, use PD = 100%, where (5.1a) (I-P) (5.1b) (SI)
81
T u Tu
= air temperature, F (C); = air velocity, fpm (m/s)for u < 10 fpm (0.05 m/s), use u = 10 fpm (0.05 m/s); = turbulent intensity. We use Tu = 100(2k)0.5/ u [%] , (5.2)
where k = the turbulent kinetic energy. The PPD can be calculated via (ISO 1990) PPD = 100 95exp(-0.03353PMV4 0.2179PMV2) [%] . The predicted mean vote, PMV, in the equation is determined by PMV = 3.155 [0.303 exp(0.114M) + 0.028] L PMV = [0.303exp(0.036M) + 0.028] L where L=MW {1.196 10-9fcl[(Tcl + 460)4 (Tr + 460)4] + fclhc(Tcl T) + 0.97 [5.73 0.022 (M W) 6.9 Pa] + 0.42(M W 18.43) + 0.0173 M (5.87 0.69 Pa) 0.00077 M (93.2 T)} or L=MW {3.96 10-8fcl[(Tcl + 273)4 (Tr + 273)4] + fclhc(Tcl T) + 3.05 10-3 [5733 6.99 (M W) Pa] + 0.42 (M W 58.15) + 1.7 10-5 M (5867 Pa) + 0.0014 M (34 T)} with M = W = fcl = T = Tcl = Tr = hc = Pa = (5.5b) (SI) (5.4a) (I-P) (5.4b) (SI) (5.3)
(5.5a) (I-P)
metabolism, Btu/h (W); external work, Btu/h (W); cloth factor; local air temperature, F (C); cloth temperature, F (C); mean radiant temperature, F (C); convective heat transfer coefficient between the cloth and air, Btu/hft2F (W/m2C); partial water vapor pressure, in. water (Pa).
82
The fcl, Tcl, and hc are determined by the following equations: fcl = 1.05 + 0.645Icl for Icl 0.078 fcl = 1.00 + 1.290Icl for Icl < 0.078 (5.6) (5.7)
Tcl = 96.3 0.156(M W) Icl{1.196 10-9fcl[(Tcl + 460)4 (Tr + 460)4] (5.8a) (I-P) + fclhc(Tcl T)} Tcl = 35.7 0.028(M W) Icl{3.96 10-8fcl[(Tcl + 273)4 (Tr + 273)4] (5.8b) (SI) + fclhc(Tcl T)} The convective heat transfer coefficient, hc, is determined from: hc = 0.361(Tcl T)0.25 for 0.361(Tcl T) 0.25 0.151 u0.5 hc = 2.38(Tcl T)0.25 for 2.38(Tcl T) 0.25 12.1 u0.5 hc = 0.151 u0.5 for 0.361(Tcl T) 0.25 < 0.151 u0.5 hc = 12.1 u0.5 for 2.38(Tcl T) 0.25 < 12.1 u0.5 (5.9a) (I-P) (5.9b) (SI) (5.10a) (I-P) (5.10b) (SI)
where Icl = clothing insulation, Fft2h/Btu (Cm2/W) u = air velocity, fpm (m/s) The mean age of air, , is defined as the averaged time for all air molecules to travel from the supply diffuser to that point. It can be derived from the measured transient history of the tracer-gas concentration. Li and Jiang (1996) show that the mean age of air is governed by a transport equation,
---- ( ) + ------ ( u j ) = ------ ------ +, xj x j x j t
(5.11)
with the following as the boundary conditions: = 0 at the supply diffuser and ------ = 0 at the exhaust and walls. xj The ventilation effectiveness is defined as
ce cs -, = -------------c cs
(5.12)
where = ventilation effectiveness; ce = contaminant concentration at the exhaust air, ppm;
83
cs c
= contaminant concentration at the supply air, ppm; = contaminant concentration in the room air, ppm. In addition, a CFD program will calculate the airflow pattern and the distributions of air velocity, temperature, and contaminant concentration. We will be able to evaluate the thermal comfort and indoor air quality provided by displacement ventilation in terms of airflow pattern, temperature distribution, PD, PPD, contaminant concentration distributions, mean age of air, ventilation effectiveness.
The energy consumption of the HVAC system and the first costs and maintenance costs of the system for displacement ventilation are also important parameters for evaluating the performance of a displacement ventilation system. The research results are reported in the next chapter. 5.2 PERFORMANCE EVALUATION OF DISPLACEMENT VENTILATION For the 56 cases as listed in Table 4.1, we have calculated the airflow patterns and the distributions of air temperature, PD, PPD, CO2 concentration, the mean age of air, and ventilation effectiveness. This section uses the results for a classroom shown in Figure 4.1c as an example. The classroom has a teacher, 24 pupils, an exterior window/wall, overhead lighting, and diffusers in all four corners. More detailed information is provided in case CR01 in Table 4.1. The results for other cases can lead to similar conclusions (Chen et al. 1999). Airflow Pattern The classroom uses one diffuser in each of the four corners. Figure 5.1 shows the airflow patterns in three different sections of the room: a horizontal section at the ankle level, section A-A, and section B-B (see Figure 5.1a for the section location). The relatively cold air from the diffusers falls toward the floor. The falling flows behave like jets, and the jets can meet in the center of the room. The wall temperature determines the flow near the walls. The heated objects, such as the occupants, generate strong plumes in the room that bring the contaminants from the lower zone to the upper zone. If the walls are assumed to be isothermal and have the same temperature as the average room air temperature, the wall flow in the lower part is upward and, in the upper part of the wall, the flow is downward. This is because room air temperature in the lower part is lower than that of the walls and, in the upper part, higher than that of the walls. However, in our experiment, we found that the temperature of
84
Figure 5.1
The airflow pattern in the classroom: (a) at ankle level, (b) at section A-A, and (c) at section B-B.
85
the walls is higher than the room air temperature. Therefore, the flows near the walls are all upward, as shown in Figure 5.1. The air velocity in the room with displacement ventilation is generally small (less than 40 fpm or 0.2 m/s), except in the thermal plumes, and the flow near the floor and walls. In the winter, the downward flow near the exterior window and wall might bring pollutants from the upper zone to the lower zone. To prevent such a downward flow requires a heating device placed near the exterior window or wall at the floor level. The heating capacity should be slightly larger than the heating load from the exterior window or wall. Temperature Distribution Figure 5.2 shows that the air temperature is nearly uniform in the horizontal direction except in the region close to an occupant. The supply air is heated first by the floor and mixed with room air by induction. As a result, the air temperature near the floor, Tf, is higher than that of supply air. To avoid draft, the supply air temperature cannot be too low. The supply air temperature depends on the room geometry, cooling load, and heat source type. Generally, the supply air temperature is in the range of 65F to 68F (18C to 20C). There is a vertical gradient of the air temper-
Figure 5.2
The air temperature distribution in the classroom: (a) at section A-A and (b) at section B-B.
86
ature. The vertical temperature gradient depends on the distribution of the heat sources. The gradient in the lower part will be larger than that in the upper part if most of the heat sources are in the lower part of the room. Although Figure 5.2 shows a high air temperature near the ceiling, the ceiling surface temperature is several degrees lower. In general, radiation from the ceiling may not be felt, unless the air temperature near the ceiling is very high. Percentage Dissatisfied Due to Draft (PD) and Predicted Percentage Dissatisfied (PPD) The PD and PPD are generally less than 15% in the occupied zone if the design uses the guidelines to be shown in Chapter 7. Figures 5.3 and 5.4 show that only in the region very close to the diffuser (1.6 ft or 0.5 m) the PD and PPD are higher. Therefore, displacement ventilation presents a satisfactory comfort level. Contaminant Concentration Distributions The present study uses CO2 as an indicator for contaminants. With displacement ventilation, the CO2 concentration in the lower zone is lower than that in the upper zone, as shown in Figure 5.5. The CO2 level at the inlet is 400 ppm. The indoor CO2 sources are from the occupants, and the occupants are the heat sources as well. The heat generates thermal plumes that can bring the CO2 to the upper zone. In most buildings, many contaminant sources are associated with the heat sources, such as printers, computers, and other heated equipment. Displacement ventilation provides better indoor air quality than mixing ventilation for these cases. The convective flow around a human body brings the air at the lower zone to the breathing zone. Therefore, the occupant actually breathes air with lower contaminant concentrations than those at the nose level in the middle of the room. However, displacement ventilation may not provide better indoor air quality than mixing ventilation if the contaminant sources are not associated with heat sources, such as the volatile organic compounds from building materials. The concentration is indeed very sensitive to the locations of the contaminant and heat sources, wall thermal conditions, and disturbances, such as moving objects in the room. Mean Age of Air and Ventilation Effectiveness Figure 5.6 illustrates the mean age of air in the classroom. Clearly, the mean age of air in the lower part of the room is much younger than that in the upper part of the room. The mean age of air at the breathing level in the classroom is about 600 s. For complete mixing ventilation, the mean age of air in the classroom is 900 s. The distribution of ventilation effectiveness shown in Figure 5.7 also indicates that the classroom with displacement ventilation system has higher ventilation effectiveness. With a perfect mixing ventilation system, the ventilation effectiveness is one. A perfect or complete mixing is impossible in practice. The corresponding mean age
87
Figure 5.3
The PD distribution in the classroom: (a) at ankle level, (b) in section A-A, and (c) in section B-B.
88
Figure 5.4
The PPD distribution in the classroom: (a) at ankle level, (b) in section A-A, and (c) in section B-B.
89
Figure 5.5
The CO2 distribution in the classroom: (a) in section A-A and (b) in section B-B.
Figure 5.6
The distribution of mean age of air in the classroom: (a) in section A-A and (b) in section B-B.
90
Figure 5.7
The ventilation effectiveness distribution in the classroom: (a) in section A-A and (b) in section B-B.
of air will be older in the occupied zone than with displacement ventilation, and the ventilation effectiveness will be lower than one. Hence, we can conclude that displacement ventilation does provide much better indoor air quality than mixing ventilation. 5.3 DISCUSSION We are particularly interested in studying cases with a high cooling load. Our survey in several buildings in the greater Boston area shows that the cooling load can be as high as 40 Btu/(hft2) or 120 W/m2. Some investigators suggested that the maximum cooling load a displacement ventilation system can handle is about 13 Btu/(hft2) (40 W/m2) without increasing ventilation rate. In order to design displacement ventilation for U.S. buildings, the ventilation rate must be increased because of the high cooling loads. A displacement ventilation system may maintain a comfortable environment with a cooling load up to 40 Btu/(hft2) (120 W/m2) with increased ventilation rate. We have studied a case with such a high cooling load in a small office. As shown in Figures 5.8 and 5.9, the PD and PPD in the occupied zone can still be less than 15%. The air temperature difference between head and foot levels is 3F (1.6C). In this case, the ventilation rate is 15 ach. The high ventilation rate would require a large fan, duct, and air-handling unit. It may not be economically feasible compared to
91
Figure 5.8
Distribution of the percentage of dissatisfied people due to draft in a small office with a cooling load of 40 Btu/(hft2) (120 W/m2): (a) at ankle level and (b) at head level.
Figure 5.9
Distribution of the predicted percentage of dissatisfied people for thermal comfort in a small office with a cooling load of 40 Btu/ (hft2) (120 W/m2): (a) at ankle level and (b) at head level.
mixing ventilation. In other words, energy efficiency and cost might limit the maximum cooling load that will be discussed in the next chapter. In addition, the available area of the walls for installing supply diffusers will limit the maximum cooling load. In winter conditions, downward flow near the exterior window and wall might bring polluted air from the upper zone to the lower zone. A heating device placed near the exterior wall at the floor level may prevent the downward flow. If the displacement diffuser is used to supply warm air, the air will move upward. The advantage of displacement ventilation for better indoor air quality will disappear.
92
Figure 5.10 illustrates the computed flow pattern with a baseboard heater with a capacity equal to the heating load from the exterior wall and window, while Figure 5.11 shows the flow pattern with a heating capacity equal to 80% of the heating load. It can be seen from the figures that the heating capacity should be at least equal to the heating load.
Figure 5.10 Flow pattern with a heating capacity equal to the heating load from the exterior wall and window.
Figure 5.11 Flow pattern with a heating capacity equal to 80% of the heating load from the exterior wall and window.
93
5.4 CONCLUSIONS With proper design, displacement ventilation can maintain a thermally comfortable indoor environment. The air velocity is smaller than 40 fpm or 0.2 m/s. The temperature difference between the head and foot level of a sedentary occupant is less than 3.6F (2 K). The percentage of dissatisfied people due to draft (PD) and the predicted percentage of dissatisfied are less than 15%. Compared with mixing ventilation, displacement ventilation provides better indoor air quality when the contaminant sources are associated with the heat sources. Thermal plumes bring the contaminants to the upper zone and the contaminant concentrations in the lower zone are lower. The mean age of air is younger and the ventilation effectiveness is higher in a room with displacement ventilation than with mixing ventilation. A high cooling load in a room would require a high ventilation rate that may limit the application of displacement ventilation.
94
CHAPTER 6
Energy and Cost Analysis

Proper design of displacement ventilation requires information about its energy consumption and first costs. A good ventilation system should save energy and be cost-effective. This chapter presents energy and cost analysis of displacement ventilation. The investigation uses a mixing system for comparison. Energy simulation methods range from manual to detailed computer simulation methods. Manual methods, such as degree-day and bin methods (ASHARE 1997), are still widely used in practical design, although they are not accurate. The modeling strategy used in building energy simulation is in a sequence of load, system, and plant (Sowell and Hittle 1995), no matter whether a manual or a detailed computer simulation method is used. Degree-day uses only one value of temperature, while the bin method calculates energy over several intervals (bins) of temperature. However, detailed methods often calculate energy in an hour-by-hour interval. Although the manual methods are simple, they could not, for example, be used for the comparison of energy consumption by displacement and mixing ventilation systems. A detailed computer simulation can consider the difference between displacement ventilation and mixing ventilation. That method is also powerful for analyzing a number of alternatives in order to make an optimal design for an HVAC system. Therefore, the present investigation uses a detailed computer simulation method. The detailed methods calculate cooling and heating loads hour-by-hour for an entire year for a building. Then the secondary systems are simulated to calculate the required energy flows at the air handlers or other equipment supplied by the central plant. The next step is to calculate the source energy requirements in the central plant. Finally, one would calculate the costs of the source energy, and sometimes introduce capital and other costs for a complete life-cycle economic analysis. 6.1 LOAD CALCULATIONS This study used the ACCURACY program (Chen and Kooi 1988) for load calculation. The fundamentals of the program are discussed here.
95
Room Air Energy Balance Equation The energy balance equation of room air is
N
i=1
V room C p T q i , c A i + Q lights + Q people + Q appliances + Q infiltration Q heat _ extractions = -----------------------------------t
(6.1) where
N
qi , c Ai
i=1
= = =
convective heat transfer from enclosure surfaces to room air, number of the enclosure surfaces, area of surface i, cooling loads of lights, people, appliances and infiltration, respectively, heat extraction via HVAC device, room air energy change, air density, room volume, air specific heat, temperature change of room air, and sampling time interval.
N Ai
Qlights, Qpeople, Qappliances, and Qinfiltration =
Qheat_extraction
V room C p T ------------------------------t
= = = = = = =
Vroom Cp T t
In order to determine the convective heat flux from surface i, qi,c, we need the energy balance equations for building enclosure. Energy Balance on a Wall, Ceiling, Floor, Roof, or Slab For a wall, ceiling, floor, roof, or slab, as shown in Figure 6.1, we have the following energy balance equation:
N
q i + q i , t = q ik + q i , c
k=1
(6.2)
where qi qi,t qik = = = conductive heat flux on surface i, transmitted solar heat flux re-absorbed by surface i, emitted radiative heat flux from surface i to surface k.
96
Figure 6.1
Energy balance on the interior surface of a wall, ceiling, floor, roof, or slab.
The ACCURACY program determines qi by Z-transfer functions. The radiative heat flux is
q ik = h ik , r ( T i T k ) ,
(6.3)
where hik,r = radiative heat transfer coefficient between surfaces i and k, = temperature of interior surface i, Ti = temperature of interior surface k, Tk and
q i , c = h c ( T i T air ) ,
(6.4)
where = convective heat transfer coefficient, and hc Troom = room air temperature. Most existing energy simulation programs assume Tair to be uniform in the entire room/building. The assumption is appropriate for a room with a mixing ventilation system where the room air temperature is relatively uniform. However, a single temperature is not good for displacement ventilation because the non-uniform temperature distribution in displacement ventilation can have a major impact on energy consumption of the HVAC system. The air temperature at the boundary layer of a wall is an important factor for the heat transfer through convection in the air-wall interface. This study uses the air temperature at 4 in. (0.1 m) from a wall surface (Ti,air in Figure 6.2) as the Tair. If the air temperature in the center of the occupied zone is controlled to be Troom, Equation 6.4 becomes qi,c = hi,c (Ti Ti,air) = hi,c(Ti Troom) hi,c Ti,air , (6.5)
97
Figure 6.2
Schematic presentation of the coupling between flow and energy programs.
Figure 6.3
Energy balance on the interior surface of a window.
where Ti,air = Ti,air Troom . The Ti,air can be directly obtained from CFD simulation, as shown in Figure 6.2. However, this could be very time consuming if the CFD simulation would be done hourly for a whole year. Hence, the present investigation uses the temperature model developed and used in Chapter 4 to determine Ti,air. Energy Balance Equation for Window For the window shown in Figure 6.3, we have the following energy balance equation:
98
q i + q i , s + q i , t = q ik + q i , c
k=1
(6.6)
where qi = conductive heat flux on window i, qi,s = inward heat flux of the absorbed solar radiation by window i, qi,t = transmitted solar heat flux reabsorbed by window i, and qik = emitted radiative heat flux from window i to room surface k. The ACCURACY program can calculate qi, qi,s, and qi,t. The qik and qi,c are determined in the same way as those for walls. Let qi,in = qi + qi,t for walls, ceiling, floor, etc. qi,in = qi + qi,s + qi,t for windows. Then we have [H] [T] = [q] + [h T] where
N
(6.7)
h 1, c + h 1 k , r
k=1
h 12, r
N
h1 N , r h2 N , r
N
[H] =
h 21, r h N 1, r
h 2, c + h 2 k , r
k=1
(6.8)
h NN 1, r h N , c + h Nk , r
k=1
T1 [T] = T2 TN
(6.9)
q 1, in + h 1, c T room [q] = q 2, in + h 2, c T room q N , in + h N , c T room
(6.10)
99
h 1, c T 1, air [ h T ] = h 2, c T 2, air h N , c T N , air
(6.11)
Solving Equations 6.1 to 6.11 together, the surface temperatures of room enclosure, Ti, and heat extraction (heating/cooling load) can be obtained. 6.2 SECONDARY SYSTEMS AND PLANTS Secondary systems simulate mass, energy, and moisture balance at various components (coils, fans, etc.) and junctions (mixing boxes, deviators) in the airhandling systems. Most programs, whole-building simulation programs, provide a number of fixed menus of common secondary systems. A typical program is EnergyPlus or DOE-2. Some programs, component-based programs such as TRNSYS (Klein et al. 1994), provide component-based or modular simulators. These programs would allow the user to interconnect freely the components that are packaged as algorithms. Generally, those programs are most useful for detailed studies of special systems, such as active solar heating and cooling systems and for control studies. The whole-building simulation approach seems the best to compare the energy consumption by displacement and mixing ventilation systems, and it is used for the present investigation. Figure 6.4 shows the air-handling systems used. Note that the figure shows only one of many possible systems. Another preferred system is to use 100% outdoor air to achieve better air quality. Then energy in the exhaust air can be conserved by using exchangers. According to the results shown in Chapter 5, the displacement ventilation system needs a separated heating system for winter heating. The present investigation used a baseboard heater, as shown in Figure 6.4. Further, in order to conserve energy, the air-handling systems use economizers and heat exchangers. The mainstream programs model chillers by curve fits of manufacturers data and boilers with a single seasonal efficiency. More detailed models are also available, such as inclusion of combustion calculations in boilers. The present investigation uses a variable-air-volume system with constant air supply temperatures, except in the shoulder season when the supply air temperature fluctuates for maximum use of free cooling. We use two COP values for chillersone for displacement ventilation and another for mixing ventilation. 6.3 ENERGY ANALYSIS FOR U.S. CONDITIONS This section compares energy consumption by displacement ventilation with that by a mixing-type system for an individual office building, a classroom building, and a workshop building. The study is for five climatic regions in the U.S.: hot,
100
Figure 6.4
The air-handling systems used by the displacement and mixing ventilation.
humid (e.g., New Orleans, LA), hot, dry (e.g., Phoenix, AZ), moderate (e.g., Nashville, TN), maritime (e.g., Seattle, WA), and cold (e.g., Portland, ME). Table 6.1 lists the building characteristics and thermal conditions used in the study. In addition, the investigation uses the following assumptions/conditions: A fixed fan efficiency = 0.60 A fixed boiler efficiency = 0.75 A fixed COP for chiller = 2.9 for mixing ventilation and 3.0 for displacement ventilation Pressure drop of the air-handling system: 1900 Pa (7.64 in. of water) Supply air temperature for the mixing system = 12.8C (55F) for cooling and 40C (104F) for heating Supply air temperature for the displacement system is determined by Equation (4.19) Minimum outdoor air = 10 L/s per person (20 cfm per person) for the mixing ventilation and 7.7 L/s per person (15 cfm per person) for displacement ventilation to ensure the same indoor air quality
101
Table 6.1
Space Type Space Size
Building Characteristics and Thermal Conditions

Small Office (SO) Classroom (CR) Workshop (WS)
5.2 m 3.7 m 2.4 m 11.7 m 9.0 m 3.3 m 26.2 m 21 m 4.5 m (17 ft 12 ft 8 ft) (38 ft 30 ft 11 ft) (86 ft 68 ft 15 ft) U = 0.72W/m2K (R = 7.9 ft2hF/Btu) U = 0.72W/m2K (R = 7.9 ft2hF/Btu)
Exterior Wall Seattle & U = 0.72W/m2K envelope Portland (R = 7.9 ft2hF/Btu)
Phoenix, U = 0.96W/m2K U=0.96W/m2K U = 0.96W/m2K New (R = 5.92 ft2hF/Btu) (R = 5.92 ft2hF/Btu) (R = 5.92 ft2hF/Btu) Orleans & Nashville Glazing 1) Double glazing. 1) Double glazing. 1) Double glazing. U = 4.6 W/m2K U = 4.6 W/m2K U = 4.6 W/m2K 2 2 (R = 1.24 ft hF/Btu) (R = 1.24 ft hF/Btu) (R = 1.24 ft2hoF/Btu) 2) 52% of exterior wall area 2) 44% of exterior wall area 2) 61% of exterior wall area
3) shading coefficient 3) shading coefficient = 3) shading coefficient = = 0.5 if facing south & 0.5 (facing south) 0.5 (facing south) = 0.8 if facing north. Occupancy schedule Internal load (sensible and latent) 8:00 a.m. 7:00 p.m. Monday Friday 2 persons: 260 W (887 Btu/h) 2 computers: 250 W (853 Btu/h) lights: 204 W (696 Btu/h) T = 25C (77F) for cooling T = 23C (73.5F) for heating 8:00 a.m. 7:00 p.m. Monday Friday 25 persons: 3,250 W (11,088 Btu/h) lights: 1,264 W (4,313 Btu/h) 8:00 a.m. 7:00 p.m. Monday Friday 112 persons: 14,560 W (49,676 Btu/h) equipment: 3,362 W (11,470 Btu/h) lights: 5,502 W (18,772 Btu/h) T = 25C (77F) for cooling T = 23C (73.5F) for heating
Room temperature setpoint
T = 25C (77F) for cooling T = 23C (73.5F) for heating
102
Figure 6.5
Comparison of monthly energy consumption between the displacement and mixing ventilation systems for an individual office in Seattle, WA (maritime climate).
Office Building Figure 6.5 shows the monthly energy consumption of a small office with a south-facing wall and window in Seattle, WA. The displacement ventilation system uses more fan energy than the mixing ventilation system. Although the exhaust air temperature with the displacement ventilation system is higher than that with the mixing ventilation system, the air temperature difference between the air exhaust and supply is smaller. This is because the air supply temperature in the displacement ventilation system is much higher. Typically, for the displacement system, Texhaust Tsupply = 81F 64F = 17F or (27C 18C = 9C) , and for the mixing system, Texhaust Tsupply = 77F 55F = 22F or (25C 13C = 12 K) . (6.13) (6.12)
To remove the same amount of cooling load requires a larger amount of supply air with the displacement ventilation. The difference is especially large during summer, when the cooling load is high. However, the fan energy for displacement ventilation in August in Seattle is lower than that for the mixing system. The outdoor temperature in August in Seattle is low, and the outdoor air conditions are good for free cooling. Displacement ventilation has a larger temperature difference between the supply and exhaust air during a free cooling period, which is typically 16F (9 K),
103
while the mixing system has only 9F (5 K). Therefore, the total amount of air for the displacement ventilation is smaller. Consequently, the fan energy consumed is also smaller. Very similar results can be found in other climatic regions, e.g., June, August, and September in Portland, ME, and March and November in Phoenix, AZ (Chen et al. 1999). In the winter, the heat for the displacement system is mainly supplied via the baseboard heater. The displacement system supplies only fresh air, and the amount of air supply is much lower than that of the mixing system. The fan energy consumption also should be lower. Since the office has a high internal heat gain, as do most U.S. office buildings, cooling is required during some office hours even in winter. For these cooling times, the displacement ventilation system uses more fan energy, as explained above. The fan energy consumed during winter is similar between the two ventilation systems due to the combined effect. Figure 6.5 also indicates that the energy consumed by the chiller in the displacement ventilation system is also much less. Since the air supply temperature is higher in the displacement ventilation system than in the mixing ventilation system, this allows the displacement system to use more free cooling during the shoulder seasons. The COP value is slightly higher with displacement ventilation (3.0) than with the mixing ventilation (2.9). On the other hand, the exhaust air temperature in the displacement system is 4F (2 K) higher than that in the mixing system. More importantly, the mean room air temperature in the displacement ventilation system is higher than that in the mixing system, due to the temperature stratification. The cooling load in the summer months is lower in the displacement ventilation. All of these factors contribute to smaller energy consumption of the chiller. The energy consumed by the boiler with displacement ventilation is also smaller than that with mixing ventilation. This is especially evident during the winter. With displacement ventilation, the ventilation effectiveness is higher. Maintaining the same air quality, the total amount of fresh air can be reduced. As a result, the energy needed to heat the fresh air becomes less with the displacement ventilation system, as shown in Figure 6.5. We have further studied the impact of different building orientations on the energy consumption. Figure 6.6 shows the annual energy consumption for the individual office in three different building zones: one having an exterior wall and window facing south, one having an exterior wall and window facing north, and one having no exterior walls and windows (core region of a building). All of the other thermal and fluid boundary conditions are the same. The results show that the energy consumption trend is the same for all three building zones. The energy consumption by the boiler is the highest for north-facing zones because of the high heat loss through the exterior wall and window during the heating period. The chiller does not use much more energy in the south-facing zones because the heat gain due to solar radiation is comparable to the heat loss to the moderate outdoor air temperature in Seattle, WA. In the core region where no exte-
104
Figure 6.6
Annual energy consumption of the displacement and mixing ventilation systems for an individual office at different locations of a building in Seattle, WA (maritime climate).
rior windows and walls exist, no heating is needed. Therefore, the separated heating system with baseboard heater can be eliminated in the core region. In most cases, the sum of the energy consumed by displacement ventilation is slightly smaller than that with mixing ventilation. The fan uses higher energy in the displacement ventilation system because of the high cooling loads found in U.S. buildings. The study for the five climatic regionsSeattle, WA (maritime); Portland, ME (cold); Phoenix, AZ (hot and dry); New Orleans, LA (hot and humid); and Nashville, TN (moderate)shows very similar results (Chen et al. 1999). Classroom and Workshop The investigation also compares annual energy consumption by the displacement and mixing ventilation systems for a classroom and a workshop in the five climatic regions. Figure 6.7 shows the results for Seattle, WA. The results of the classroom facing south are also used for the comparison. The energy consumed is normalized by the floor area. A classroom has less heated equipment than an individual office; the occupants are the major heat sources. Each occupant has a smaller floor area in a classroom than in an individual office. Therefore, the amount of fresh air per square foot floor area is higher in a classroom. The fresh air consists of a large portion of cooling and
105
Figure 6.7
Annual energy consumption per unit floor area of the displacement and mixing ventilation systems for three different types of rooms in Seattle, WA (maritime climate).
heating load. Hence, the energy saving in winter becomes more significant in the classroom with displacement ventilation than in the individual office, as shown in Figure 6.7. The workshop used in the present study has a population density similar to the classroom but with more heated equipment. As a result, the workshop needs little auxiliary heating in winter. This investigation has studied an individual office, classroom, and workshop for all five climatic regions (Chen et al. 1999). Generally, the displacement ventilation system uses more fan energy and less chiller and boiler energy than the mixing ventilation system. The overall energy used by the displacement ventilation system is slightly less. 6.4 FIRST COST ANALYSIS FOR U.S. CONDITIONS The first cost analysis has been performed for the individual office building. This study has assumed that the building has 100 identical individual offices, as listed in Table 6.1. This would allow us to select reasonably sized chillers, boilers, and air-handling units and to distinguish the difference between the displacement and mixing ventilation systems. We have further assumed that one-third of the
106
offices are facing south, one-third are facing north, and the other one-third are in the core region. The equipment capacity is selected to simultaneously handle the maximum load in the three zones of the building. With the equipment capacity, the first costs of the air-handling units, chillers, and boilers can be estimated by using the 1998 R.S. Means building construction cost data. Figure 6.8 shows results for the five climatic regions. The costs are for materials and labor but do not include project overhead. Figure 6.8a shows that the first costs of the air-handling units are higher for the displacement ventilation system than for the mixing ventilation system. This is because the displacement ventilation system needs to handle a larger amount of air. In contrast, the displacement ventilation system needs a smaller chiller, as shown in Figure 6.8b, due to a higher air supply temperature and smaller cooling load. However, the boiler size is almost the same between the two ventilation systems (Figure 6.8c). Although the displacement ventilation system needs a slightly lower boiler capacity, it falls to the same category by using the building construction data. The overall first costs, including the costs for the air-handling unit, chiller, and boiler, are shown in Figure 6.9a. The costs do not include those for air distribution, such as ducts. The displacement ventilation system supplies air at floor level and returns at ceiling level. It may be more desirable to use wall cavities for the ducts. However, the mixing ventilation system should use a false ceiling, which will definitely have an impact on construction costs. Though the first cost analysis provides an insight on the costs of major units, the results show only a rough estimation. Figure 6.9a shows that the overall first cost for displacement ventilation is smaller than that for mixing ventilation. This is mainly due to a small chiller, as illustrated in Figure 6.8b. The results for the displacement ventilation system shown in Figure 6.9a do not include the first costs of a separated heating system needed in the perimeter zones of the building. If we include the first cost of the separated heating system, the total first cost for the displacement ventilation become slightly higher. Therefore, the displacement ventilation system is more desirable for the core region of a building where no heating is required. 6.5 CONCLUSIONS This investigation has studied energy consumption of an individual office, a classroom, and a workshop for five U.S. climatic regions. The study has been done for three building zones in an individual office building: one having a south-facing exterior wall and window, one having a north-facing exterior wall and window, and one having no exterior walls and windows. The study uses the energy balance method to calculate hourly cooling load. The load calculation considers the non-uniform temperature distribution in room air. The secondary systems are analyzed by a whole-building simulation program. The energy analysis shows that displacement ventilation may use more fan energy and less chiller and boiler energy than mixing ventilation. The total energy used is
107
Figure 6.8
Comparison of first costs of air-handling units, chillers, and boilers for the displacement and mixing ventilation systems.
108
a) Without the separated heating system for displacement ventilation.
b) With the separated heating system for displacement ventilation. Figure 6.9 Comparison of the total first costs between the displacement and mixing ventilation systems.
109
slightly less for displacement ventilation, although the ventilation rate is increased to handle the high cooling loads found in U.S. buildings. The study has also analyzed first costs of the displacement and mixing ventilation systems for the office building. The displacement ventilation system needs a larger air-handling unit, a smaller chiller, and a boiler similar to that for the mixing ventilation system. The overall costs are lower for displacement ventilation if the system is applied for the core region of a building. In the perimeter zones, the displacement ventilation system needs a separated heating system, which will slightly increase the first and maintenance costs.
110
CHAPTER 7
Design Guidelines
Since U.S. buildings have a high cooling load and complex geometry, the design guidelines available from the literature need revision and further development for U.S. buildings. This chapter presents a ten-step design procedure for displacement ventilation systems. The design guidelines are for the determination of the key parameters in the displacement ventilation system, such as ventilation rate, location and type of supply diffuser, and supply air temperature. Step (1): Judge the Applicability of Displacement Ventilation Displacement ventilation is suitable when the contaminant sources are associated with heat sources and the ceiling height is no less than eight feet. There is also a limitation on the cooling load that can be handled by displacement ventilation. This study shows that the maximum can be as high 38 Btu/(hft2) (120 W/m2) if the ventilation rate is increased and if there is sufficient space for installing large diffusers. When the cooling load is high, the energy consumption with displacement ventilation will increase significantly. Displacement ventilation is especially effective in the premises with high ceilings (open atriums, cinemas, theaters, and industrial buildings). In addition, the displacement ventilation system may require a separated heating system for perimeter zones of a building to prevent downflow from cold windows/walls in winter. This will increase first and maintenance costs. Hence, the displacement ventilation system is the best for the core region where no heating is needed. Step (2): Calculate Summer Design Cooling Load Use a cooling load program or the ASHRAE manual method to calculate the design cooling load of the space in summer. If possible, assume a 1F/ft (2C/m) vertical temperature gradient in the space in the computer calculation because the air temperature in a room with displacement ventilation is not uniform.
111
Itemize the cooling load into the occupants, desk lamps, and equipment, Qoe (Btu/h, W); the overhead lighting, Ql (Btu/h, W); and the heat conduction through the room envelope and transmitted solar radiation, Qex (Btu/h, W).
Step (3): Determine the Required Flow Rate of the Supply Air for Summer Cooling Displacement ventilation creates a thermal stratification. To maintain a comfort level, the design air temperature difference between the head and foot level of a sedentary occupant, Thf, should be less than 3.6F (2 K). The required ventilation rate, n, can be determined according to Equation 4.25:
3600 - ( a Q + a l Q l + a ex Q ex ) n = ----------------------------T hf C p HA oe oe 1 - ( a Q + a l Q l + a ex Q ex ) n = ----------------------------T hf C p HA oe oe
(7.1a) (I-P)
(7.1b) (SI)
where n Thf Cp H A = ventilation rate (ach) = 3.6F (2 K) = air density (lb/ft3, kg/m3) = specific heat of air (Btu/lbF, J/kgK) = space height (ft, m) = floor area (ft2, m2)
aoe, al, and aex = 0.295, 0.132, and 0.185, respectively; the coefficients stand for the fractions of the cooling loads entering the space between the head and feet of a sedentary occupant ASHRAE Standard 55-1992 requires the temperature difference between the head and foot level of a standing person not to exceed 5F (3 K). Since the vertical temperature gradient in the space between 3.6 ft (1.1 m) and 5.5 ft (1.7 m) is generally smaller than that between 0.3 ft (0.1 m) and 3.6 ft (1.1 m), the design with the above formula can also provide a comfortable condition for a standing person. The flow rate required for summer cooling, Vh, is then Vh = nAH / 60 [cfm] Vh = n AH / 3600 [m3/s] (7.2a) (7.2b)
112
Design Guidelines
Step (4): Find the Required Flow Rate of Fresh Air for Acceptable Indoor Air Quality Use ANSI/ASHRAE Standard 62 (ASHRAE 2001) to find the required flow rate for acceptable indoor air quality, Vr. This flow rate is based on mixing ventilation with a ventilation effectiveness of one. As mentioned earlier, displacement ventilation has higher ventilation effectiveness. For the same indoor air quality, displacement ventilation requires less fresh air than mixing ventilation. The required flow rate of fresh air for displacement ventilation, Vf, can be calculated by Vf = Vr / , where = the ventilation effectiveness at the breathing level. The can be determined by
= 3.4 ( 1 e
0.28 n
(7.3)
) ( Q oe + 0.4 Q l + 0.5 Q ex ) Q t .
(7.4)
Step (5): Determine the Supply Airflow Rate Choose the greater of the required flow rate for summer cooling, Vh, and that for indoor air quality, Vf, as the design flow rate of the supply air, V. V = max{Vf, Vh} If V = Vf, the air-handling system uses 100% fresh air. Step (6): Calculate the Supply Air Temperature Ts can be determined by the air temperature at the floor level, Tf, and the dimensionless temperature, f. Ts = Tf f Qt / (60CpV) Ts = Tf f Qt / (CpV) where Tf = Th Thf Here, Th is the room design air temperature, and
1 - ( 0.295 Q oe + 0.132 Q l + 0.185 Q ex ) . T hf = ------------ Cp V
(7.5)
(7.6a) (I-P) (7.6b) (SI)
(7.7)
(7.8)
The f can be calculated by Mundt's formula (1992).
113
1 f = ------------------------------------------------------60 V C p 1 1 - ------------------- ----- + -----+1 A r cf 1 f = ------------------------------------------------V Cp 1 1 - +1 ------------- ----- + -----A r cf
(7.9a) (I-P)
(7.9b) (SI)
where = radiative heat transfer coefficient from ceiling to floor, Btu/hft2F r (W/m2K); cf = convective heat transfer coefficient from the floor surface to room air, Btu/hft2F (W/m2K). The radiative and convective heat transfer coefficient can be found from ASHRAE HandbookFundamentals. As a rough estimate, they are close to 1 Btu/ (hft2F) or 5 W/(m2K). Then the exhaust air temperature, Te, can be easily calculated via an energy balance for the entire space. Te = Ts + Qt / (CpV) (7.10)
In practice, one air-handling system often serves several (N) rooms (N > 1). For easy control, the supply air temperature should be the same for these rooms. To ensure the thermal comfort in all the rooms, we choose the highest supply air temperature calculated by Equation 7.6 as the supply air temperature for all the rooms. Tss = max{Ts,i}, i = 1, , N (7.11)
where Tss = supply air temperature for all the rooms, and Ts,i = supply air temperature calculated for room i. In this case, the supply flow rates for these rooms, Vi (i = 1, , N), have to be recalculated. For simplification, Vi can be estimated via
Qt , i V i = ------------------------------------- , C p ( T e , i T ss )
(7.12)
where Qt,i = cooling load in the ith room, and Te,i = exhaust air temperature in the ith room calculated by Equation 7.10. Since Tss Ts, i (i = 1, , N), the supply flow rates with Tss are larger than the ones with Ts,i. The air temperature differences between head and foot levels with Tss are smaller than those with Ts,i. The Vi calculated by Equation 7.12 is not very accu-
114
Design Guidelines
rate, but it is acceptable for most design. Th for the room should be recalculated. If it is too high compared to the design room air temperature, Vi should be increased in small increments until Th is acceptable. An exact Vi can be calculated by solving Equations 7.6 through 7.10 together. The total supply flow rate of the system is
N
Vs = Vi .
i=1
(7.13)
Note that if the difference between the Ts,i is larger then 5F (3 K), the designer should consider using two supply air temperatures, Tss. This would allow some of the spaces being supplied with a lower air temperature and a smaller amount of supply air. Step (7): Determine the Ratio of the Fresh Air to the Supply Air, rf For the room with hazardous or toxic contaminants, such as biotech and chemistry laboratories, the displacement ventilation system uses 100% fresh air, or rf = 100%. The energy loss through the exhaust air can be recovered by a heat exchanger. For other spaces, such as offices, we recommend the use of filtered return air to save energy. The fresh air ratio is determined via
Vf , i - r f = max ------ Vi i = 1, , N ,
(7.14)
where Vf,i = the required flow rate for acceptable indoor air quality determined at Step 4. Step (8): Select Supply Air Diffuser Size and Number The supply air velocity must have an upper limit to avoid draft. Our investigation shows that the maximum face velocity of a diffuser is 40 fpm (0.2 m/s). Based on this face velocity, the total face area of the diffusers for the ith room, Ai, is Ai = Vi / 40 [ft2] Ai = Vi / 0.2 [m2] (7.15a) (I-P) (7.15b) (SI)
If each diffuser has a face area of Ad, the number of diffusers needed for the room, Nd, is Nd = Ai / Ad . (7.16)
According to the space layout, the designer should place the diffusers in the room under the following rules:
115
There should not be large obstacles near the diffusers. The diffusers should be placed on the walls opposite to the exterior walls/windows. The diffusers can be placed in the center of a room, such as around a column. More diffusers should be placed in the spaces with higher cooling load.
Step (9): Check the Winter Heating Situation The displacement ventilation system is ideal for cooling. The system should not be used for heating because the buoyancy and low supply air velocity will drive the hot supply air to the ceiling level. However, Straub (1962) used a diffuser to discharge hot air horizontally at floor level for heating. With a high air supply velocity, the air temperature in the room under heating condition can be rather uniform. For the building perimeter where heating may be necessary in winter, a separate heating system is necessary to offset the heating load from exterior walls and windows when displacement ventilation is used. The perimeter heating system can be baseboard convectors, radiators, and fan coil units at floor level, etc. The heating capacity should be slightly larger than the winter design heating load from exterior walls and windows. This will ensure an upward flow near the exterior walls and windows. The perimeter heating device should be placed near to the exterior walls and windows on the floor level. If a fan coil unit is used, the airflow direction should be upward. The displacement ventilation system for winter heating will condition the room air as if the room has only an internal cooling load. Although heating (by a perimeter heating system) and cooling (by displacement ventilation) are used simultaneously in the winter, we do not waste energy. This is because the supply air temperature is still much higher than the outdoor air temperature. The outdoor air must be heated to the supply air temperature in any system. The ventilation air is heated, but not to as high a temperature as that in mixing ventilation. Step (10): Estimate the First Costs and Annual Energy Consumption Estimate the first costs and annual energy consumption when necessary. The results presented in Chapter 6 are general and valid for most office, classroom, and workshop buildings in the United States. This step is only necessary when a designer wishes to compare the displacement ventilation system with a special type of mixing ventilation system.
116
CHAPTER 8
Conclusions
Many studies have been conducted on performance evaluation and the development of design guidelines for displacement ventilation. However, most of the investigations were for low cooling load (less than 13 Btu/hft2 or 40 W/m2 floor area) in a cold climate, such as Scandinavia. The design guidelines are not suitable for U.S. buildings that have a higher cooling load and different building layouts. In addition, simple models for determining temperature difference between the head and foot level of a sedentary occupant and for ventilation effectiveness at breathing level were not available from the literature. In order to develop design guidelines for U.S. buildings, we used a CFD program to develop simple models by establishing a database of 56 cases of individual offices, large offices, classrooms, and workshops with displacement ventilation. The CFD program was validated by detailed flow fields and the distributions of air temperature and contaminant concentration measured in a climate chamber. The agreement is good between the results computed by CFD and the experimental data. The two simple models developed can predict the temperature difference between head and foot levels and ventilation effectiveness at breathing level. The calculated results agree with the results from CFD and from the literature. Compared with conventional mixing ventilation, displacement ventilation may provide better indoor air quality when the contaminant sources are combined with heat sources. The displacement ventilation may maintain a thermally comfortable environment. The percentage of dissatisfied people due to draft (PD) and the predicted percentage of dissatisfied with the thermal comfort (PPD) are normally less than 15%. The mean age of air in a room with displacement ventilation is younger than that with mixing ventilation. With a careful design, the air temperature difference between the head and foot levels can be less than 4F (2C). With increased ventilation flow rate, the displacement ventilation system can remove a cooling load as high as 40 Btu/hft2 (120 W/m2) of floor area without draft. A high ventilation rate will increase the energy consumed by the fan and require a larger air-handling unit. Nevertheless, displacement ventilation can use more free
117
cooling, so it needs a smaller chiller. The total energy consumption by displacement ventilation is slightly smaller than that by mixing ventilation. The first costs are similar between the displacement and mixing ventilation systems. Since displacement ventilation requires a separate heating system for the perimeter zones of a building, it will increase slightly the first costs. The investigation has developed a ten-step design guideline for displacement ventilation systems in U.S. buildings. The system seems the best for the core region of a building or a building with low cooling load.
118
Nomenclature
A Ar C c Cp d g H h h l Q q R s T U u U u V = = = = = = = = = = = = = = = = = = = = = area, ft2 (m2) Archimedes number, ghTr/us2 dimensionless contaminant concentration contaminant concentration, ppm specific heat at constant pressure, Btu/(lbF) (kJ/[kgK]) hydraulic diameter of a heat source, ft (m) gravity, ft/s2 (m/s2) height of a room, ft (m) height of a diffuser, ft (m) heat transfer coefficient, Btu/hft2F (W/m2K) length of a line heat source, ft (m) heat source or cooling load, Btu/h (W) heat flux, Btu/hft2 (W/m2) thermal resistance, hFft2/Btu (Cm2/W) vertical temperature gradient, F/ft (C/m) temperature, F (C) dimensionless velocity, u/us velocity, fpm (m/s) dimensionless fluctuating velocity, u u s fluctuating velocity, fpm (m/s) volume flow rate, ft3/h (m3/h)
Greek Symbols = = heat transfer coefficient, Btu/(hFft2) (W/[Cm2]) gas thermal expansion coefficient, 1/K
119
= = =
difference dimensionless temperature, (T Ts) / (Te Ts) density, lb/ft3 (kg/m3)
Subscripts a c e f r s t = = = = = = = air convection or ceiling exhaust floor radiation or room supply or surface total, time, transmission
120
References
Akimoto, T., T. Nobe, and Y. Takebayashi. 1995. Experimental study on the floorsupply displacement ventilation system. ASHRAE Transactions 101(2). Axelrad, R. 1989. Economic implications of indoor air quality and its regulation and control. The NATO committee for the Challenges of Modern Society, Enrice. ASHRAE. 2001. ANSI/ASHRAE Standard 62-2001, Ventilation for Acceptable Indoor Air Quality. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. ASHRAE. 1992. ANSI/ASHRAE Standard 55-1992, Thermal Environmental Conditions for Human Occupancy. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. ASHRAE. 1997. 1997 ASHRAE HandbookFundamentals, Chapters 27-30. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. Baturin, V. 1972. Fundamentals of Industrial Ventilation. Pergamon Press. Belin, K. 1978. Allmaen ventilation med displacerande stroemning, BFR-rapport R 77:1978, Stockholm. Brohus, H., and P.V. Nielsen. 1994. Contaminant distribution around persons in rooms ventilated by displacement ventilation. Proceedings of RoomVent '94: Air Distribution in Rooms, Krakow, Poland. Brohus, H., and P.V. Nielsen. 1996. Personal exposure in displacement ventilated rooms. Indoor Air 6: 157-167. CHAM. 1996. PHOENICS Version 2.1, CHAM, Ltd., U.K. Cheesewright, R., K.J. King, and S. Ziai. 1986. Experimental data for validation of computer codes for prediction of two-dimensional buoyant cavity flows. In Significant Questions in Buoyancy Affected Enclosure or Cavity Flows, (ed. by J.A.C. Humphrey, C.T. Adedisian and B.W. le Tourneau), ASME, 75-81. Chen, Q. 1988. Indoor airflow, air quality and energy consumption of buildings, Ph.D. thesis. Delft University of Technology, Delft.
121
Chen, Q. 1995. Comparison of different k-e models for indoor airflow computations. Numerical Heat Transfer, Part B, 28, pp. 353-369. Chen, Q. 1996. Prediction of room air motion by Reynolds-stress models. Building and Environment, 31(3), pp. 233-244. Chen, Q., and J. Kooi van der. 1988. ACCURACYA computer program for combined problems of energy analysis, indoor airflow and air quality. ASHRAE Transactions 94(2): 196. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. Chen, Q., J. Kooi van der, and A. Meyers. 1988. Measurements and computations of ventilation efficiency and temperature efficiency in a ventilated room. Energy and Buildings 12(2): 85-99. Chen, Q., C.A. Meyers, and J. Kooi van der. 1989. Convective heat transfer in rooms with mixed convection. Proceedings of International Seminar on Air Flow Patterns in Ventilated Spaces, University of Liege, Belgium, pp. 69-82. Chen, Q., L.R. Glicksman, X. Yuan, S. Hu, Y. Hu, and X. Yang. 1999. Performance evaluation and development of design guidelines for displacement ventilation, Final Report for ASHRAE RP-949, 230 pages, Department of Architecture, Massachusetts Institute of Technology, Cambridge, Mass. Clark, R.P., and O.G. Edholm. 1985. Man and His Thermal Environment. Edward Arnold (Publishers) Ltd. Danielsson, P.O. 1987. Convective flow and temperature in rooms with displacement system. ROOMVENT 87, Stockholm. Fanger, P.O., A.K. Melikov, H. Hanzawa, and J. Ring. 1989. Turbulence and draft. ASHRAE Journal 31(7): 18-23. Fisk, W.J. 2000. Health and productivity gains from better indoor environments and their relationship with building energy efficiency. Annu. Review Energy Environ. 25, 537-566. Fitzner K. 1989. Foerderprofil einer Waermequelle bei verschiedener Temperaturgradienten und der Einfluss auf die Raumstroemung bei Quellueftung, Ki Klime-Kaelte-Heizung, Nr. 10. Flatheim, G. 1984. Air-conditioning without draft and noise. Proc. Indoor Air 84, Stockholm. Fukao, H., M. Oguro, K. Hiwatashi, and M. Ichihara. 1996. Environment evaluation in an office with floor-based air-conditioning system in an office building. ROOMVENT 96, Yokohama, Japan, Vol. 3, pp. 315-322. Glicksman, L.R. L.K. Norford, G.M. Okutan, and K.J. Holden. 1996. Scale model studies of displacement ventilation. Proc. of ROOMVENT '96 Conference, Yokohoma, Japan, July 17-19, Vol. 3, pp. 347-354. Heiselberg, P. 1993. Draught risk from cold vertical surfaces. Proc. of the 6th International Conference on Indoor Air Quality and Climate INDOOR AIR 93, Helsinki.
122
References
Heiselberg, P., and M. Sandberg. 1990. Convection from a slender cylinder in a ventilated room. Proceedings of ROOMVENT '90: International Conference on Engineering Aero- and Thermodynamics of Ventilated Rooms, Oslo. Holmberg, R.B., K. Folkesson, L-G. Stenberg, and G. Jansson. 1987. Experimental analysis of office climate using various air distribution methods. ROOMVENT 87, Stockholm. ISO. 1990. ISO standard 7730, Moderate thermal environments-determination of the PMV and PPD indices and specification of the conditions for thermal comfort. Geneva, International Standards Organization. Jarmyr, R. 1982. Displacerande inblasningNagra erfarenheter, Tekn. Meddelanden o. 247, Inst. For Uppv.- o Vent. Teknik, KTH, Stockholm. Kegel, B. and U.W. Schulz. 1989. Displacement ventilation for office buildings. Proc. of 10th AIVC Conf., Finland, Vol. 1. Klein, S.A., W.A. Beckman, et al. 1994. TRNSYS: A transient simulation program. Engineering Experiment Station Report 38-12, University of Wisconsin, Madison. Kofoed, P., and P.V. Nielsen. 1990. Thermal plume in ventilated rooms. ROOMVENT 90, Oslo. Kruehne, H., and K. Fitzner. 1993. stroemungsuntersuchungen am Klimasystem Quellueftung und Deckenkuehlung, Ki Klima-Kaelte-Heizung, No. 11. Launder, B.E., and D.B. Spalding. 1974. The numerical computation of turbulent flows. Computer Methods in Applied Mechanics and Energy, 3, pp. 268-289. Li, X., and Y. Jiang. 1996. Calculation of age-of air with velocity field, Paper presented in Post-IAQ96 Seminar, Beijing. Li, Y., M. Sandberg, and L. Fuchs. 1992. Vertical temperature profiles in rooms ventilated by displacement: Full-scale measurement and nodal modelling. Indoor Air, Vol. 2, pp. 225-243. Mattsson, M., and M. Sandberg. 1994. Displacement ventilationInfluence of physical activity. ROOMVENT 94, Krakow. Melikov, A.K., and J.B. Nielsen. 1989. Local thermal discomfort due to draft and vertical temperature difference in rooms with displacement ventilation. ASHRAE Transactions 95(2). Mierzwinski, S., and Z. Popiolek. 1981. The convective plume above the human body as a factor of the ventilation process, Technical University, Gliwice, Poland. Mundt, E. 1990. Convective flow above common heat source in rooms with displacement ventilation. Proc. of the International Conference on Engineering Aero- and Thermaodynamics of Ventilated Room, ROOMVENT 90, Oslo. Mundt, E. 1992. Convection flows in rooms with temperature gradientsTheory and measurements. Proceedings of ROOMVENT '92: Third International Conference on Air Distribution in Rooms, Copenhagen.
123
Mundt, E. 1996. The performance of displacement ventilation system, Ph.D. thesis, Royal Institute of Technology, Sweden. Murakami, S., S. Kato, and J. Zeng. 1997. Flow and temperature field around human body with various room air distributionCFD study on computational thermal manikin: Part 1. ASHRAE Transactions 103(1). Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. Nichel, J. 1990. Air Distribution in displacement ventilation (In Danish), VVS, Teknisk Forlag A/S, Copenhagen, marts. Nielsen, P.V. 1988. Displacement ventilation in a room with low-level diffusers. Presented at the Kaelte-Klima-Tagung 1988, Deutscher Kaelte- und Klimatechnischer Verein e. V., Munich. Nielsen, P.V. 1992. Air distribution systemRoom air movement and ventilation effectiveness. Proc. of the ISRACVE Conference, Tokyo, Society of Heating, Air-Conditioning and Sanitary Engineers of Japan. Nielsen, P.V. 1993. Air distribution in roomsRoom air movement and ventilation effectiveness. International Symposium on Room Convection and Ventilation Effectiveness, ISRACVE, ASHRAE. Nielsen, P.V. 1993. Displacement ventilationTheory and design. Department of Building technology and Structural Engineering, Aalborg University, ISSN 0902-8002 U9306. Nielsen, P.V. 1996. Temperature distribution in a displacement ventilated room. ROOMVENT 96, Yokohama, Japan, Vol. 3, pp. 323-330. Nielsen, P.V., L. Hoff, and L.G. Pedersen. 1988. Displacement ventilation by different types of diffusers. Proc. of the 9th AIVC Conference, Warwick. Nielsen, P.V., A. Restivo, and J.H. Whitelaw. 1978. The velocity characteristics of ventilated rooms. J. of Fluid Engineering 100: 291-298. Niemela, R., and H. Koskela. 1996. Air flow patterns in a large industrial hall with displacement ventilation. ROOMVENT 96, Yokohama, Japan, Vol. 3, pp. 363-370. Niu, J. 1994. Modelling of Cooled-Ceiling Air-Conditioning Systems, Ph.D. Thesis, Delft University of Technology, The Netherlands. Olesen, B.W., M. Koganei, G.T. Holbrook, J. Seelen, and J.E. Woods. 1994. Evaluation of a vertical displacement ventilation system. Building and Energy 29:303-310. REHVA. 2002. Displacement Ventilation in Non-Industrial Premises, ed. Skistad, H. et al. Saeteri, J. 1992. A breathing mannequin for measuring local ventilation effectiveness. ROOMVENT 92, Aalborg. Sandberg, M. 1985. Luftutbyteseffektivitet, ventilationseffektivitet, temperatureffektivitet I cellkontor. System med luft som energibaerare. Statens institut foer byggadsforskning, Meddelande M85:24.
124
References
Sandberg, M., and C. Blomqvist. 1989. Displacement ventilation systems in office rooms. ASHRAE Transactions 95(2). Seppanen, O.A., W.J. Fisk, J. Eto, and D.T. Grimsrud. 1989. Comparison of conventional mixing and displacement air-conditioning and ventilating systems in U.S. commercial buildings. ASHRAE Transactions 95(2). Skaret, E. 1985. Ventilation by DisplacementCharacterization and Design Application. Ventilation 85, edited by Goodfellow, Elsevier Science Publishers., B.V. Amsterdam. Skistad, H. 1989. Fortrengningsventilasjon I komfortanlegg med lavimpuls lufttilforel I oppholdssonene, Norsk VVS Teknisk Forening, Oslo. Skistad, H. 1994. Displacement Ventilation. Taunton, Somerset, England: Research Studies Press Ltd. Sowell, E.F., and D.C. Hittle. 1995. Evolution of building energy simulation methodology. ASHRAE Transactions 101: 850-855. Spengler, J.D. 1995. From the environmentally-challenged city to the ecological city. Proc. of Conference on Cities in North America, NY. Stabi, V. 1988. Ed. by Sorensen, H.H., Teknisk Forlag A/S, Kobenhavn. Straub, H.E. 1962. What you should know about room air distribution? Heating, Piping & Air Conditioning, January, 209-220. Stymne, H., M. Sandberg, and M. Mattsson. 1991. Dispersion pattern of contaminants in a displacement ventilated room. Proceedings of the 12th AIVC Conference, Ottawa. Svensson, A.G.L. 1989. Nordic experiences of displacement ventilation systems. ASHRAE Transactions 95(2). Taki, A.H., D.L. Loveday, and K.C. Parsons. 1996. The effect of ceiling temperatures on displacement flow ad thermal comfortExperimental and simulation studies. ROOMVENT 96, Yokohama, Japan, Vol. 3, pp. 307-314. Tanabe, S., and K. Kimura. 1996. Comparisons of ventilation performance and thermal comfort among displacement, under floor and ceiling based air distribution system by experiments in a real sized office chamber. ROOMVENT 96, Yokohama, Japan, Vol. 3, pp. 299-306. Woods, J.E. 1989. Cost avoidance and productivity in owning and operating buildings. Occupational Medicine: State of the Art Reviews 4(4): 753-770. Wyon, D.P., and M. Sandberg. 1990. Thermal manikin prediction of discomfort due to displacement ventilation. ASHRAE Transactions 96(1). Xu, W. 1998. New turbulence models for indoor airflow simulation, Ph.D. thesis. Department of Architecture, Massachusetts Institute of Technology, Cambridge, Mass. Yakhot V., S.A. Orzag, S. Thangam, T.B. Gatski, and C.G. Speziale. 1992. Development of turbulence models for shear flows by a double expansion technique. Phys. Fluids A 4(7), pp. 1510-1520.
125
Yuan, X., Q. Chen, and L.R. Glicksman. 1998. A critical review on displacement ventilation. ASHRAE Transactions 104(1): 78-90. Zhivov, A.M., and A.A. Rymkevich. 1998. Comparison of heating and cooling energy consumption by HVAC system with mixing and displacement air distribution for a restaurant dining area in different climates. ASHRAE Transactions 104(2).
126
Index
A
air temperature difference between the head and foot level 55, 56, 70-73, 79, 112, 117 airflow pattern 4, 38, 46, 53, 84, 85 air-handling systems 30-33, 100, 101
C
CFD program 4, 5, 34, 35, 39, 42, 45, 46, 52, 53, 56, 57, 70, 72, 79, 81, 84, 117 classrooms 34, 56, 57, 61, 65, 73, 75, 79, 81, 117 climatic regions 33, 100, 104-107 comfort aspects 26 computer simulation 35, 95 contaminant distribution 21, 26, 56, 57, 121 contaminant source type 21, 26 convection from human bodies 24 cooled ceiling panel temperature 25, 28 cooled ceiling panels 3, 12, 13, 25, 26, 28, 29, 32, 33 cooling loads 3, 28, 30, 31, 33, 91, 96, 105, 110, 112
D
database of displacement ventilation 4, 56 design guidelines 4, 7, 21, 33, 34, 42, 111, 117, 122 design procedure 111 diffuser type 8, 13, 18, 55 dimensionless temperature 7, 10, 11, 113, 120 displacement ventilation system 2-4, 7, 28-30, 34, 46, 55, 56, 72, 84, 87, 91, 100, 103-107, 110, 111, 115-118, 121, 124, 125 draft 1, 7, 13, 14, 18, 19, 21, 26, 28, 29, 34, 39, 46, 51, 55, 81, 86, 87, 92, 94, 115, 117, 122, 123 draft risk assessment 26
E
energy analysis 29, 100, 107, 122 energy balance equation for window 98 energy balance on a wall 96 environmental test facility 36 evaluation criteria 81
127
exhausts 2, 7 experimental investigation 35 experimental results 4, 10, 42
F
first cost analysis 32, 106, 107 flow rate in a plume 16 flow visualization system 37
G
governing equations 43, 45
H
heat source type 8, 10, 21, 55, 86 high cooling load 12, 33, 34, 91, 94, 105, 110, 111 hot-sphere anemometer system 37
I
indoor air quality 1, 2, 4, 21, 34-36, 39, 51, 56, 78, 81, 84, 87, 91, 92, 94, 101, 113, 115, 117, 121, 122 industrial workshops 34, 56, 57, 62, 66, 73, 75, 79, 81
L
large offices with partitions 56, 57, 73, 74, 79, 81 load calculations 95
M
maximum height of a plume 15, 16, 21 mean age of air 26, 41, 42, 51-53, 81, 83, 84, 87, 94, 117
N
numerical technique 45
P
percentage of dissatisfied people due to draft 81, 92, 94, 117 performance evaluation 84, 117, 122 perimeter spaces 3, 4 predicted percentage of dissatisfied for the thermal comfort 81
128
Index
R
ratio of the fresh air to the supply air 115 renormalization group k- model 42 required flow rate 112, 113, 115 room air energy balance 96
S
secondary systems 95, 100, 107 sick building syndrome 1 sidewall diffuser 2 small offices 56, 57, 59, 63, 73, 74, 79, 81 space height 7, 8, 13, 21, 25, 55, 57, 112 stratification height 21, 25, 33, 51 stratified temperature 3 supply air diffuser size and number 115 supply air temperature 2, 3, 13, 31, 32, 34, 39, 70, 86, 100, 101, 111, 113-116
T
temperature distribution 7, 13, 28, 70, 81, 84, 86, 97, 107, 124 temperature gradient 3, 7, 9-13, 15, 16, 21, 26, 28-30, 33, 34, 46, 69, 70, 72-75, 87, 111, 112, 119, 123 temperature profiles 7-9, 12, 13, 28, 47, 123 test procedure 39 thermal plumes 3, 14, 21, 22, 24, 25, 46, 79, 86, 87, 94 thermocouple system 38 tracer-gas system 38, 39 turbulence models 35, 36, 125
V
validation of CFD program 45, 52 ventilation effectiveness 5, 34, 56, 70, 77-79, 83, 84, 87, 91, 94, 104, 113, 117, 124 ventilation effectiveness model 77 vertical temperature gradient 3, 7, 9, 12, 13, 15, 16, 26, 29, 30, 33, 34, 70, 72, 87, 111, 112, 119
W
wall surface temperature 21, 25
129
This document is the errata for the following book published by ASHRAE: Q. Chen and Glicksman, 2003. System Performance Evaluation and Design Guidelines for Displacement Ventilation, ASHRAE, Atlanta, GA. Page 112 in replacement of Eqs. (7.1a) and (7.1b):
n=
n=
1 (a Q + a l Q l + a ex Q ex ) Thf C p HA oe oe
3600 (a oe Q oe + a l Q l + a ex Q ex ) Thf C p HA
(I-P) (SI)
(7.1a) (7.1b)
Page 113 in replacement of Eq. (7.8):
Thf =
Thf =
1 (0.295 Q oe + 0.132 Q l + 0.185 Q ex ) C p V

1 (0.295 Q oe + 0.132 Q l + 0.185 Q ex ) 60C p V
(I-P) (SI)
(7.8a) (7.8b)
Page 114 in replacement of Eq. (7.10): Te = Ts + Qt/(60CpV) Te = Ts + Qt/(CpV) Page 114 in replacement of Eq. (7.12): (I-P) (SI) (7.10a) (7.10b)
Vi =
Vi =
60C p (Te,i Tss )

C p (Te,i Tss ) Q t ,i
Q t ,i
(I-P) (SI)
(7.12a) (7.12b)

ASHRAE - SystemPerformanceEvaluation

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

ASHRAE - SystemPerformanceEvaluation

Caricato da

Copyright:

Formati disponibili

System Performance Evaluation and Design Guidelines for Displacement Ventilation

About the Authors

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Qingyan Chen Leon Glicksman

American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

PUBLISHER W. Stephen Comstock

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Sketch of displacement ventilation.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Simplified vertical temperature profile in a room with displacement ventilation.

Temperature profiles in office rooms.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Dimensionless temperature near the floor vs. supply flow rate.

1 sH -------- = 1 f = 1 ------------------------------------------------ . V cp 1 T e 1 ------------ ----- + ------ +1 A r cf

Dimensionless temperature gradient vs. supply flow rate.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Comparison of the temperatures between the models and the experiment.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

A thermal plume above a heated object.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Volume flow rate around and above a person.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Figure 2.10 A typical velocity distribution near a diffuser (Nielsen 1993).

umax, x = Kdr(h/x)uf where h uf = diffuser height

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Figure 2.11 Measured value for some diffusers (Nielsen 1993).

Figure 2.12 Design chart for choosing a diffuser.

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

The Costs of Annual Energy Consumption

System Performance Evaluation and Design Guidelines for Displacement Ventilation

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Experimental Study and Validation of CFD Program

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Sketch of the test facility.

Experimental Study and Validation of CFD Program

Height Length Width Height Partition height window width

Climate Chamber 10 ft (3.08 m)

Partition wall Partition window Other walls Ceiling Floor Door

System Performance Evaluation and Design Guidelines for Displacement Ventilation

The control interface of the HVAC system.

Experimental Study and Validation of CFD Program

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Experimental Study and Validation of CFD Program

Specifications for the Experimental Measurements

Type Small office Large office classroom

Type Small office Large office classroom

System Performance Evaluation and Design Guidelines for Displacement Ventilation

c(t) c(0) - dt 1 ---------------------------c() c(0)

The Measured Parameters

Experimental Study and Validation of CFD Program

,eff = effective diffusion coefficient; and

System Performance Evaluation and Design Guidelines for Displacement Ventilation

For kinetic energy of turbulence:

For dissipation rate of turbulent kinetic energy:

Experimental Study and Validation of CFD Program

where q hc Tw = heat flux = convective heat transfer coefficient = wall temperature

System Performance Evaluation and Design Guidelines for Displacement Ventilation

Experimental Study and Validation of CFD Program