and Design Guidelines for
Displacement Ventilation
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 Tech-nology. 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 edito-rial 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 ASHRAE’s International Journal of HVAC&R Research.
Qingyan Chen
Leon Glicksman
American Society of Heating, Refrigerating
and Design Guidelines for
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 inves-tigated, 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 endorse-ment, 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 illustra-tions 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 means—electronic, photocopying, recording, or other—without permission in writing from ASHRAE.
ASHRAE S
TAFF
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 ComstockPREFACE . . . .vii
ACKNOWLEDGMENTS. . . ix
CHAPTER 1—INTRODUCTION . . . 1
1.1 Displacement Ventilation. . . 2
1.2 Special Features in U.S. Buildings . . . 3
1.3 Objective of This Book . . . 4
CHAPTER 2—LITERATURE 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 3—EXPERIMENTAL 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 4—MODELS 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
CHAPTER 5—PERFORMANCES EVALUATION
OF DISPLACEMENT VENTILATION. . . 81
5.1 Evaluation Criteria . . . 81
5.2 Performance Evaluation of Displacement Ventilation . . . 84
5.3 Discussion . . . 91
5.4 Conclusions . . . 94
CHAPTER 6—ENERGY 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 7—DESIGN GUIDELINES . . . 111
CHAPTER 8—CONCLUSIONS . . . 117
NOMENCLATURE . . . 119
REFERENCES . . . 121
This book presents system performance evaluation and design guidelines for displacement ventilation.
The authors first reviewed the literature concerning the performance of tradi-tional 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 full-scale 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 contami-nant 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.
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.
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 infil-tration 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 build-ing envelope is made tighter. However, such a reduction of air supply causes an increase in the concentration of indoor pollutants and sometimes generates a non-uniform distribution of air temperature and contaminant concentration. Draft (ther-mal 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 govern-ment 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
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 verti-cal 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. build-ing in which the coolbuild-ing 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/(h⋅ft2) (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 strat-ification 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 Scandina-vian 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 parti-tions to form individual work staparti-tions 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 follow-ing 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. Literature review to identify the existing results and problems
2. Experimental study to get reliable data, including the distribution of velocity, turbulence intensity, temperature, tracer-gas concentration, etc.
3. Validation of a computational fluid dynamics (CFD) program by the experimental data to determine the accuracy of the program
4. Numerical simulation of a large number of cases by the CFD program to establish a database on the performance of displacement ventilation
5. Model development to develop models needed for design guidelines 6. Energy and cost analysis to assess the impact of energy and first costs 7. 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 venti-lation. 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 ther-mal and flow conditions for different types of U.S. buildings. Based on the computed
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.
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 distri-bution 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 temper-ature 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, θ = (T – Ts / Te – Ts), in normal offices obtained from several different investigations. The dimen-sionless temperature near the floor, θf, 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
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
Figure 2.1 Simplified vertical temperature profile in a room with displacement ventilation.
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 θf 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 θf is a function of ventilation flow rate. The θf decreases from 0.4 to 0.2 when the air change rate increases from 3 to 5 ach.
Simi-lar results can be found from Mundt (1990) and Li et al. (1992).
Mundt (1990) developed a formula to calculate θf 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 θf is a function of ventilation rate given as
(2.1)
where
V = ventilation flow rate, ρ = air density,
Cp = specific heat at constant pressure, A = floor area,
αr = radiative heat transfer coefficient from ceiling to floor, and
αcf = 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 αr = 0.9 Btu/ (h⋅ft2⋅°F) (5 W/m2⋅K) and αcf = 0.7 Btu/(h⋅ft2⋅°F) (4 W/m2⋅K). 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 5°F (2.8°C) in the experiment. Equation 2.1 accounts for the impact of cooling load on θf 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 – θf) (Te – Ts) / H (2.2) or θf 1 VρCp A --- 1 αr --- 1 αcf ---+ 1+ --- , =
(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 refer-ences 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 concen-trated heat source, the gradient of dimensionless temperature decreases slightly as the Archimedes number (gβh∆Te/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
sH Te ∆ --- 1–θf 1 1 Vρcp A --- 1 αr --- 1 αcf ---+ 1+ --- . – = =
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).
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 0.23°F/ft (0.38 K/m) in the early morning to 0.42°F/ft (0.7 K/m) at noon and then decreased slightly to 0.39°F/ft (0.65 K/m) in the evening. The nondimensional temperature near the floor, θf, 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 multi-node 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 temper-ature 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 Figure 2.6 Comparison of the temperatures between the models and the
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 venti-lation could be transformed into mixing ventiventi-lation.
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 eleva-tion 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 diffus-ers 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 occu-pied 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 temper-ature 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 estima-tion sometimes gives erroneous predicestima-tion 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.
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.
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 = 0.005Qc1/3 (y + y0) 5/3, (2.4)
where
y + y0 = 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 Qc = convective heat emission from the source.
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 surround-ings.
In an environment with temperature stratification, such as a space with displace-ment 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. There-fore, 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) (2.6)
ymax = 0.98 Qc1/4 s–3/8 – y0 (2.7)
where
V = flow rate in a plume, m3/s ymax = maximum height of the plume, m s = air temperature gradient, K/m Qc = convective heat emission, W y = height above the heated object, m
y0 = 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 person’s head. According to Clark and Edholm (1985), the flow rate may increase as the surrounding temperature decreases because of the increase of a person’s metabo-lism. 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 0.9°F/ft (0.5 to 1.5°C/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).
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).
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 18°F (10°C) between the cold wall surface and the room air.
Impact of Diffusers
Since relatively cold air is supplied directly to the occupied zone in displace-ment 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 typi-cal 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 ambi-ent 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 (gβh∆Te/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:
umax, x = Kdr(h/x)uf (2.10)
where
h = diffuser height
uf = 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). Skis-tad (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).
Figure 2.11 Measured value for some diffusers (Nielsen 1993).
Conclusions
Stratification height, which is a function of the flow rates of supply air and ther-mal 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 loca-tion, human body convecloca-tion, 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 contam-inant 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).
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 venti-lation efficiency increases as the ventiventi-lation 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 Figure 2.14 Concentration profiles with different types of tracer gases
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).
combined with a weak heat source. The thermal plumes are too weak to reach the upper zone.
Mundt’s (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 ther-mal 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 concen-tration in the inhaled air is 0.58ce—the same as that at 1.7 ft (0.5 m) below the breath-ing level—although the concentration outside the thermal boundary layer around the person at the breathing level is 0.96ce. They found the concentration of inhaled 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).
contaminant Ci may be expressed as a linear function of the stratification height, yst, as follows:
Ci = Cy –(Cy – Cf) yst / yb (yst < yb) (2.11) where
Cy = concentration outside the thermal boundary layer around the person at the breathing height, yb
Cf = concentration at the floor
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.
Skistad’s (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 68°F (20°C).
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 Koskela’s (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 contam-inant 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 venti-lation may instead take the form of mixing ventiventi-lation. Brohus and Nielsen (1996)
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 mounted displacement system, and the floor-mounted system is better than a ceiling-floor-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 concen-tration 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 ceil-ings, 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-Figur e 2.18 P re dicted per centage of discomf o rt (a) ab o v
e chair height and (b) belo
w chair height (W y o n and Sandber g 1990).
ature. The results showed that the ankle and foot (below chair height) are more sensi-tive 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 72°F to 81°F (22.1°C to 27.0°C). 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 temper-atures of about 68°F (20°C), and velocities of up to 0.83 ft/s (0.25 m/s) seem accept-able 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 envi-ronment 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 occu-pied zone, they found that 33% of measured locations had higher than 15% of dissat-isfied people due to draft. Also, 40% of the locations were found to have a temperature difference between head and foot of larger than 5.4°F (3.0°C).
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/(h⋅ft2) (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 cool-ing load. Olesen et al. (1994) found that no thermal comfort problems existed under the tested conditions with the cooling loads up to 14 Btu/(h⋅ft2) (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/(h⋅ft2) (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 signif-icant 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 59°F (15°C), 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.
Conclusions
Large vertical temperature gradient and draft are the two main causes of discomfort with displacement ventilation. Previous research shows that displace-ment ventilation without cooled ceiling panels is suitable for spaces with a cooling load less than 13 Btu/(h⋅ft2) (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/(h⋅ft2) (50 W/m2). It is important that the surface temperature of the panels should not be lower than 59°F (15°C). Low flow rate fans at floor level may reduce the vertical temperature differ-ence 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 measure-ments for a building based on a yearly basis.
Energy Analysis
Seppanen et al. (1989) evaluated the energy performance of displacement venti-lation systems and mixing ventiventi-lation systems in U.S. office buildings. The study is 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).
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/(h⋅ft2) (14 W/m2), and the maximum load is 7.5 Btu/(h⋅ft2) (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 temper-ature 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) (M–missing ventilation, D–displacement ventilation).
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 investiga-tions.
Niu’s (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.
Table 2.1 Heating and Cooling Loads for Each Location and Representative Zone in the U.S.
Atlanta El Paso Minneapolis Seattle
Max. Ave. Max. Ave. Max. Ave. Max. Ave.
North zones Heating (Btu/h⋅ft2) Heating (W/m2) Cooling (Btu/h⋅ft2) Cooling (W/m2) 19.6 62.0 22.2 70.0 5.2 16.3 10.1 31.9 16.8 53.1 24.6 77.5 4.3 13.6 11.7 36.9 14.1 44.5 17.5 55.3 4.6 14.4 7.8 24.6 9.1 28.7 15.7 49.5 3.2 10.0 7.1 22.3 Core zones Heating (Btu/h⋅ft2) Heating (W/m2) Cooling (Btu/h⋅ft2) Cooling (W/m2) 0.0 0.0 7.5 23.6 0.0 0.0 4.4 14.0 0.0 0.0 7.5 23.6 0.0 0.0 4.4 14.0 0.0 0.0 7.5 23.6 0.0 0.0 4.4 14.0 0.0 0.0 7.5 23.6 0.0 0.0 4.4 14.0 South zones Heating (Btu/h⋅ft2) Heating (W/m2) Cooling (Btu/h⋅ft2) Cooling (W/m2) 19.2 60.5 37.6 118.7 5.0 15.8 14.1 44.5 16.8 53.0 41.4 130.5 4.7 14.8 17.6 55.6 16.0 50.3 36.9 116.4 5.0 15.9 12.4 39.0 9.1 28.6 35.9 113.2 3.3 10.3 11.3 35.7
Table 2.2 The Costs of Annual Energy Consumption Ventilation
Systems Air-Handling Systems
Supply Air Temperature °F (°C)
Energy Consumption ($/m2)
Displacement Variable air volume 61 (16) 108 Mixing Variable air volume 55 (12.5) 104 Mixing Variable air volume 61(16) 126 Displacement Constant air volume - 241 Mixing Constant air volume - 222
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 4°F to 6°F (2°C to 3°C) 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) (M—mixing ventilation, D–displacement ventilation).
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 air-handling 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 avail-able.
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 gradi-ent
• 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. Determine the required airflow rate for removal of surplus heat based on the cool-ing load and the air temperature difference between supply and exhaust opencool-ings. 2. Find the required airflow rate for removal of pollutants according to ventilation
standards.
3. Choose the larger of the two flow rates determined at Steps 1 and 2 as the venti-lation rate.
4. Determine supply air temperature under assumptions of θf = 0.5 and constant vertical temperature gradient.
5. 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 guide-book (REHVA 2002).
Despite the simple design guidelines, there are problems. Figure 2.2 shows that θf 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 informa-tion on indoor airflow, indoor air quality, and thermal comfort for various configu-rations 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 clar-ification so that designers can use the design guidelines with confidence. U.S. build-ings, especially perimeter zones of such buildbuild-ings, 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 work-shops.
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 perfor-mances. 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.
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 trans-port in buildings: experimental investigation and computer simulation. Experimen-tal 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 anal-ysis. Unfortunately, not many of the experimental data include such detailed infor-mation. 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-ε model (Launder and Spalding 1974) is probably most widely used in engineering calcula-tions due to its relative simplicity. However, the model sometimes provides poor results for indoor airflow. Many modifications have been applied to the standard
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 differ-ent turbulence models. His study concluded that the re-normalization group (RNG) k-ε model (Yokhot et al. 1992) is the best among the eddy-viscosity models tested. This chapter will also compare this model’s 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 iden-tical. 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