Displacement Ventilation - Biblioteka...2017/06/23 · Displacement Ventilation rehva Federation of...
Transcript of Displacement Ventilation - Biblioteka...2017/06/23 · Displacement Ventilation rehva Federation of...
Displacement Ventilation
rehvaFederation of European Heating, Ventilation and Air Conditioning Associations
GUIDEBOOK NO 23
Risto Kosonen (ed.)Arsen Melikov
Elisabeth MundtPanu Mustakallio
Peter V. Nielsen
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
Displacement
Ventilation
Risto Kosonen (ed.)
Arsen Melikov
Elisabeth Mundt
Panu Mustakallio
Peter V. Nielsen
REHVA
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
DISCLAIMER
This Guidebook is the result of the efforts of REHVA volunteers. It has been written with
care, using the best available information and the soundest judgment possible. REHVA and
its volunteers, who contributed to this Guidebook, make no representation or warranty,
expressed or implied, concerning the completeness, accuracy, or applicability of the infor-
mation contained in the Guidebook. No liability of any kind shall be assumed by REHVA
or the authors of this Guidebook as a result of reliance on any information contained in this
document. The user shall assume the entire risk of the use of any and all information in this
Guidebook.
-----------------------------------------------------------------------------------------------------------
Copyright © 2017 by REHVA
REHVA is the Federation of European Heating, Ventilation and Air Conditioning
Associations, www.rehva.eu
All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopies or any other information storage and re-
trieval system, without permission in writing from the publisher.
Requests for permission to make copies of any part of the work should be addressed to:
REHVA Office, 40 Rue Washington, 1050 Brussels – Belgium
e-mail: [email protected]
ISBN 978-2-930521-17-6
Printed in Finland, Forssan Kirjapaino Oy, Forssa
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
iii
List of contents
1 DISPLACEMENT VENTILATION IN A NUTSHELL........................................... 1
2 TERMINOLOGY, SYMBOLS AND UNITS ............................................................ 4
2.1 Terms and definitions ................................................................................... 4
2.2 Symbols ........................................................................................................ 5
3 ROOM AIR DISTRIBUTION .................................................................................... 8
3.1 Need for Ventilation ..................................................................................... 8
3.2 Ventilation and room air distribution principles .......................................... 8
3.3 Displacement ventilation and thermal comfort .......................................... 11
3.4 Displacement ventilation and air quality .................................................... 12
4 PERFORMANCE OF DISPLACEMENT VENTILATION .................................. 15
4.1 Displacement Ventilation Method ............................................................. 15
4.2 Air flow pattern .......................................................................................... 15
4.3 Temperature distribution ............................................................................ 16
4.4 Convection flows – the engines of displacement ventilation ..................... 20
4.5 Contamination distribution ......................................................................... 28
4.6 Ventilation effectiveness ............................................................................ 29
5 CALCULATION OF SUPPLY AIRFLOW RATE ................................................ 34
5.1 Temperature based design methods ........................................................... 34
5.2 Calculation of vertical room air temperature distribution .......................... 35
5.3 Vertical position of the heat source ............................................................ 39
5.4 Calculation examples when using temperature based design models ........ 39
6 AIR DIFFUSERS FOR DISPLACEMENT VENTILATION ................................ 42
6.1 Commonly used diffusers .......................................................................... 42
6.2 Radial air flow or plane air flow from low-velocity diffusers ................... 44
6.3 Air flow from low –velocity diffusers ....................................................... 44
6.4 Air distribution from a low-velocity diffuser giving a radial flow in the
occupied zone ............................................................................................. 45
6.5 Air distribution from wall-mounted diffusers giving plane flow in the
occupied zone ............................................................................................. 53
6.6 Air distribution from floor-mounted diffusers ........................................... 54
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
iv
7 DESIGN OF DISPLACEMENT VENTILATION .................................................. 56
7.1 Design criteria ............................................................................................ 56
7.2 Design of air distribution ........................................................................... 56
7.3 Integration with separate heating and cooling systems .............................. 60
7.4 Control of indoor conditions ...................................................................... 64
8 CASE STUDIES ......................................................................................................... 67
8.1 Air distribution with four typical air supply methods in a classroom ........ 67
8.2 Comparison of calculated and measured vertical temperature gradients for
displacement air distribution ...................................................................... 70
8.3 Field measurements for a multipurpose arena ........................................... 72
9 RESEARCH FINDINGS ........................................................................................... 74
9.1 A CFD Benchmark test for manikins in displacement flow ...................... 74
9.2 Full-scale tests and CFD- simulations of indoor climate conditions ......... 74
9.3 Test on the performance of displacement ventilation– proper simulation of
occupants .................................................................................................... 77
9.4 Airborne cross infection risk in a room with displacement ventilation ..... 80
9.5 Displacement ventilation design based on occupants’ response ................ 83
9.6 Convective boundary layer around human body ....................................... 87
10 REFERENCES ........................................................................................................... 91
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
v
Preface
Displacement ventilation is primarily a
means of obtaining good air quality in oc-
cupied spaces that have a cooling demand.
It has proved to be a good solution for
spaces where large supply air flows are re-
quired.
Some advantages of displacement ventila-
tion:
• Less cooling needed for a given tempera-
ture in the occupied space;
• Longer periods with free cooling;
• Potential to have better air quality in the
occupied spaces;
• The system performance is stable with all
cooling load conditions.
Displacement ventilation has been origi-
nally developed in Scandinavian countries
over 30 years ago and now it is also a well-
known technology in different countries
and climates. Historically, displacement
ventilation was first used for industrial ap-
plications but nowadays it is also widely
used in commercial premises.
However, displacement ventilation has not
been used in spaces where it could give
added values. For that there are two main
reasons: firstly, there is still lack of
knowledge of the suitable applications of
displacement ventilation and secondly,
consulters do not know how to design the
system.
REHVA published 2002 the first version of
displacement ventilation guide. The aim of
this revised Guidebook is to give the state-
of-the art knowledge of the technology. The
idea of this guidebook is to simplify and
improve the practical design procedure.
This guide discusses methods of total vol-
ume ventilation by mixing ventilation and
displacement ventilation and the guide
book gives insight of the performance of
the displacement ventilation. It also takes
into account different items, which are cor-
related, to well-known key words: free con-
vection flow; stratification of height and
concentration distribution; temperature dis-
tribution and velocity distribution in the oc-
cupied zone and occupant comfort.
The guide book discusses two principal
methods which can be used when the sup-
ply air flow rate of displacement ventilation
system is calculated: 1) temperature based
design, where the design criterion is the air
temperature in the occupied zone of the
room and 2) air quality based design where
the design criterion is the air quality in the
occupied zone. Some practical examples of
the air flow rate calculations are presented.
The air flow diffusers are the critical factor:
most draught problems reported in rooms
with displacement ventilation are due to
high velocity in the zone adjacent to the dif-
fuser. This guide explains the principle for
the selection of diffuser.
This guide also shows practical case studies
in some typical applications and the latest
research findings to create good micro cli-
mate close to persons is discussed.
These and some other aspects are discussed
in this book. Authors believe you will find
this guide useful and interesting when you
design or develop new ventilation solu-
tions.
The authors
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
vi
Foreword
REHVA, now 54 years old, is an organisation of European professionals in the field of building
services (heating, ventilating and air–conditioning). REHVA represents more than 100,000 experts
from 27 European countries. REHVA’s mission is to promote energy efficient and healthy technol-
ogies for mechanical services of buildings, and to disseminate knowledge among professionals and
practitioners in Europe and beyond. REHVA Guidebooks are the most important tools to diffuse
knowledge on latest developments, and advanced technologies providing practical guidance to practi-
tioners. REHVA has published 22 guidebooks to date, this one on Displacement Ventilation is the 23rd.
– Anita Derjanecz, REHVA Managing Director
Member countries of REHVA
Belgium | Croatia | Czech Republic | Denmark | Estonia | Finland | France | Germany | Hungary |
Italy | Latvia | Lithuania | Moldavia |Netherlands | Norway | Poland | Portugal | Romania | Russia |
Serbia | Slovakia | Slovenia | Spain | Sweden | Switzerland | Turkey | United Kingdom
Working Group
This book was developed with a working group consisting of the following experts:
• Risto Kosonen (Aalto University, Finland)
• Arsen Melikov (DTU Technical University of Denmark)
• Elisabeth Mundt (KTH Royal Institute of Technology, Sweden)
• Panu Mustakallio (Halton Oy, Finland)
• Peter V. Nielsen (Aalborg University, Denmark)
Reviewers
This book was reviewed with a working group consisting of the following experts:
• Hazim B. Awbi (Reading University, United Kingtom)
• Klaus Fitzner (Technical University of Berlin, Germany)
• Jarek Kurnitski (Tallinn Technical University, Estonia)
• Alfred Moser (Science Services, Switzerland)
• Marco Perino (Technical University of Torino, Italy)
• Jorma Railio (SULVI, Finland)
Acknowledgements
The authors wish to thank REHVA's Technology & Research Committee and Publishing & Marketing
Committee as well as REHVA's staff members for all their valuable contributions to the guidebook.
The authors would also like to thank Tim Dwyer for proof-reading the manuscript and Jarkko
Narvanne for the graphical design and typesetting of the guidebook.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
1
1 Displacement ventilation in a nutshell
The idea Displacement ventilation, as presented in
this book, is considered to be the technique
of supplying clean, cool air at floor level,
letting warm air and contaminants rise to
the ceiling and extracting the contaminated
air at ceiling level (Figure 1.1).
Figure 1.1. The concept of displacement
ventilation.
Best suited for Displacement ventilation is primarily a
means of obtaining good air quality in oc-
cupied spaces that have a cooling demand.
It has proved to be a good solution for:
• Gyms;
• Meeting rooms;
• Classrooms;
• Tall rooms: Convention centres, Lobbies,
Sport arenas, Auditoriums, Theatres, Mu-
seums, Airports, Shopping centres, etc.
Displacement ventilation is usually prefer-
able in the following cases:
• Where the contaminants are warmer
and/or lighter than the surrounding air;
• Where the supply air is cooler than the
ambient air;
• In tall rooms, for example, where the
room heights are more than 3 metres;
• When there are heat loads in the upper
part of room;
• Where large supply air flows are required
in rooms.
Less suited for Displacement ventilation may be less pref-
erable than mixing ventilation in the fol-
lowing cases:
• Where surplus heat is the main problem,
and relatively low specific outdoor air-
flow rate is needed;
• Where there are space constraints for sup-
ply diffusers and duct work;
• When the requirement is to cool in low
height rooms (in offices, consider mixing
and cooling panels or chilled beams);
• Where there are significant disturbances
to air flow near the floor (for example,
furniture);
• Where the contaminants are cooler/
denser than the ambient air.
Strong points Some advantages of displacement ventila-
tion are:
• Less cooling needed for a given tempera-
ture in the occupied space;
• Longer periods with free cooling;
• Potential to have better air quality in the
occupied spaces;
• The system performance is stable with all
cooling load conditions.
Weak points Some weak points are:
• Possibility of cold draughts along the
floor - use the suitable air supply units,
and take care of the zone in front of the
diffusers;
• Sensitive to furniture arrangement in
rooms;
• Displacement principle may be disturbed
by walking occupants;
• Wall mounted diffusers often require sig-
nificant wall space and reduce occupied
zone.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
Rehva Displacement Ventilation Guidebook
2
The air supply diffuser – a crucial factor Most draught problems reported in rooms
with displacement ventilation are due to
high velocity in the zone adjacent to the dif-
fuser (Figure 1.2). It is important to choose
a diffuser that is suited for the application
and only utilise diffusers from manufactur-
ers that supply robust documentation to-
gether with the products.
Figure 1.2. Diffuser has a limited near (adja-
cent) zone with high velocities where the risk
of draught is high (see Chapter 6).
Collaboration with the architect is required The diffusers require a certain amount of
wall area, or space in, or on, the floor. Close
cooperation with the architect is required to
find a suitable location for the air diffusers.
The supply units can also be designed to fit
different architectural requirements: units
could be invisible (Figure 1.3) or be an ex-
posed architectural element (Figure 1.4).
Air flow rates To reach the same air quality in the occu-
pied zone, displacement ventilation typi-
cally requires a lower air flow rate than
mixing ventilation. When the main task is
to remove excess heat, both mixing and dis-
placement systems are likely to require
similar air flow rates.
Figure 1.3. Diffusers integrated in benches
with air supplied through perforated side
plates (courtesy of Halton).
Figure 1.4. Free-standing diffusers as archi-
tectural elements (courtesy of Halton).
The occupied zone – the coolest part of the room In displacement ventilation, the air temper-
ature increases from floor to ceiling (Fig-
ure 1.5). This means that the occupied zone
is the coolest part of the room. Vertical tem-
perature profiles measured with different
individual types of heat load (occupants,
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
1. DISPLACEMENT VENTILATION IN A NUTSHELL
3
warm floor, warm window and warm ceil-
ing) are shown in Figure 1.5 (Kosonen et
al. 2016). With the load dominated by oc-
cupants or by a warm floor, a two layer
structure is generated with heat and pollu-
tion accumulated in the upper part of the
room. The data indicate an obvious mixing
layer. Across the mixing layer, the room air
temperature could be assumed to be con-
stant or exhibiting just a slight increase.
The warm window produces a near linear
temperature profile with no clear two-layer
structure. With the heated ceiling, the con-
vection heat remains mainly in the upper
portion of the room.
Figure 1.5. Vertical temperature profiles in
room with displacement ventilation with differ-
ent heat loads (temperature ratio= 𝜃−𝜃𝑠
𝜃𝑒−𝜃𝑠).
With displacement, the supply air tempera-
ture is typically about 3 K to 5 K cooler
than the room air temperature at a height of
1,1 m. In areas where people are moving
for example, in shopping centres, the sup-
ply air could be 6 K to 8 K lower than the
room air temperature. Depending on the
particular design, the temperature differ-
ence between the supply and exhaust air is
typically between 6 K and 15 K.
Compared with mixing ventilation, dis-
placement ventilation supplies air at a
higher temperature and this implies longer
periods of the year where free cooling can
be applied, and so less energy consumption
for cooling the supply air.
Do not heat occupied rooms with displacement ventilation If an occupied room is to be warmed by the
ventilating air, displacement ventilation, as
described in this book, is not suitable. If
warm air is supplied at floor level, in a room
cooler than the supply air, it will rise due to
buoyancy, and be extracted when it reaches
the ceiling (Figure 1.6). Thus, the supply
air will short circuit into the outlet and little
of the clean and heated warm air will reach
the occupied space. Displacement air distri-
bution can be used for heating up spaces
prior to them being occupied.
Figure 1.6. Supply of warm ventilation air
means short-circuiting.
0
1
2
3
4
5
0 0,4 0,8 1,2
H(m)
Temperature ratio
Occupants
0
1
2
3
4
5
0 0,4 0,8 1,2
H(m)
Temperature ratio
Warmwindow
0
1
2
3
4
5
0 0,4 0,8 1,2
H(m)
Temperature ratio
Warmfloor
0
1
2
3
4
5
0 0,4 0,8 1,2
H(m)
Temperature ratio
Warm ceiling
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4
2 Terminology, symbols and units
2.1 Terms and definitions
Adjacent zone: The zone in front of a dis-
placement air distribution diffuser where
draught discomfort may occur.
Air change rate: The ratio of the volumet-
ric airflow rate supplied to a space related
to the volume of that space. It is usually
measured in air changes per hour, and nor-
mally relates to the outdoor air change rate.
Air exhaust opening: Air terminal device
used to extract air from a space.
Air flow rate: Mass or volume flow of air
passing a given plane divided by the time.
Air flow: Continuous movement of air.
Air jet throw: The distance an air stream
travels on leaving a diffuser before its ve-
locity is reduced to a specific value.
Air pollution: Any material in the atmos-
phere that affects people and their environ-
ment (pollutants include materials such as
liquids, solids, aerosols, gases and odours).
Air stratification: The layering of air
within a space, due to density differences.
Air supply diffuser: Air terminal device
used to supply ventilation air to a space.
Air temperature: Dry-bulb temperature of
the air.
Air velocity: Rate of motion of air in a
given direction measured as distance per
unit time.
Buoyancy: The vertical force exerted on a
volume of air that has a density different
from the ambient air.
Displacement ventilation [displacement
air distribution]: Room ventilation cre-
ated by room air displacement, by intro-
ducing air at low level in a space at a lower
air temperature than the room air.
Draught risk rating: Percentage of occu-
pants predicted to be dissatisfied due to
draught.
Draught: Unwanted local cooling of the
human body caused by air movement.
Face velocity: Average air discharge ve-
locity from the diffuser (supplied airflow
rate divided by face area).
Indoor air quality: Attributes of the res-
pirable atmosphere (climate) inside a
building including gaseous composition,
humidity, temperature and contaminants.
Isovel: Boundary contours of equal local
mean air velocity.
Local air velocity: Velocity in a specific
point in an air stream at a specific time.
Local mean air velocity: Magnitude of the
time-averaged vector of velocity at a point
of an air stream. The velocity vector and
its components on an orthogonal coordi-
nate system in any point of a turbulent
stream is subject to fluctuations with re-
spect to time. The time averaged vector of
velocity is a vector for which each compo-
nent is averaged with respect to time.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
2. TERMINOLOGY, SYMBOLS AND UNITS
5
Mechanical ventilation: Ventilation with
the aid of powered air movement compo-
nents.
Mean velocity: Instantaneous velocity
averaged for a period of time.
Mixing ventilation [mixing air distribu-
tion]: Air diffusion where the mixing of
supply air and room air is intended.
Occupied zone: Volume of a space be-
tween the floor and 1,8 m above the floor
and more than 1,0 m from outside
walls/windows, 0,5 m from inner walls
and excluding the adjacent zone generated
by displacement diffuser.
Plume: The air flow rising from a hot
body (or descending from a cold body).
Reference air temperature in a room
with displacement ventilation: Average
of at least five measurements of the mean
(in time) air temperature at a height of
1,1 m from the floor within the occupied
zone outside the area directly influenced
by the flow from displacement air supply
diffuser.
Speed: Magnitude of mean velocity.
Temperature: Measurement of warmth
or coldness with respect to an arbitrary
zero or absolute zero. A physical prop-
erty related to the average kinetic energy
of the atoms or molecules of a substance
(according to Collins English Diction-
ary).
Turbulence intensity: Ratio of the stand-
ard deviation of the air velocity fluctua-
tions around the local mean velocity to the
local mean air velocity.
Turbulent flow: Flow that is character-
ized by irregular eddies associated with
momentum transfer between fluid layers.
Ventilation flow rate: Volume flow rate at
which ventilation air is supplied or removed.
Ventilation: Designed supply and re-
moval of air to and from a treated space.
2.2 Symbols
Latin letters
A Area or floor area [m²]
Ar Archimedes number [-]
Af Floor area [m²]
Awl Lower wall area [m²]
Awu Upper wall area [m²]
ao Air diffuser supply area [m²]
af Air diffuser face area [m²]
arf Archimedes coefficient [-]
B Width (of an air diffuser) [m]
D Diameter (of a source) [m]
Fmo Mixing factor of convection flow
[-]
H Height of diffuser or room [m]
Im Entrainment factor of convection
flow [-]
Ka Air diffuser constant, jet discharge
[-]
KDr Air diffuser constant, low velocity
discharge, radial flow [-]
KDp Air diffuser constant, low velocity
discharge, plane flow [-]
Kob Air diffusion constant for flow bet-
ween obstacles
Lp Sound pressure level [dBA]
N Number of convection sources [-]
T Absolute temperature [K]
(= + 273)
Tu Turbulence intensity [ %]
W Depth of an air diffuser [m]
bm Flow adjustment factor for air sup-
ply diffuser [-]
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
Rehva Displacement Ventilation Guidebook
6
bn Half-width of the adjacent zone
[m]
c Contaminant concentration
[mg/m³, ppm, etc.]
ce Contaminant concentration at the
extract air [mg/m³, ppm, etc.]
cexp Contaminant concentration in
breathing air [mg/m³, ppm, etc.]
cmean Mean contaminant concentration
in the room [mg/m³, ppm, etc.]
coz Mean contaminant concentration
in the occupied zone [mg/m³, ppm,
etc.]
cs Contaminant concentration in the
supply air [mg/m³, ppm, etc.]
cp Specific heat at the constant pres-
sure of the air = 1004 J/kg K
(1 J = 1 Ws)
d Diameter [m]
do Diameter of “vena contracta”, i.e.
the most contracted cross section
of a plume [m]
e Entrainment coefficient in the dis-
charge flow from the diffuser [-]
g Acceleration of gravity [m/s²]
h Height [m]
ln Length of the adjacent zone [m]
l0.2 Length of the adjacent zone (to the
0,2 m/s isovel) [m]
n Number (of people)
RH Relative humidity [ %]
pd Dynamic pressure = ½ v² [Pa]
ps Static pressure [Pa]
ptot Total pressure = pd + ps [Pa]
ptot Total pressure drop across a dif-
fuser [Pa]
qB Ventilation rate for emissions from
building, m³/(s·m²)
qs,l Supply air volume flow per m
width of the diffuser/room
[m³/(s·m)]
qp Ventilation rate for occupancy per
person, m³/s, pers
qs Supply air volume flow [m³/s]
qv Air volume flow [m³/s]
qv,z Vertical air volume flow [m³/s]
qv,l Vertical air volume flow in a
plume above a line source
[m³/(s·m)]
hTx Height of mixing layer [m]
hmx Height of lower wall [m]
s Vertical temperature gradient =
z [K/m]
v Velocity [m/s]
vs Face velocity = qo /As [m/s]
vx Horizontal velocity (x-direction)
[m/s]
vx,max Maximum velocity in the vertical
velocity profile at the floor
vy Horizontal velocity (y-direction)
[m/s]
vz Vertical velocity (z-direction)
[m/s]
vs Face velocity = qo /As [m/s]
v Mean velocity [m/s]
x Length co-ordinate [m]
y Width co-ordinate [m]
zexp Height to the breathing zone [m]
z Height co-ordinate [m]
Stratification height [m]
zmax Maximum height for a plume in
stratified surroundings [m]
zo Height between virtual (point)
source and the source [m]
zp Height between virtual (point)
source and the chosen reference
height [m]
zt Equilibrium height for a plume in
stratified surroundings and height
of mixing layer[m]
z*, z** Non-dimensional height for a
plume in stratified surroundings
vf Face of diffuser [m/s]
w Velocity of convection flow [m/s]
Greek letters
Heat flux [W, W/m]
tot Total heat flux = cf + r
[W, W/m]
cf Convective heat flux [W, W/m]
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
2. TERMINOLOGY, SYMBOLS AND UNITS
7
r Radiative heat flux [W, W/m]
Heat transfer coefficient
[W/(m² K)]
cf Convective heat transfer coeffi-
cient [W/(m² K)]
r Radiative heat transfer coefficient
[W/(m² K)]
o Angular spread of the radial flow
from an air diffuser [rad]
Thermal expansion coefficient of
air = 1/ (θ + 273 °C) ~ 1/300 K-1
Thickness of stratified flow near
the floor; δ is the height where ve-
locity vx= 0,5 · vx,max [m]
a Air change efficiency, a measure
of how quickly the air in the room
is replaced [-]
c Mean ventilation effectiveness.
Also called contaminant removal
effectiveness. It is a measure of
how quickly an airborne contami-
nant is removed from the room [-] coz Ventilation effectiveness of the oc-
cupied zone. Also called air quality
index of the occupied zone [-] cP Local ventilation index. Also
called air quality index at a given
point P [-] cexp Personal exposure index. Also
called air quality index of the in-
haled air [-]
Temperature effectiveness [-]
θ Temperature difference [K]
s Under-temperature in the supply
air = oz − s [K] or e − s [K]
θz Difference between maximum air
temperature in a plume and ambi-
ent air temperature [K]
θ Temperature [°C]
a Air temperature [°C]
θob Air temperature in the obstacle
opening [°C]
θe Exhaust air temperature [°C]
θf Floor temperature [°C]
θaf Air temperature near (0,05 m) the
floor [°C]
θmx Air temperature at mixing layer [°C]
θoc Air temperature at 0,65 m height [°C]
θoz Mean air temperature in the occu-
pied zone [°C]
θs Supply air temperature [°C]
θr Upper room air temperature [°C]
θsu Surface temperature [°C]
θwl Surface temperature of lower wall
[°C]
θwu Surface temperature of upper wall
[°C]
Dimensionless temperature of the
air near the floor [-]
Air density. For normal room tem-
peratures = 1,20 kg/m³ (θ = 21 °C)
Ratio of the cooled ceiling output
to the total cooling output
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
8
3 Room air distribution
This chapter discusses methods of total vol-
ume ventilation by mixing ventilation and
displacement ventilation. Other methods
for achieving thermal comfort and good air
quality in spaces which are now under de-
velopment (including localized radiant and
convective system, stratum ventilation,
etc.) and advanced air distribution (such as
personalised ventilation) are not considered
in this chapter.
3.1 Need for Ventilation
In buildings, the supply of clean outdoor air
is needed for breathing and the removal of
internal heat loads, gases and particulates.
Heat is generated by occupants, equipment
(PC’s, lighting, etc.) and solar radiation.
Vapour, gases and particulates are gener-
ated by occupants, building materials, of-
fice equipment, etc. and also introduced by
infiltration from outdoors.
The air supplied to spaces is either filtered
outdoor air or filtered outdoor air mixed
with re-circulated filtered room air. Fur-
thermore, it may be needed to humidify or
dehumidify the supplied air. Displacement
ventilation aims to provide occupants with
clean air for breathing more effectively
than fully mixed air distribution. With dis-
placement ventilation, it is possible to uti-
lize buoyancy flows that transfer contami-
nant from the occupied zone towards the
upper room zone and so improves the qual-
ity of air inhaled by occupants. Simultane-
ously, with better air quality, displacement
ventilation creates a vertical temperature
gradient in the room, with a high tempera-
ture near the ceiling. This may result in
lowering cooling peak power and cooling
energy consumption when only the envi-
ronment of the occupied area is actively
controlled.
It is possible to achieve good indoor condi-
tions in an energy efficient manner by the
use of well-designed displacement ventila-
tion.
3.2 Ventilation and room air distribution principles
a) Target levels
The aim of air conditioning and distribution
into rooms is to maintain, in the most eco-
nomical way, (accounting for energy usage
and cost efficiency) the desired thermal en-
vironment and air quality in the occupied
zone so as to meet target levels during dif-
ferent operating conditions. Depending on
the design criteria, the designer may choose
different room air distribution methods in
order to achieve the specified targets.
b) Methods of room air distribution
The room air distribution method is critical
for the air conditioning of spaces. Often air
distribution in rooms is assisted by radiant
or convective heating and cooling methods.
However, it must be noted that in some
cases a strategy of room air distribution can
also be fulfilled without any mechanical in-
stallations using only buoyancy forces. The
classification of ideal room air distribution
methods is summarized in Figure 3.1
(Hagström et al. 2000). Note that piston
flow requires large amounts of air.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
3. ROOM AIR DISTRIBUTION
9
Fitzner (1996) points out that piston flow
from the floor and upwards exists for Ar-
chimedes numbers less than 360.
3602
vT
hgAr
(3.1)
where:
g = acceleration of gravity = 9,81 m/s²
H = height of the room [m]
= e – s = temperature difference
between exhaust and supply air [K]
T = absolute temperature of the supply air [K]
v = mean air velocity upwards
= air volume flow/floor area [m/s]
For Ar > 360, buoyancy forces will domi-
nate and create a thermally stratified flow.
The Archimedes number can be also ex-
pressed as Ar ~ ∆θs/vs2 in a given geometry,
or as ∆θs/qs2, called (Arratio), because the
supply area a0 is constant within a given ge-
ometry in the following explanations.
Figure 3.1. The ideal performance of the total volume room air distribution principles.
Description
Strategy
θ
Temperature
effectiveness and
ventilation
effectiveness 1∞-
-
oz s
e s c c - c
c - c
oz s
e s
θ θ θθ
θ θ
θ θθε ε
, RF
temp,
c
, RF, c , RF, c
e
s
e
s
e
s
e
s
, RF, c , RF, c
Main
characteristics
Air quality;
DISPLACEMENT MIXING
PISTON STRATIFICATION
ZONING
Unidirectionalflow through theroom
Room dimension Room dimension Room dimension Room dimension
Flow patterncontrolled by lowmomentum supplyair, strong enoughto overcomedisturbances
Utilise densitydifferences
Flow patterncontrolled bybuoyancy
Air flow fromclean zones tocontaminatedzones
Flow patterncontrolled partlyby buoyancy andpartly by supply airmomentum
Uniformconditions inall parts ofthe room
Flow patterncontrolled byhigh momentumsupply air
s = supplye = exhaust
contaminants,
humidity
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
10
Several boundary conditions including room
geometry, type and location of supply and
exhaust air openings, sources and sinks of
heat including their strength and location,
enclosure surface temperatures, etc. will in-
fluence air distribution in spaces. It can be
very complicated to describe all the details
of the boundary conditions because they are
particular to each room, but a few primary
and common conditions and parameters,
which are considered most important, will
be taken into account in the following dis-
cussion. These primary variables are:
• Cooling mode or heating mode;
• Archimedes ratio ∆θs/qs2, or flow rate of
air supplied to the room, qs and tempera-
ture difference between exhaust and sup-
ply air, ∆θs;
• The ratio between the total area of the
supply openings and the wall/ceiling/
floor area, a0/A;
• Location, high or low, of the air supply
opening(s);
• Heat load per floor area.
The ratio between the total area, ao of the
air supply openings and the surface area, A
of wall/ceiling/floor on/in which the supply
openings are located, ao/A, is an important
parameter for the air distribution in the
room. The ratio, ao/A, is considered to be
small for values smaller than 10-3, medium
for values between 10-3 to 0.3, and large for
values larger than 0.3. The values smaller
than 10-3 are typical for diffusers designed
for mixing ventilation, and the value 6·10-3
is typical for displacement ventilation dif-
fusers.
Figure 3.2 shows a design graph (qs − ∆θs
graph) for a constant value of ao/A. The area
on the right side of the curve defines mo-
mentum driven flow while on the left side
defines a flow driven by the buoyancy
forces (Nielsen 2011). The curve indicates
the position of the critical Archimedes
number where the air movement changes
between the two different flow types.
Figure 3.2. Principle determination of airflow
in a room with a given ao /A ratio based on the
critical Archimedes ratio. Convective flow is
dominant on the left side of the graph while in-
let momentum flow is dominant on the right.
In cooling mode, the air distribution pattern
in a room can be addressed in a three-dimen-
sional graph defined by the flow rate of air
supplied to the room, qs, the difference be-
tween exhaust and supply air temperature,
∆θs, and the ratio between the total area of
the supply openings and the wall area, ao/A
as shown in Figure 3.3 (Nielsen 2011).
The whole “family” of air distribution pat-
terns can, in the case of cooling, be described
in two three-dimensional charts, “family
trees”, one for a high location of the supply
opening and one for a low location of the air
supply opening. The charts are shown in
Figure 3.4 and Figure 3.5 (Nielsen 2011).
Figure 3.3. Three-dimensional system which de-
fines the room air distribution for the cooling case.
Δθs
qs
Δθs /qs² large and
convective flow is
dominating
Δθs /qs² small and
inlet momentum is
dominating
qs
Δθo
ao /A
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
3. ROOM AIR DISTRIBUTION
11
A high location of the air supply openings
makes it difficult to work with stratifica-
tion, and to obtain high air change effiency
when cooling, see Figure 3.4. Most of the
flow is characterized by strong mixing, ei-
ther due to the high momentum of the sup-
plied airflow or due to the interaction of the
supplied cold flow moving downwards
with that of the upward thermal plumes
generated by heat sources. Full mixing is
typical for this air distribution pattern. In
the case of downward flow from a full dif-
fuse ceiling (ao /A = 1,0), with a very high
flow rate, it is possible to established piston
flow and supply clean air to the working
zone. This system is often called a laminar
flow system.
Figure 3.4. Different room air distribution sys-
tems for cooling with high location of supply
openings.
Figure 3.5 shows the location of displace-
ment ventilation in the “family tree” of
room air distribution – which is the subject
of this guide book. It can be achieved with
low level supply openings which makes it
possible to work with a high ventilation ef-
fectiveness because of the stratification ef-
fect. Displacement ventilation works with
large supply openings to obtain a low mo-
mentum flow into the room.
Displacement air distribution can, to some
extent, be obtained with high level supply
openings if the openings are large with a
low momentum flow and if the heat
sources in the room are located outside the
downward flow from the openings. The ef-
fect is seen in the case of vertical ventila-
tion, Figure 3.4 (Nielsen et al. 2007).
Figure 3.5. Different room air distribution sys-
tems for cooling with low level supply openings.
3.3 Displacement ventilation and thermal comfort
Thermal comfort is that condition of mind
which expresses satisfaction with the ther-
mal environment. Peoples’ thermal sensa-
tion is related to the thermal balance of their
body as a whole. This balance is influenced
by physical activity and clothing, as well as
several environmental parameters: air tem-
perature, mean radiant temperature, air ve-
locity and air humidity. The ranges of envi-
ronmental parameters for whole body ther-
mal comfort are described in handbooks
and standards (ISO Standard 7730 2005,
EN15251 2007, ASHRAE Standard 55
2013). Nevertheless, a subject in thermal comfort can be negatively affected by ad-verse local environmental conditions.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
12
In practice occupants may experience lo-
cal thermal discomfort at one or more parts
of the body. Local thermal discomfort due
to draught, vertical temperature differ-
ence, radiant temperature asymmetry and
a cold/ warm floor may occur alone or in
combination.
In rooms with displacement air distribution
non-uniformity in the vertical temperature
field may cause local discomfort due to
“warm head” and “cool feet” when the dif-
ference in air temperature between the head
level (1,1 m above the floor) and the ankle
level (0,1 m above the floor) is large. The
standards recommend vertical temperature
difference between 1,1 m and 0,1 m above
the floor to be in the range of 2 K to 4 K
depending on the category of the aimed in-
door thermal environment (ISO 7730 2005,
EN 15251 2007, ASHRAE 55 2013).
In rooms with displacement air distribution
the relatively low temperature and high ve-
locity near the floor may case draught at the
feet. Draught is defined as unwanted local
cooling of the body due to air movement.
The risk of draught increases when airflow
temperature decreases and mean velocity
and turbulence intensity increase. The per-
centage of occupants dissatisfied due to
draught, DR (%), can be predicted by the
following equation (ISO Standard 7730
2005, EN 15251 2007):
)143(0,37)050)(34( 0,2 ,Tuv,vDR a
(3.2)
In this equation θa [C] is the air tempera-
ture, v [m/s] is the mean velocity, and Tu
[ %] is the turbulence intensity of the flow.
The equation is valid when v is higher than
0,05 m/s; for v smaller than 0,05 m/s, v =
0,05 m/s should be used; for DR > 100 %,
DR = 100 % should be used.
Equation (3.2) can be used when air tem-
perature, mean velocity and turbulence in-
tensity at the location of the occupants are
known (obtained by airflow predictions or
measurements).
Measurements of these parameters at four
heights 0,1, 0,6, 1,1 and 1,7 m above the floor
are recommended in the standards (ISO 7726
1998, ASHRAE 55 2013). However, in
rooms with displacement air distribution the
highest velocity typically occurs below
0.1 m. Therefore, measurement at a height of
0.05 m above the floor is also recommended
in order to identify the highest velocity. The
turbulence intensity is typically approxi-
mately 40 % in case of mixing air distribution
and approximately 20 % in case of displace-
ment air distribution. Field surveys reveal that
combined discomfort due to draught and ver-
tical temperature difference is not typically a
serious problem in rooms with displacement
ventilation (Melikov et al. 2005).
3.4 Displacement ventilation and air quality
The primary aim of ventilation of occupied
spaces is to provide people with clean air for
breathing. In this respect displacement air
distribution may perform better than mixing
air distribution because the free convection
layer around the human body is less dis-
turbed and its ability to transport clean air
from the lower room level to the breathing
zone can be better utilised. This has been
documented in numerous studies based on
CFD predictions and physical measure-
ments performed under laboratory condi-
tions. Only contaminant sources with heat
production can be treated effectively by dis-
placement ventilation (Wildeboer and Mül-
ler 2006, Cermak and Melikov 2006). How-
ever, as will be discussed in the following
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
3. ROOM AIR DISTRIBUTION
13
chapters, the ability of displacement air dis-
tribution to provide room occupants with
clean air for breathing depends on several
other factors, such as type and location of
heat sources, movement of occupants, etc.
Cleanliness of the inhaled air is important
for occupants’ health and perceived air qual-
ity. In addition to air cleanliness, tempera-
ture and relative humidity of the inhaled air
are also important. Sick building syndrome
symptoms decrease and perceived air qual-
ity improves when cleanliness of the inhaled
air increase and its temperature decrease
(Fang et. al 2004, Melikov and Kaczmar-
czyk 2012). Elevated facial air movement
improves perceived air quality and reduces
the negative impact of elevated pollution,
temperature and relative humidity of the in-
haled air (Melikov and Kaczmarczyk 2012).
These findings are important for the perfor-
mance of displacement air distribution.
Limited research on human response to the
environment generated by displacement
ventilation is reported in the literature (Car-
rer et al. 2012). At a comfortable room air
temperature of 23 °C (1,1 m above floor
level) and typical indoor pollution sources,
Sick Building Syndrome (SBS) symptoms
(eye irritation intensity and eye dryness),
perceived air quality and thermal comfort
were reported by people to be at the same
level in rooms ventilated by displacement
air distribution and those employing mix-
ing air distribution (Dalewski et al. 2014).
Compared to mixing air distribution the
positive impact of inhaling clean air on per-
ceived air quality in the case of displace-
ment ventilation may be diminished by the
high temperature of the inhaled air mainly
originated from the free convection layer
around the body. Due to the low velocity air
may be perceived less fresh in rooms with
displacement ventilation than in rooms with
mixing ventilation (Dalewski et al. 2014).
This has been reported also in field surveys
in rooms with displacement ventilation
(Melikov et al. 2005).
The temperature and flow rate of the supply
air are important parameters for the design
of displacement ventilation. Maintaining
relatively high room air temperatures may
lead to energy saving. It is reported that at
a target air temperature in the occupied
zone (1,1 m height) of 26 °C or 29 °C the
increase of the flow rate and temperature of
the air supplied by displacement ventilation
(i.e. small difference between temperature
of room and supply air) leads to an im-
provement of the perceived air quality
while a decrease of the flow rate and tem-
perature of the supplied air (large differ-
ence between temperature of room and sup-
ply air) leads to improvement of occupants’
thermal comfort (Dalewski et al. 2014).
This is discussed further in Chapter 9.
Perceived air quality (PAQ) is used in the
standards to define the minimum amount of
outdoor air needed for ventilation. Indoor
air pollution is generated by occupants (bio
effluents) and emissions from building ma-
terials. The total amount of outdoor air re-
quired for ventilation is defined as:
Bps qAqnq (3.3)
where
qs = total ventilation rate for the breathing
zone, m³/s;
n = design value for the number of the
people in the room;
qp = ventilation rate for occupancy per
person, m³/s per person;
A = room floor area, m²;
qB = ventilation rate for emissions from
building, m³/(s m²).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
14
Because displacement air distribution is
considered to be more efficient than mix-
ing air distribution in providing clean air
to the breathing zone of occupants, the to-
tal amount of outdoor air may be reduced.
If the displacement ventilation is not suffi-
cient to remove heat generated in the
room, it may be combined for example,
with radiant cooling (see Chapter 7).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
15
4 Performance of displacement ventilation
4.1 Displacement Ventilation Method
The air flow pattern in a ventilated room is
principally divided into two types, mixing
(dilution) ventilation and displacement
ventilation. In mixing ventilation, the ven-
tilation air is supplied in such a way that the
room air is mixed and the contaminant con-
centration is the same in the whole room. In
displacement ventilation, which is the sub-
ject of this book, a stratified flow is created
using the buoyancy forces in the room. The
air quality in the occupied zone is then gen-
erally better than with mixing ventilation.
The ventilation system supplying the air to
the room is not considered in this book, but
only the air flow within the room.
Displacement ventilation has for many years
been used in industrial premises with high
thermal loads. Since the mid-80’s it has also
been used more extensively in non-industrial
premises, especially in the Scandinavian
countries. Displacement ventilation presents
the opportunity to improve both the temper-
ature effectiveness (Chapter 4.3.3) and the
ventilation effectiveness (Chapter 4.6). The
principle is based on air density differences
where the room air separates into two layers,
an upper polluted zone and a lower clean
zone (Figure 4.1). As already discussed in
Chapter 1 this is achieved by supplying cool
air with a low velocity in the lower zone and
extracting the air in the upper zone. Free
convection from heat sources creates verti-
cal air movement in the room. When the heat
sources in the room are also the contamina-
tion sources, the convection flows transport
the warm polluted air up to the upper zone.
The convection flow rates relative to the
ventilation flow rate determine the height of
the boundary between the two zones. The
sum of the warm convection flow rates to the
upper zone minus the downward directed
flow rates from cold surfaces to the lower
zone is equal to the ventilation air flow rate
supplied to the room. An increased ventila-
tion flow rate at fixed convection flow rates
thus moves the boundary upwards and a de-
creased flow rate moves the boundary down-
wards.
Figure 4.1.Schematic illustration of the air flow
that might be found in a room ventilated by dis-
placement ventilation.
4.2 Air flow pattern
In a displacement ventilated room, the air
flow pattern is governed by the convection
flows from heat sources and sinks present in
the room. This means that a distinctive fea-
ture of displacement ventilation is the for-
mation of horizontal air layers. The warmest
air layers are at the top and the coolest air
layers at the bottom. The air moves easily
within a horizontal layer but the transporta-
tion between the layers needs a stronger
force (Figure 4.2). This means that the ex-
tract should be positioned at the layer in
which the pollution concentrations are high-
est or where the highest temperatures occur.
In most cases this means that the extract
should be in the upper part of the room.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
16
The vertical air movement is caused by
convection flows from warm sources or
cold sinks. Warm objects such as people,
computers, lamps etc. create rising convec-
tion flows. Depending on the power and ge-
ometry of the heat source the convection
flows will rise all the way to the ceiling or
settle at a lower height (Figure 4.3).
Figure 4.2. Horizontal air movement in con-
nection with the extract.
Figure 4.3. Vertical air movement caused by
convection.
The supply air temperature must be lower
than the room air temperature, which is nor-
mally the case when there is a cooling load in
the room. If the supply air temperature is
warmer there will be a short-circuit (Fig-
ure 4.4). However, the vertical air flow has a
certain amount of entrainment, which causes
some circulation in the rest of the room, this
is sometimes used for heating an empty room
prior to the time of occupation by means of a
displacement ventilation system.
The airflow pattern in rooms with displace-
ment ventilation is sensitive to other flows.
Walking occupants will cause mixing of the
clean and cool air with the polluted and
warm air at the higher level. This will dis-
turb the displacement principle. The tem-
perature of the inhaled air will decrease and
pollution concentration will increase. A
person walking close to the air supply dif-
fuser will cause more disturbance than a
walking person more distant from the dif-
fuser. However, the displacement airflow
pattern will recover in a relatively short
time (Halvonova and Melikov 2010).
Figure 4.4. Short-circuit of air flow in a room
when the supply air temperature is warmer than
the room air temperature.
4.3 Temperature distribution
As already discussed in Chapter 1, the risk
of draught at the feet and discomfort due to
vertical temperature difference exist in
rooms with displacement ventilation be-
cause the cold supply air (sis released at
low level directly to the occupied zone
(Figure 4.5) and warm exhausted air (e) is
removed at the ceiling level. The room air
temperature (at different heights will
not, however, vary by much in the horizon-
tal direction, except close to the diffuser.
Figure 4.5. Temperature stratification in a dis-
placement ventilated room.
Hei
ght
ab
ov
e fl
oo
r [m
]
0 0,2 0,4 0,6 0,8 1 1,20,0
0,5
1,0
1,5
2,0
2,5
Temperature ratio -
- s
se
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
17
4.3.1 Temperature at the floor
The temperature of the supply air in the
floor area rises due to induction and con-
vection, as radiation from the other warmer
surfaces in the room in turn heat the floor.
A dimensionless temperature of the air near
the floor is often presented as
se
saf
(4.1)
where:
f = the air temperature near the floor
s= the supply air temperature
e = the exhaust air temperature
The total temperature difference together
with the air volume flow rate gives the
amount of heat removed from the space:
sepstot cq Φ (4.2)
where:
tot = the heat removed from the space [W]
qs= supply air volume flow [m³/s]
= the air density = 1,2 kg/m³
cp= the specific heat of the air =
1004 J/kg K
The following equation can be used to esti-
mate the dimensionless temperature of the
air near the floor (Mundt 1990):
111
1
cfr
ps
A
cq
(4.3)
where
A = the floor area [m²]
r= the heat transfer coefficient due to
radiation [≈ 5 W/m²K]
cf = the heat transfer coefficient at the
floor due to convection [≈ 4 W/m²K]
In Figure 4.6 the dimensionless tempera-
ture of the air near the floor is shown as a
function of the ventilation flow rate per m²
floor area. The points shown in the figure
are from measurements with distributed
heat sources presented in eleven different
references (Mundt, 1996).
Figure 4.6. Dimensionless temperature of the
air near the floor as a function of the ventilation
flow rate per m² floor area with different heat
transfer coefficients due to convection.
4.3.2 Vertical temperature distribution
The vertical temperature distribution de-
pends on the vertical location of the heat
sources. When the heat sources are in the
lower part of the room the temperature gra-
dient is larger in the lower part and the tem-
perature more constant in the upper part. On
the other hand, when the heat sources are lo-
cated mostly in the upper zone, the tempera-
ture gradient is smaller in the lower part and
increases in the upper part (Figure 4.7).
The type and location of the source has a
significant effect on the relative tempera-
ture difference (Figure 4.7). Point sources
and horizontal sources (warm floor) create
a clear mixing layer.
Ventilation flow rate per m² floor area,
qs / A [x10−3 m3/s m²]
0
0,2
0,4
0,6
0,8
1,0
0 1 2 3 4 5 6 7 8
κ =
(θa
f −θ s
) /
(θe −
θ s)
αcf
= 5 W/m²K
αcf
= 3 W/m²K
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
18
Over the mixing layer, the room air temper-
ature could be assumed to be constant or to
only slightly increase. Vertical heated
sources (for example, a warm window) pro-
duce a nearly linear profile with no clear
mixing layer (Figure 1.5). In rooms where
the heat sources are located at a high level,
displacement ventilation is efficient for
keeping the occupied spaces cool (Fig-
ure 4.8). The air temperatures near the
floor, f, and the vertical temperature gra-
dient are not only a function of flow rate
and load, they are also a function of the type
of heat source in the room.
According to Nielsen (1996) and Brohus
and Ryberg (1999) the relative air tempera-
ture near the floor, (see equation 4.1) var-
ies between 0,3 and 0,65 for different types
of heat sources (Figure 4.9).
A concentrated heat load such as a small
furnace in an industrial environment can
give a value of 0,3. A ceiling light will
give a vertical temperature gradient with a
floor temperature of = 0,5, which is gen-
erated by radiation from the light source.
When people are the primary heat source
will have a value of 0,58, and evenly dis-
tributed heat sources will give a value of
0,65. It is obvious that this can vary in the
same magnitude as that associated with dif-
ferent flow rates.
Heat sources in the lower part of the
room
Heat sources in the upper part of the
room
Hei
ght
above
floor
[m]
0 0,2 0,4 0,6 0,8 1 1,20,0
0,5
1,0
1,5
2,0
2,5
Temperature ratio -
- s
se
Figure 4.7. Relative change in the vertical tem-
perature in a displacement ventilated room with
the heat sources at different levels.
Figure 4.8. Roof heated by sun - an example where displacement ventilation is efficient.
Temperature
Hei
gh
t ab
ov
e fl
oo
r
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
19
The different temperature gradients are
shown in Figure 4.9 where it is assumed
that the vertical temperature distribution is
a linear function of the height. If many dif-
ferent heat sources are present in the room
it is advised to use the “50 % rule” (Chapter
4.3.4). In real situations, vertical tempera-
ture stratification is often non-linear.
Distributed heat sources
Sedentary persons
Ceiling light
Point heat source
Heig
ht
ab
ov
e f
loo
r
0 0,3 0,5 0,65 1
0,58
Temperature ratio -
- s
se Figure 4.9. Vertical temperature distribution
for different types of heat loads with assumption
of linear vertical temperature distribution.
4.3.3 Temperature effectiveness
As the exhaust temperature is higher than
the air temperature in the occupied zone, a
temperature effectiveness can be defined:
soz
se
(4.4)
where
oz = the mean air temperature in the occu-
pied zone
4.3.4 Simplified assumptions for the
temperature distribution
As shown in Figure 4.5 and Figure 4.7, the
temperature increases with height, and the
temperature profile depends on the location
of the heat sources and the flow rate. For
most practical purposes, temperature pro-
files are assumed as shown in Figure 4.10.
Figure 4.10. The "50 %-rule" for vertical tem-
perature distribution.
The “50 %-rule” for the vertical tempera-
ture distribution indicates that the air tem-
perature at the floor is half-way between the
supply air temperature and the extract air
temperature. This is a general experience
that may be used as a first approximation
for most normal rooms and normal air dif-
fusers.
Example:
If the heat balance and air flow rate in the
room yields a temperature increase of
e − s10 K, then the temperature at
the floor level will become approxi-
mately 5 K higher than the supply air
temperature.
In rooms with higher ceilings than normal,
it is often found that the temperature in-
crease from supply air temperature to that
of the air at the floor is less than 50 % of the
total temperature increase. In these cases, a
“33 % rule” may be appropriate.
Temperature
Air temperature
at floor, θaf
Supply air
temperature, θs
Extract airtemperature, θe
Hei
ght
above
flo
or 50% 50%
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
20
4.4 Convection flows – the engines of displacement ventilation
Natural convection flows are the engines of
displacement ventilation. A natural convec-
tion flow is the air current that rises above
warm objects like people or computers,
rises along a warm wall, or descends from
cold objects like windows or outer walls,
due to buoyancy (Figures 4.11 - 4.13). To
understand displacement ventilation, one
has to understand the nature of the natural
convection flows, and to know the magni-
tude of these flows.
The convection flow rising above a hot ob-
ject, including the human body, is called a
thermal plume, or simply a plume. Empiri-
cal, analytical and computational fluid dy-
namics are commonly used methods to
evaluate air temperatures, velocities and air
flow rates in thermal plumes above differ-
ent heat sources and convection flows at
vertical surfaces.
All plumes encountered in practical venti-
lation are turbulent flows, and follow the
similarity laws for fully turbulent flows.
The amount of air in the convection flow
increases with height due to entrainment of
the surrounding air. The amount of air
transported in a natural convection flow de-
pends on the temperature and the geometry
of the source and the temperature of the sur-
rounding air. As the driving force in con-
vection flows is the buoyancy force caused
by the density difference (i.e. the tempera-
ture difference) a temperature gradient in
the room influences the plume rise height.
With development of low power consum-
ing office equipment, lighting, high quality
windows, etc., which generate weak buoy-
ancy flows, the importance of the natural
Figure 4.11. Convection flows - the engine of
displacement ventilation.
Figure 4.12. Convection flows at vertical sur-
faces.
Figure 4.13. Thermal plume above a horizontal
source.
Hot wall
>
Cold wall
<
su
su su
su
Flowqv
Flowqv
z
Flowqv
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
21
convection flow around the human body,
especially in rooms with displacement air
distribution, will increase. Apart from the
impact on the room air distribution (due to
the generated thermal plume) the free con-
vection flow transports pollution generated
by the human body and in its surroundings
to the breathing zone and therefore is im-
portant for occupant exposure and inhaled
air quality.
In the comfortable temperature range, the
maximum velocity in the convective
boundary layer (CBL) around the human
body may be as high as 0,25 – 0,30 m/s. It
decreases when the difference between the
body surface temperature and the surround-
ing air temperature decreases (Licina et al.
2014, 2015, 2015a, 2015b, 2016). The ve-
locity and temperature distribution in the
CBL, as well as the thickness of the bound-
ary layer is influenced by numerous factors
including body posture, clothing style and
thermal resistance, presence of obstacles in
the vicinity of the body (such as a desk that
greatly reduces the strength of the natural
convection flow). (Licina et al. 2014).
Breathing also influences the natural con-
vection flow (Özcan et al. 2003, 2005). The
natural convection flow and its importance
for human thermal comfort and inhaled air
quality is discussed in Chapter 9.
4.4.1 Point and line sources
Thermal plumes above point and line
sources (Figure 4.14) have been studied for
many years. Among the earliest publica-
tions are those from Zeldovich (1937) and
Schmidt (1941). Turner (1973) gives a
comprehensive record of most of the phe-
nomena encountered in connection with
buoyancy effects in fluids. Analytical equa-
tions to calculate velocities, temperatures
and air flow rates in thermal plumes over
point and line heat sources with given heat
loads were derived based on the momentum
and energy conservation equations and as-
suming Gaussian velocity and excessive
temperature distribution in thermal plume
cross-sections (Mundt, 1996).
Figure 4.14. Plumes from a point source and
from a line source.
These equations correspond with those pro-
duced experimentally by other researchers
(Mierzwinski, 1981, Popiolek, 1981) and
are listed in Table 4.1. The equations in
Table 4.1 were derived assuming that the
size of the heat source was very small and
did not account for the actual source dimen-
sions.
The coefficients in the equations differ
slightly in different references depending
on the entrainment coefficients used. cf is
the convective heat flux in W or W/m from
the heat source and z is the height above the
level of the heat source. The convective
heat flux cf can be estimated from the en-
ergy consumption of the heat source tot by
cf k tot (4.5)
The value of the coefficient k is 0,7–0,9 for
pipes and ducts, 0,4–0,6 for smaller compo-
nents and 0,3–0,5 for larger machines and
components (Nielsen, 1993).
Point source Line source
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
22
4.4.2 Convection flow along vertical
and horizontal surfaces
Convection flow along vertical surfaces is
also of significance. When the vertical ex-
tension of the surface is small, the convec-
tion flow is mainly laminar and at larger di-
mensions the flow is turbulent. The basic
equations for a surface with a constant tem-
perature are given in Table 4.2 (Jaluria
1980, Etheridge and Sandberg1996).
is the temperature difference between
the surface and the surrounding air and z is
the height from the bottom of the surface.
The flow changes from laminar to turbulent
at Gr·Pr=7·108, which for air and moderate
temperature differences means around z =
1 m and for air at higher temperatures
around z = 0,5 m.
Convection flows from horizontal surfaces
are very difficult to determine in the same
basic way as for point, line or vertical
sources. The reason is that the flows behave
in a very unstable way and leave the flat
surface from different positions at different
times, partly depending on the total air
movement in the room. These surfaces are
mostly treated as plumes from extended
sources see Chapter 4.4.3.
4.4.3 Extended sources
In reality heat sources are seldom a point, a
line or a plane vertical surface. The most
common approach to account for the real
source dimensions is to use a virtual source
from which the air flow rates are calculated
(Mundt 1992 and Skistad 1994) (Fig-
ure 4.15). The virtual origin is located
along the plume axis at a distance z0 on the
other side of the real source surface.
Figure 4.15. Illustration of the position of the
virtual source.
Table 4.1. Characteristics of thermal plumes above point and line sources.
Parameter Point source Line source
Centreline velocity, vz [m/s] vz = 0,128 cf 1/3 z – 1/3 vz = 0,067 cf 1/3
Centreline excessive temperature, z [K] z = 0,329 cf 2/3 z – 5/3 z = 0,094 cf2/3 z –1
Air flow rate, qv,z [m³/s for point source, m³/sm for line source] qv,z = 0,005 cf 1/3 z 5/3 qv,z = 0,013 cf 1/3 z
Table 4.2. Characteristics of convection flows along vertical surfaces.
Parameter Laminar region Turbulent region
Maximum velocity, vz [m/s] z, vz 10 z, vz 10
Thickness of boundary layer [m] 250250050 ,, z, 7010110 ,, z,
Air flow rate, qv,z [m³/sm width] 750250002870 ,,
z,v z,q 2140002750 ,,
z,v z,q
b) Extended source
Virtual source
z
a) Point source
z0
Flowqv
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
23
The adjustment of the point source model
to realistic sources using the virtual source
method gives a reasonable estimate of the
air flow rate in thermal plumes.
The weak part of this method is how to es-
timate the location of the virtual located
point source. The method of a "maximum
case" and a "minimum case" provides a tool
for such estimation (Figure 4.16) (Skistad
1994). According to the "maximum case",
the real source is replaced by the point
source such that the border of the plume
above the point source passes through the
top edge of the real source (for example, a
cylinder). The "minimum case" is when the
diameter of vena contracta of the plume is
about 80 % of the upper surface diameter
and is located approximately 1/3 diameter
above the source. The spreading angle of
the plume is set to 25º. For the low-temper-
ature sources, Skistad (1994) recommends
the "maximum case", whereas the "mini-
mum case" best fits the measurements for
larger, high temperature sources. The
“maximum case” gives z0 = 2,3·D and the
“minimum case” z0 = 1,8·D with z0 defined
in Figure 4.16.
Figure 4.16. Convection flow above a vertical
cylinder.
For a flat heat source Morton et al. (1956)
suggest the position of the virtual source to
be located at z0 = 1,7 – 2,1·D below the real
source. Mundt (1996) calculates the thick-
ness of the boundary layer (see Table 4.2)
at the top of a vertical extended heat source
and adds this to the source radii and then
calculates the position of the virtual source
as z0 = 2,1·(D+2·) before using the point
source equation. According to Bach et al.
(1993), the volume flow from the vertical
surfaces should be added to the volume
flow calculated by the equations for point
or line sources.
Example:
Calculate the convection flow rate 0,5 m
above a cylinder with height 1 m and diameter
0,4 m. The convective heat flux is 50 W.
For the maximum case (Figure 4.16)
m 902552)512tan2( ,D,,/Dzo
and
m 4150900 ,,,hzz
and from Table 4.1
35310050 z,q /cfz,v
which gives
m³/s032041500050 3531 ,,,q z,v
In the minimum case (Figure 4.16)
m 7208041)512tan2(80 ,D,,/D,zo
and
m 0915013072030 ,,,,hDzz
which gives
m³/s0210091500050 3531 ,,,q z,v
(The position of the virtual source is in this
case D,D, 471318041 below the up-
per edge of the source.)
Minimum case
Maximum case
D
z
H
h
D
z
H
h D/3d0 d0
z0
z0
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
24
4.4.4 Plume interaction
When a heat source is located close to a wall,
the plume may be attached to the wall, Fig-
ure 4.17a. In this case the entrainment will
be reduced compared to the entrainment in a
free plume. The air flow rate from a heat
source can then be calculated as half of the
flow from a source with a heat emission of
2cf (Nielsen, 1993). See equation (4.6).
351/3
3531
Φ003202
)Φ2(0050z,
z,q cf
cf
v,z
(4.6)
Figure 4.17. Thermal plumes.
If the heat source is located in a corner the
air flow rate is equal to 40 % of the air flow
from a heat source with a heat emission of
4 cf (Nielsen, 1993).
35310020 z,q z,v Φ (4.7)
When several heat sources are positioned
close to each other the plumes merge into a
single plume (Figure 4.17b). The total
flow from N identical sources is then given
by (Nielsen, 1993)
z,vN,z,v qNq 31 (4.8)
where
qv, z = the flow in the plume from one of
the sources
When the heat sources are more separated
the total flow is equal to the sum of the
flows from each heat source.
4.4.5 Plumes and temperature gradi-
ents
When there is temperature stratification in a
room, like in a room ventilated by displace-
ment ventilation, the plumes are influenced
by the temperature stratification. The driv-
ing force for the plume is the temperature
difference between the plume and the sur-
roundings and when this difference dimin-
ishes the plumes will disintegrate and spread
horizontally in the room (Figure 4.18). The
individual plumes rise to particular levels as
plume 1 and 2 in Figure 4.18. The total ef-
fect on the temperature gradient is the com-
bination of the various heat loads.
Figure 4.18. Schematic illustration of the air flow
pattern in a room ventilated by displacement.
a) Plume attached
to a wall
b) Interaction between
two plumes
Plume 1
Plume 2
Plume 3
Plume 1
Plume 2
Room
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
25
Batchelor (1954) noticed the influence of a
temperature gradient surrounding the
plume and Morton et al (1956) gave a solu-
tion for calculating the maximum plume
rise from a point source in surroundings
with a temperature gradient. The volume
flow rates in plumes in a room with temper-
ature stratification is slightly decreased
compared to the volume flow rates calcu-
lated with the equations presented for a
non-stratified media, Mundt (1992). Jin
(1993) studied the maximum plume rise
height for plumes above welding arcs.
In the presence of a temperature gradient,
the convective plume reaches the equilib-
rium height (zt) where the temperature dif-
ference between the plume and the ambient
air disappears, see Figure 4.19. Also, there
is another level in the plume, where the air
velocity equals zero. This is referred to as
the maximum height of the plume (zmax ).
Figure 4.19. Vertical plume in a room with tem-
perature gradients and stratification.
The plume spreads horizontally between
these two heights. The convective flow be-
low zt can be calculated from the following
model (Mundt, 1996).
Point source
The position of the virtual source is calcu-
lated. A dimensionless height z* above the
virtual source is calculated
4183 Φ862 /cf
/* sz,z
(4.9)
where:
s = vertical temperature gradient
(/z) in the room [K/m]
cf = convective heat from the
source [W]
As can be seen from Figure 4.19, only z*
values less than 2,1 are relevant to further
calculations. The volume flow rate at the
height z* is then given by
18543
Φ002380 Zs,q cfv
with
321 0620380003900040 *** z,z,z,,Z
(4.10)
where:
qv = the volume flow rate in m³/s.
The maximum height zmax is given by Equa-
tion (4.9) for z* = 2,8
8341Φ980 s,z cfmax (4.11)
and the height zt by Equation (4.9) for
z* = 2,1
8341Φ740 s,z cft (4.12)
z*
2,1
2,8
z**
2,0
2,95
Point source
Line source
s = > 0d
dz
zt
zmax
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
26
Line source
The position of the virtual source is calcu-
lated. A dimensionless height z** above the
virtual source is calculated
3121 Φ785 /cf
/** sz,z
(4.13)
where:
s = vertical temperature gradient
(/z in the room [K/m]
cf = convective heat from the
source [W/m]
As can be seen from Figure 4.19, only z**
values less than 2,0 are relevant to further
calculations. The volume flow rate at the
height z** is then given by
22132
Φ004820 Zs,q cfl,v
with
322 0180029047700040 ****** z,z,z,,Z
(4.14)
where
qv, 1 = the volume flow rate in m³/(s m)
The maximum height zmax is given by Equa-
tion (4.13) for z**=2,95.
2131Φ510 //
cfmax s,z (4.15)
and the height zt by Equation (4.13) for
z**=2,0.
2131Φ350 s,z cft (4.16)
4.4.6 Plumes from real objects
From the theories above and practical ex-
periments, Nielsen (1993) has summarised
the convection flows above some common
objects found in non-industrial environ-
ments (Figure 4.20). The line drawn in the
upper figure is calculated by the equation
for the air flow rate in Table 4.1. The con-
vection flow above a sitting person is thus
approximately 0,02 m³/s (Figure 4.21). In
order to keep the inhaled air at a lower pol-
lution concentration than the surrounding
air at the same level, a lower air flow may
however be used in calculations, see Chap-
ter 4.6.
Figure 4.20. Convection volume flow at nor-
mal room temperatures above a sedentary per-
son, upper figure and above some objects.
(Mundt, 1992/Nielsen, 1993).
Fluorescent
lamp 36 W
Desk
lamp
60 W
Vertical temp
gradient
s = 0,3 K/m
Equation, Table 4.1
Convec
tion f
low
rat
e,q
vz [
x10
-3m
³/s]
Conv
ecti
on f
low
rat
e,q
vz [
x10
-3m
³/s]
Height above object, z [m]
Height above object, z [m]
Personal
computer
75W
s = 0,09 K/m
1,0 2,0 3,0 4,0 5,0
0,3 0,5 1,0 1,2 1,4
10
20
30
50
80
80
50
30
10
5
3
200
100
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
27
The convective flow above the human
body, referred as thermal plume, is influ-
enced by body posture, surrounding air
temperature and its stratification, design of
clothing (tight or loose) and its thermal in-
sulation, furniture design for example, chair
design and desk positioning relative to the
seated human body, etc. (Homma and Ya-
kiyama 1988, Zukowska 2011a, Zukowska
et al. 2010a and b, 2011b, 2012a and b). A
standing person generates a symmetrical
thermal plume which is relatively easy to
characterize. However, for a sitting person
the boundary layer develops asymmetri-
cally due to the impact of the thermal flow
rising from the thighs and lower legs (Fig-
ure 4.22). The characteristics of an asym-
metrical thermal plume above a sitting hu-
man body (room occupant) can be accu-
rately calculated (Zukowska et al. 2010b).
The normal height of the ceiling in rooms
is often insufficient to allow full develop-
ment of the plume and the formation of
symmetrical profiles of air temperature and
velocity distribution.
Figure 4.21. Convection flow in plume above a
sedentary person in a normal environment.
Figure 4.22. Maps of temperature excess (K)
above room temperature (above) and air veloc-
ity (m/s) (below) measured 0,7 m above the
head of a sitting thermal manikin resembling
room occupant (Zukowska et al. 2010b).
Hei
ght
abo
ve f
loor
[m]
0
0,5
1,0
1,5
2,0
2,5
s = dθ /dz
= 1,5 K/m
q = 0,020 m³/svz
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
28
4.5 Contamination distribution
The contamination distribution in a dis-
placement-ventilated room depends on the
position of the contamination sources and if
the heat sources are also the contamination
sources. In the ideal case with warm con-
centrated contamination sources, all con-
taminants are transported directly into the
upper zone by the convection flows, see
Figure 4.23. According to Krühne and Fitz-
ner (1995), Cermak et al. (2006), Cermak
and Melikov (2006) if the contamination
sources are cold and evenly distributed at
the floor, the contamination distribution
will be like the temperature distribution
(Figure 4.10).
However, if the source is too weak, the
plume might disintegrate at a lower level
and the contaminants will then be trapped
at this level (Figure 4.24) and only slowly
transported indirectly by the stronger con-
vection flows to the upper zone.
A typical situation is the stratification of
human exhalation in a room with a vertical
temperature gradient of 0,5 K/m as shown
in Figure 4.25B (Bjørn and Nielsen 2002).
The contaminant concentration is also in-
fluenced by the downward directed convec-
tion flows that might occur at the outer
walls in cold seasons, especially when the
walls are poorly insulated.
Figure 4.23. Schematic illustration of the contamination distribution in a room ventilated by dis-
placement ventilation and with warm contaminant sources.
Figure 4.24. Schematic illustration of the contamination distribution in a room ventilated by dis-
placement ventilation, when the contaminant source (the person) is not the warmest source.
0 0,2 0,4 0,6 0,8 1,0
Hei
gh
t ab
ove
floo
r, z
[m
]
0
0,5
1,0
1,5
2,0
2,5
Contamination ratio, c /croom e
Hei
ght
above
floor,
z [
m]
Temperature,
0
0,5
1,0
1,5
2,0
2,5
Contamination, c room
room
plume 1
plume 2
croom
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
29
A
B
Figure 4.25. A) Exhalation in surroundings
with a small vertical temperature gradient,
0,1 K/m. The exhalation rises to the ceiling. B)
Stratified exhalation from a manikin (person) in
a room with a larger vertical temperature gra-
dient, 0,5 K/m.
These downward flows will then transport
the contaminants from the upper zone back
to the lower zone (Yamanaka et al. 2007).
However as long as there is a positive con-
centration gradient in the room, the contam-
inant concentration in the occupied zone
will always be lower than by mixing venti-
lation.
The influence of a poorly insulated roof
will, in the cold season, decrease the con-
centration gradient, due to the downfall of
cold air, just like with the cold walls (Fig-
ure 4.26). However, if the roof is heated by
the sun this will help stabilise the displace-
ment ventilation as it heats the air in the up-
per zone (Figure 4.8).
Figure 4.26. Poor building air tightness and in-
sulation may reduce the benefit of displacement
ventilation, and make it more like mixing venti-
lation.
4.6 Ventilation effectiveness
In order to assess and compare different air
distribution patterns different definitions of
ventilation effectiveness have been intro-
duced. These are discussed in detail in the
REHVA’s Guidebook No. 2 on Ventilation
Effectiveness (Mundt et al. 2004). In defin-
ing ventilation effectiveness, a distinction
must be made between two terms:
• the contaminant removal effectiveness,
c, which is a measure of how quickly an
airborne contaminant is removed from
the room (Brouns and Waters, 1991) and
• the air change efficiency, a, which is a
measure of how quickly the air in the
room is replaced (Sutcliffe, 1990).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
30
In a displacement ventilated room, the air
change efficiency is mostly higher (a 60–
70 %) than in a room ventilated by mixing
ventilation (a 50 %), (Mundt, 1994). A
good survey of the relationship between the
different versions of ventilation effective-
ness is given by Nielsen (1993). Flows in
rooms can be defined based on the air
change efficiency (Figure 4.27). Perfect
mixing ventilation is defined by an air
change efficiency a equal to 50 % and a
contaminant removal effectiveness c equal
to 1. Short circuit flows lead to values
smaller than 50 % for the air change effi-
ciency. The quality of displacement venti-
lation systems depends on the contaminant
source. Only contaminant sources with heat
production can be treated effectively by
displacement ventilation (Wildeboer and
Müller 2006).
Figure 4.27. Definition of different flow types
based on ventilation effectiveness measures.
4.6.1 Contaminant removal effective-
ness
The contaminant removal effectiveness is
defined by:
sm ean
sec
cc
cc
(4.17)
where
ce = the contaminant concentration in
the exhaust
cs = the contaminant concentration in
the supply
cmean = the mean contaminant concentra-
tion in the room
or for the occupied zone:
soz
sec
cc
cc
(4.18)
where
coz = the mean contaminant concentra-
tion in the occupied zone
4.6.2 Personal exposure index.
Thermal flow around a person and flow gen-
erated by the movement of a person may give
an inhaled concentration that is different from
the concentration at head height if the meas-
urements are made without a person.
This can be expressed by the following per-
sonal exposure index, Brohus and Nielsen
(1996a):
sexp
secexp
cc
cc
(4.19)
where
cexp = the inhaled concentration.
Short circuit flow,contaminant sourceclose to exhaustopening
0 % 50 % 100 %
Short circuit flow
Perfectmixing
Displacementventilation
Displacementventilation,contaminantsource withoutheat generation
Air change efficiency εa
Pis
ton f
low
Conta
min
ant
rem
ova
lef
fect
iven
ess
εc
<1
1
>1
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
31
It is possible to work with a stratification
height that is lower than the height of the
breathing zone. The personal exposure in-
dex will often be larger than the local ven-
tilation index because clean air is moved
from the lower part of the room up to the
breathing zone by the free-convection
boundary layer around the person (Fig-
ure 4.28 and 4.29).
Figure 4.28. Thermal flow around a person
may give cleaner breathing air.
Figure 4.29. Iso-concentration map showing
the dispersion pattern of a tracer gas emitted
directly above a 4 W heat source in the lower
zone. The dummies are situated in the measur-
ing plane (Stymne et al, 1991).
Usually the stratification height will be
around 1 m in a room when the air distribu-
tion is designed for an appropriate temper-
ature distribution. The concentration in the
inhalation air cexp of a standing person can
be found from (Brohus and Nielsen 1996a).
)( fp
exp
stpexp cc
z
zcc zst < zexp (4.20)
where zst is stratification height, zexp is
height to the breathing zone, cp the concen-
tration at breathing height outside the
breathing zone and cf the concentration in
the lower zone (floor level). cexp is equal to
cp when zst > zexp (Figure 4.30).
Figure 4.30. Inhalation of air in a room with
stratified flow and stratification layer below the
breathing zone.
The transport of clean air in the personal
boundary layer can only take place when
people are not moving. Figure 4.31 shows
that the effect disappears when a person is
moving with a speed equal to, or greater
than, 0,2 m/s in a room with a stratified
layer (concentration cp at head height)
(Bjørn and Nielsen 2002).
Figure 4.31.Concentration in inhalation cexp
relative to concentration in front of the breath-
ing zone cp versus speed of movement.
zexp
cexp cp
cf cR c
zst
z
cexp / cp
1,4
1,2
1,0
0,8
0,6
0,4
0,2
0,00,0 0,2 0,4
Speed [m/s]
0,6 0,8 1,0
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
32
Moving people will also influence the ver-
tical contaminant distribution as seen in
Figure 4.32. The figure shows the relative
concentration distribution for four seden-
tary people and for two sedentary people
and two people in motion. It is shown that
two people in motion are able to smooth the
vertical gradient slightly, but it is possible
to observe a stratification of CO2 (Nielsen
1992a and Brohus and Nielsen 1994). Hal-
vonova and Melikov (2010) also reported
that in a room with displacement ventila-
tion walking people will destroy the strati-
fication and will decrease the quality of air
inhaled by seated room occupants. The dis-
turbance decreases with the distance be-
tween the walking person(s) and the air
supply diffuser(s).
Figure 4.32. Concentration distribution in a
room with thermal manikins, sedentary people
and people in motion.
Transport of air from the lower zone to the
breathing zone is an additional positive ef-
fect in displacement ventilation, but in the
case of movement the lack of air from the
lower zone means that the concentration in
the inhalation air corresponds to the fully
mixed concentration which will be found in
rooms with mixing ventilation.
Measurements of the personal exposure in-
dex made in situations with air movement
in the occupied zone and contaminant
sources close to a person can give rise to a
very small exposure index (Brohus and
Nielsen 1996b).
Displacement ventilation should not be
used when the contamination sources are
mostly cold.
As pointed out above, the ventilation flow
rate should not be set equal to the convection
flows above the people present in a room,
because this will, in practice, lead to too high
air flow rates. Figure 4.33 shows the im-
provement in inhaled air quality relative to
the air quality in the ambient as a function of
the ventilation flow rate per person.
Figure 4.33. The ratio between the concentra-
tion in the breathing zone and in the ambient air
at the same height (Etheridge and Sandberg,
1996).
With a ventilation flow rate of
0,020 m³/(s·person), the boundary is above
the person. However, a ventilation flow rate
of 0,010 m³/(s·person) gives a concentra-
tion of only 20 % of the concentration in the
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
4. PERFORMANCE OF DISPLACEMENT VENTILATION
33
ambient at the same level. Figure 4.33 ap-
plies when the breathing zone is in the up-
per layer, i.e. when the sum of the convec-
tion flow rate is larger than the ventilation
flow rate. In these cases, the concentration
in the upper layer is almost equal to the ex-
haust concentration and the reduction in
concentration in the inhaled air can be cal-
culated from Figure 4.33. Measurements
by Mundt (1994) showed the rapid, almost
instantaneous, recreation of the thermal
flow around a person when the person
moves from one place to another in a room.
Particle transportation in a displacement-
ventilated room was studied by Mundt
(2001). The results indicate that there seem
to be little risk of resuspension of particles
from the floor into the supply air flow. The
sizes studied were however only particles
larger than 0,5 μm and more research is
needed for smaller particles. Recent studies
(Rim and Novoselac 2009, Licina et al.
2015a) reveal that the human convective
boundary layer transports particles of small
size and substantially influences the per-
sonal exposure when the pollution origi-
nates at the low level, for example, foot
level. With stratified airflow patterns, such
as displacement ventilation the inhaled con-
centration of particles generated at floor
level and the near proximity to an occupant
may be several times higher than the ambi-
ent concentrations.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
34
5 Calculation of supply airflow rate
Two principal methods can be used when
the supply air flow rate of displacement
ventilation system is calculated: 1) temper-
ature based design, where the design crite-
rion is the air temperature in the occupied
zone of the room and 2) air quality based
design where the design criterion is the air
quality in the occupied zone.
In Chapter 4.4, the thermal plume equations
are presented that are applied in the air
quality based design. Based on the known
heat loads and their locations, the supply air
flow rate is set by calculating the air flow
rate induced by convection flows.
In commercial buildings, the removal of the
excess heat is likely to be the main concern.
The cases where cooling is the main issue,
the temperature based design is the most
commonly applied method. For that reason,
this chapter focuses only on temperature
based design.
5.1 Temperature based design methods
In the design process, the challenging task
is to estimate vertical contaminant or tem-
perature gradients in the room space. While
the contaminant stratification level is
mainly affected by the relation of supply air
flow rate and convective air flow rate, ther-
mal stratification is also affected by thermal
radiation exchange between different room
surfaces. The thermal radiation from upper
level surfaces warms lower level surfaces
and thus affects the air temperature at floor
level and in the occupied zone.
In the case of steady-state conditions when
the supply air flow rate in the room is de-
creased, the vertical temperature stratifica-
tion in the occupied zone and the ceiling
temperature will increase. This implies that
the thermal radiation from the upper zone to
the lower zone will also increase and so will
increase the air temperature at the floor level.
In a real building, the thermal mass of the
building also influences the room air tem-
perature. The resulting temperature will be
dependent on the thermostat location and
the selected control strategy, i.e. variable
airflow volume (VAV) strategy or constant
airflow volume (CAV) strategy.
The displacement ventilation design meth-
ods applicable for manual calculations are
based on empirical coefficients and nomo-
grams, in which the influence of thermal ra-
diation exchange between the upper and
lower parts of the room is included (Skistad
1994, Halton Oy Design Guide 2000, etc.).
The advantage of these methods is their
ease of use and also the accuracy of the es-
timation of temperature and contaminant
distribution particularly in industrial type
applications.
In modern practice, it is more common to
use simulation software where the contam-
inant and temperature gradients are mod-
elled. In some models, the temperature dis-
tribution is modelled to be linear over the
room height, i.e. a constant vertical temper-
ature gradient is assumed (Mundt 1996,
Arens 2000, Nielsen 1995, 2003). Temper-
ature based design methodology where the
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
5. CALCULATION OF SUPPLY AIRFLOW RATE
35
space is divided into zones: the lower occu-
pied zone and the upper unoccupied zone is
introduced by Livchak and Nall (2001).
The heat loads are split into two zones ac-
cording to their type and location. The radi-
ation between upper and lower zones is also
taken into account when the air tempera-
tures of the zones are iteratively solved.
An approach to determine the required air
flow rate and the supply temperature by use
of fractional coefficients applied for three
selected heat loads is suggested (Chen et al.
1999, Chen and Gliksman 2003). The frac-
tional coefficients set the ratio of the con-
vective heat load that is released in the
room space between head and foot level.
Several nodal models have been introduced
that allow different slopes for the tempera-
ture profile between nodes. Three-node
models (Li et al. 1992, da Graça 2003 and
Mateus and da Graça 2015) and multi-node
models have been proposed that apply jet
and thermal flow elements to track individ-
ual jets from heat or mass sources (Erikson
et al. 2012). Multi-zone models where air
flow rates between the nodes are predefined
by a CFD method has also been proposed
(Rees 1998 and Griffith 2002).
The influence of coupling displacement
ventilation with chilled ceilings or floor
heating on the temperature distribution
have been analysed (Novoselac et al. 2006,
Rees and Haves 2001 and Rees and Haves
2013).
Compared with CFD-simulation the nodal
models require less computation time, they
are more suitable for engineering calcula-
tions and can be added to the whole build-
ing simulations.
Some nodal models are currently available
in energy simulation tools. Rees’ model can
be used with ESP-r (Hensen and Hamelinck
1995). Mundt (1996) and da Graça (2003)
models are implemented in DOE Ener-
gyPlus (2015). Mundt (1996) and Erikson
et al. (2012) models are incorporated in
IDA ICE (Sahlin 1996).
5.2 Calculation of vertical room air temperature distribution
A heat-balance-based method is used when
excess heat is considered the main indoor
climate concern. In rooms with displace-
ment ventilation the vertical temperature
distribution depends on the output, charac-
teristics and location of the heat sources and
on the airflow rate.
The following chapters introduce three
models that calculate the vertical tempera-
ture distribution in rooms with displace-
ment ventilation.
5.2.1 Linearized vertical temperature
distribution calculated by
Mundt model
In this model, the radiative energy transfer
from the ceiling to the floor is balanced by
convective heat transfer from the floor sur-
face to the air at floor level.
αr·Af·(θe – θf) = αcf·Af·(θf – θaf) (5.1)
The convective heat transfer from the floor
is in turn equal to heat transferred to the
supply air near the floor, neglecting any in-
duction of room air into the supply air flow
upon entering the room:
ρ · cp · qs ·(θaf – θs) = αcf·Af·(θf – θaf) (5.2)
Where qs= airflow rate (m³/s).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
36
The dimensionless temperature κ, defined
by equation 4.1, can be calculated accord-
ing to equation 4.3 introduced in Chapter 4
(Mundt 1996).
Together with an energy balance (equation
5.3), it is possible to estimate vertical room
air temperature profile and to select the re-
quired supply air flow rate.
tot = qs · ρ · cp · (θe – θs) (5.3)
Using the energy balance equation (5.3)
and the dimensionless temperature in equa-
tion 4.3, the supply air temperature may be
determined:
θs = θaf − · tot /(ρ · cp · qs) (5.4)
5.2.2 Vertical temperature distribu-
tion calculated by Nielsen model
In the Nielsen model (Nielsen 1995 and
2003), a linear temperature gradient be-
tween floor and the height of mixing layer
(stratification height) is predicted.
Over the mixing layer, the room air temper-
ature is assumed to be constant. The mixing
layer temperature (that is the same as the
exhaust air temperature) is calculated with
the energy balance (equation 5.3). The floor
temperature is determined with specific
ArA- number of supply air:
ArA = β·g·H·(θr – θs)/(qs/Af)² (5.5)
The height of the mixing layer zt is calcu-
lated:
zt = 0,62·Φcf2/5·(θe – θaf)
-3/5 (5.6)
where Φcf = total convection load.
The vertical temperature profile is given
from the floor temperature, Figure 5.1, and
the stratification height zt:
af
t
afe
zz
for z > zt (5.7)
θ = θe for z > zt (5.8)
Using equations 5.3 – 5.8 and Figure 5.1, it
is possible to determinate the room air tem-
perature distribution.
Figure 5.2 shows measurements of vertical
temperature distribution in a room with
four thermal manikins as heat sources. The
predicted temperature profiles are found
from Figure 5.1 and equation (5.6). The
predictions seem to give an improved de-
scription of the vertical temperature distri-
bution in comparison with a linear distribu-
tion over the entire height of the room.
Figure 5.1. Dimensionless floor temperature
versus the Archimedes number Nielsen (1995).
0,8
0,6
0,4
0,2
0,040020010060402010
ArA.10-3
κ
ABC
D
A = Distributed heat sourceB = Sedentary personsA = Ceiling lightA = Point heat source
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
5. CALCULATION OF SUPPLY AIRFLOW RATE
37
Figure 5.2. Vertical dimensionless temperature
distribution in a room with four thermal mani-
kins, Nielsen (1995).
5.2.3 Vertical temperature distribu-
tion calculated by Mateus and
da Graça model
The estimation accuracy of the vertical
temperature distribution can be improved
by modelling based on three-nodes (Mateus
and da Graça 2015 and Kosonen et al.
(2016). In this way, it is possible to obtain
different slopes of the vertical temperature
profile between the nodes. Three room air
temperatures are predicted: at heights of
0,05 m, and 0,65 m above floor level and at
the mixed layer.
Mateus and da Graça model (Figure 5.3) in-
cludes convective energy conservation
equations for the three-nodes and radiative
energy conservation equations for the room
surfaces: floor, ceiling and two lateral wall
areas.
In the model, two additional parameters IM
and FMO are included. IM describes entrain-
ment generated accumulated flow rate. De-
fault value of IM is 0,6. The parameter of
FMO characterizes the fraction of the con-
vective heat loads that are mixed into the
occupied zone, and, therefore, not con-
veyed directly to the mixed layer
(0 < FMO <1). The lower level mixing does
not occur in an ideal displacement system
(FMO =0). A default value of FMO can be
used 0,4.
Figure 5.3. Mateus and da Graça (2015) model
scheme.
The radiation between the surfaces is mod-
elled by dividing the wall into lower (Awl)
and upper (Awu) surface areas. The height of
the lower wall area (hTmx) is calculated with
the height of ceiling and the height of the
mixing layer (equation 5.9):
hTmx = hmx + (H – hmx)/3 (5.9)
ArA=41.103 zt / H = 0,83
ArA=28.104 zt / H = 0,62
z /
H 1,0
0,8
0,6
0,4
0,2
0,00,0 0,2 0,4 0,6 0,8 1,0 1,2
θ – θs
θe – θs
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
38
The convective heat balance for the three-
nodes can be set by the following equa-
tions:
ρ·cp·qs·(θaf – θs) – IM·ρ·cp·qs·(θoc – θaf)
= αcf,f·Af(θf – θaf) (5.10)
ρ·cp·qs·(θoc–θaf)+IM·ρ·cp·qs·(θoc–θaf)–Φcf·FMO
= αcf,wl·Awl·(θwl– θoc) (5.11)
ρ·cp·qs·(θmx – θoc) – Φcf·(1−FMO) =
αcf,c·Ac·(θc – θmx)+ αcf,wu·Awu·(θwu – θmx) (5.12)
Where Φr,f, Φr,w and Φr,c = radiant heat load
in on floor, wall and ceiling. αcf,f, αcf,wl, αcf,wu
and αcf,c are convective heat transfer coeffi-
cients on floor, lower and higher level walls
and ceiling, αr is radiative heat transfer co-
efficients, θaf is room air temperature at the
height 0,1 m, θoc is room air temperature at
the height 0,65 m, θmx is room air tempera-
ture at the mixed layer.
Without detailed information of heat
sources, the breakdown between convec-
tive and radiative heat loads of 50 % / 50 %
can be used. The height (hmx) of the mixing
layer is calculated with the following ther-
mal plume equation where the plumes are
assumed to be like a point source and lo-
cated at the floor level.
hmx = 23,95·(qs3/Φcf)
1/5 (5.13)
With the seven linear equation (5.10–5.12
and 5.14–5.17), it is possible to solve for
the seven unknown room air nodes and sur-
face temperatures.
The mixed layer temperature is assumed to
be equal to the exhaust temperature. The ra-
diant heat exchange equations for the room
surfaces are introduced in the room surface
energy conservation equations, considering
the equal impact of the radiative heat trans-
fer to all the surfaces:
αcf,c·(θc – θmx) + αr,c·(θc – (θfAf + θwlAwl + θwuAwu)/(At – Ac)) = Φr /At (5.14)
αcf,f·(θf – θaf) + αr,f·(θf – (θcAc + θwlAwl + θwuAwu)/(At – Af)) = Φr /At (5.15)
αcf,wl·(θwl – θoc) + αr,wl·(θwl – (θcAc + θfAf + θwuAwu)/(At – Awl)) = Φr /At (5.16)
αcf,wu·(θwu – θmx) + αr,wu·(θwu – (θcAc + θfAf + θwlAwl)/(At – Awu)) = Φr /At (5.17)
Where Ac, Af, Awl, Awu and At are areas of ceiling, floor, lower level wall, higher level wall
and total area of all surfaces.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
5. CALCULATION OF SUPPLY AIRFLOW RATE
39
5.3 Vertical position of the heat source
A vertical location of the heat source may
also influence the temperature distribution,
Figure 5.4A. This effect is especially
pronounced when the source may be
considered as a point heat source with
limited radiation. Figure 5.4B shows
measurements of two vertical temperature
profiles for sources with two different
vertical locations (Nielsen 1995).
Figure 5.4. A) Vertical dimensionless position
of the heat source in the room. B) Dimension-
less temperature distribution in a room with a
point heat source located at two different verti-
cal positions, (Nielsen 1995).
5.4 Calculation examples when using temperature based design models
In this chapter, three models are used in cal-
culation examples of three simplified de-
sign cases where the input data for the room
geometry, heat transfer coefficients, air
properties and the supply and target tem-
peratures are presented in Table 5.1.
Table 5.1. Input data of calculation examples.
Size of the room:
Height H 5,12 m
Width W 4,55 m
Length L 4,55 m
Heat Transfer Coefficients of the room surfaces:
Convective heat transfer coefficients:
Ceiling αc,c 1,5 W/(m²∙K)
Floor αc,f 4,0 W/(m²∙K)
Wall surface below the
mixed layer αc,wl 1,5 W/(m²∙K)
Wall surface above the
mixed layer αc,wu 1,5 W/(m²∙K)
Radiative heat transfer coefficients:
Ceiling αr,c 5,8 W/(m²∙K)
Floor αr,f 5,8 W/(m²∙K)
Wall surface below the
mixed layer αr,wl 5,8 W/(m²∙K)
Wall surface above the
mixed layer αr,wu 5,8 W/(m²∙K)
Air properties:
Density ρ 1,2 kg/m³
Specific heat capacity cp 1005 J/(kg∙K)
Thermal expansion
coefficient β 3,43 10-3 1/K
Case studies:
Supply air temperature θs 18 °C
Air temperature at the
height 1,1 m θ1.1 23 °C
The room and the combinations of heat
loads used in the following examples are
the same as described in Chapter 8.2, where
measured and calculated room air tempera-
ture profiles are compared.
Using temperature based design models of
Mundt, Nielsen and Mateus and da Graça
for the fixed room air temperature (23 °C at
1,1 m level), the required air flow rate
should be calculated by using an iterative
method.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
40
Calculation example 1: Single buoyant flow element of occupants (900 W)
Heat load of 12 people (simulated by
heated cylinders with height of 1,0 m, di-
ameter 0,3 m) generating 75 W heat, i.e.
total 900 W, heat load in the room.
Vertical temperature profiles obtained
with three models are shown in Figure
5.5. The estimated vertical temperature
distribution calculated with the three
models varies. The vertical air tempera-
ture distribution between floor and ceil-
ing is linearized with the Mundt model,
while Mateus and da Graça and Nielsen
models estimate the mixing layer starting
at 1,1 m and 3,0 m respectively. The re-
quired air flow rates were 0,073 m³/s
(Mundt), 0,096 m³/s (Nielsen) and
0,149 m³/s (Mateus and da Graça).
Figure 5.5. Temperature profiles of case 1.
Calculation example 2: Single buoyant flow elements of window (520 W)
Heat load of 520 W generated in the
room by heated window with height of
3,6 m and width of 3,6 m is considered.
The window is installed at height of
0,8 m above the floor.
The vertical temperature distribution in
the room obtained with three models is
shown in Figure 5.6. In Mundt model,
room air temperature is linearized be-
tween the floor and ceiling while Mateus
and da Graça and Nielsen models esti-
mated the mixing layer starting respec-
tively at 1,6 m and 2,6 m. The required
air flow rates were 0,049 m³/s (Mundt),
0.06 m³/s (Nielsen) and 0,072 m³/s (Ma-
teus and da Graça).
Figure 5.6. Temperature profiles of case 2.
Mundt model qs
= 0,073 m³/s
0
1
2
3
4
5
6
20 22 24 26 28 30
Hei
ght,
m
Temperature, °C
Nielsen model qs
= 0,096 m³/s
Mateus, da Graca qs
= 0,149 m³/s
Hei
ght,
m
Mundt model qs
= 0,049 m³/s
Nielsen model qs
= 0,06 m³/s
Mateus, da Graca qs
= 0,072 m³/s
0
1
2
3
4
5
6
20 22 24 26 28 30
Temperature, °C
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
5. CALCULATION OF SUPPLY AIRFLOW RATE
41
Calculation example 3: Combination of heat loads (1762 W)
In this case, heat load in the room is re-
sult of:
• Heat load (900 W) of 12 people simu-
lated by 12 heated cylinders (height
1,0 m, width 0,3 m) each generating
75 W heat;
• Heat load (520 W) from window
(height 3,6 m and width 3,6 m) in-
stalled 0,8 m above the floor;
• Heat load (260 W) from the floor
(floor heated area 2,4 m x 2 m);
• Heat load (232 W) from fluorescent
lighting units installed at ceiling.
The vertical temperature distribution cal-
culated with three models is shown in
Figure 5.7.
Figure 5.7. Temperature profiles of case 3.
The vertical air temperature distribution
is linearized between floor and ceiling
with the Mundt model while Mateus and
da Graça and Nielsen models estimated
the beginning of the mixing layer respec-
tively at 2,9 m and 2,2 m. The required
air flow rates were 0.118 m³/s (Mundt),
0.208 m³/s (Nielsen) and 0.202 m³/s
(Mateus and da Graça).
Conclusion of the calculations In this chapter, the room air temperature
gradient and required supply air flow rates
are calculated using three temperature
based design models. Two of the examples
are with single buoyant flow elements (oc-
cupant and warm window) and in one of the
examples a combination of typical heat
loads in office is used.
The examples demonstrate that the calcu-
lated supply air flow rates are quite differ-
ent with the three models. When the room
air temperature is assumed to be linear be-
tween the floor and ceiling level, the re-
quired air flow rate is lower than that calcu-
lated with the models predicting the height
of mixing layer.
In design work, it is recommended to use a
model that calculates the height of the mix-
ing layer. At the moment, the best average
accuracy of the simplified models is given
by the Nielsen and Mateus and da Graça
models. However, the accuracy is depend-
ing on the type of the flow element (see
more in Chapter 8.2).
0
1
2
3
4
5
6
20 22 24 26 28 30 32
Hei
ght,
m
Temperature, °C
Mundt model qs
= 0,118 m³/s
Nielsen model qs
= 0,208 m³/s
Mateus, da Graca qs
= 0,202 m³/s
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
42
6 Air diffusers for displacement ventilation
6.1 Commonly used diffusers
There are several types of diffusers used for
displacement ventilation. The most com-
monly used types are integrated in the
walls. Other types are placed at the walls or
in a corner, free-standing on the floor, or in-
tegrated in the floor. The layout of the room
should be considered in connection with the
selection of the type of air diffusers.
Figure 6.1 shows circular and semi-circular
diffuser and Figure 6.2A shows a wall
mounted diffuser.
The air is supplied in the lower part of the
room and it is therefore convenient to locate
the diffusers, near to corridors and other un-
occupied areas to obtain an area in front of
the diffusers where a higher velocity can be
tolerated.
The diffusers in Figure 6.1 will create a ra-
dial flow at the floor close to the diffusers
because of gravity effect on the cold air
leaving the diffusers and partly because of
the diffuser design. Plane wall mounted dif-
fusers may also create radial flow at the
floor close to the wall. Figure 6.2A and B
shows how the cold low velocity air supply
at the surface of a diffuser falls towards the
floor and creates a radial flow at the floor.
Figure 6.3 shows a wall mounted diffuser
with integral nozzles for the adjustment of
the supplied flow pattern. The flow close to
the diffuser can be directed parallel to the
wall with only a small amount of forward
directed flow. The diffuser will create a
short “adjacent zone” (discussed in Chapter
6.4.1).
Semi-circular, wall mounted
Semi- circular, corner mounted
Circular, free standing
Figure 6.1. Semi-circular and circular diffusers.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
43
A
B
Figure 6.2. A) plane wall mounted diffuser with
air velocity directed in the direction perpendic-
ular to the diffuser surface. B) Smoke experi-
ment showing how the flow creates a radial pat-
tern on the floor close to the diffuser although
the air is supplied perpendicular to the diffuser
surface.
Figure 6.3. Wall mounted low velocity diffuser
with integrated nozzles for the adjustment of the
flow patterns at the surface.
Custom made diffusers integrated into the
room design can also be seen in many situ-
ations.
With a floor mounted diffuser, there is a
draught area in the flow above the diffuser
(Figure 6.4A) but the diffuser does not create
much flow along the floor. Figure 6.4B
shows a carpet diffuser which can cover the
whole room. This type of supply generates a
very low velocity and in this case the air
movement in the room originates from the
heat load. However, care should be taken to
consider the contamination which could be
emitted from the carpet in this type of supply.
A
B
Figure 6.4. A) Floor mounted diffuser with high
velocity swirl supply. B) Supply of air through
the carpet.
Displacement diffusers can also be installed
above the occupied zone or in the ceiling.
They supply cold downward flow locally
for example by using a wall surface to sup-
port the flow into the occupied zone (Niel-
sen et al. 2010).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
44
6.2 Radial air flow or plane air flow from low-velocity diffusers
The diffusers described in Chapter 6.1 will
create a local flow in close proximity to the
diffusers. The flow in the occupied zone of
the room is not necessarily controlled by
this local flow, but will depend on the num-
ber and location of diffusers and on the
room geometry. Two types of flow can be
generated, namely (semi) radial flow or
plane flow, see Figure 6.5. A radial flow
along the floor will take place if the dis-
tance between the diffusers is as large as the
distance to the opposite wall while a plane
flow will take place if the diffusers are lo-
cated close to each other, or the room is
very long and narrow in the direction of
flow as indicated in Figure 6.5B. A dif-
fuser with a high sideway discharge, Fig-
ure 6.3, will also create plane flow even if
the diffusers are not particularly close to
each other (see Figure 6.19).
6.3 Air flow from low –velocity diffusers
Normally, the supply air is between 3 K and
5 K cooler than the room air. In these case,
the supply air falls towards the floor when
it leaves the diffuser, and spreads, like a
blanket, across the floor.
When the supply airflow is isothermal, i.e.
has the same temperature as the surround-
ing air in the lower part of the room, the
flow will be distributed horizontally into
the room according to the initial flow pat-
tern at the surface of the diffuser (Fig-
ure 6.6).
A
B
Figure 6.5. A) A diffuser creating radial flow
along the floor in the occupied zone due to the
use of a single diffuser and due to room geome-
try. B) A number of diffusers creating plane flow
in a room.
Figure 6.6. Isothermal air supply. The flow has
a constant velocity core as in a large three di-
mensional wall jet.
vx
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
45
As already discussed in Chapters 3 and 4 if
the supply temperature is higher than the
surrounding temperature, the supplied air-
flow will rise to the ceiling without spread-
ing in the occupied zone (Figure 6.7).
Therefore, displacement ventilation can be
used effectively only when the supply air is
cooler than the room air.
Figure 6.7. Supply of warm air.
The flow indicated in Figure 6.7 may be
accepted in special cases when the air dis-
tribution system is used for pre-heating
spaces in periods when unoccupied. This
could be night heating in, for example, of-
fices or pre-heating of a concert hall.
Asymmetrical heating or cooling loads may
give some recirculation in the room which
will mix the air in the case of warm air sup-
ply.
6.4 Air distribution from a low-velocity diffuser giving a radial flow in the occupied zone
6.4.1 The “Adjacent Zone”
The air from a single wall-mounted diffuser
flows over the floor and generates radial
flow. Close to the air supply diffuser there
is a zone where the flow has relatively high
velocity and low temperature. In this zone
the risk of draught may increase. This zone
is called “adjacent zone”. It is commonly
accepted that the length of the adjacent
zone, ln, is defined as the distance from the
diffuser to the point where the maximum
velocity has decreased to 0,2 m/s when the
temperature difference between the room
air (at 1,1 m height) and supply air is 3,0 K.
Draughts result from high mean velocity air
flow with high turbulence intensity and low
air temperature (Chapter 3.3). For a given
velocity (for example, 0,2 m/s) the risk of
draught will be low when the air tempera-
ture is high and conversely, the risk will be
high when the temperature is low. There-
fore, determination of the near zone based
on only the velocity is misleading.
ISO Standard 7730 (2005) suggests three
categories A, B and C of thermal environ-
ment with corresponding 10, 20 and 30 %
dissatisfied occupants due to draught. In
some cases, less stringent requirements
may be used if a room space is used, for ex-
ample, by moving people. The draught risk
is calculated based on measured or pre-
dicted local mean velocity, turbulence in-
tensity and air temperature. Figure 6.8
shows an example of an adjacent zone de-
fined based on measurements of velocity,
turbulence intensity and temperature near
the floor in a room with displacement ven-
tilation.
Figure 6.8. Adjacent zone defined by velocity of
0,2 m/s and by 10, 20 and 30 % dissatisfied oc-
cupants (Melikov and Langkilde 1990).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
46
The adjacent zone based on the draught risk
(DR) of 10, 20 and 30 % is shown. The com-
parison shows that the adjacent zone defined
by the velocity of 0,2 m/s does not comply
with the requirements in the standard and
there is risk more than 30 % of the occupants
will be dissatisfied due to draught.
From Figure 6.8, it becomes clear that the
length of the adjacent zone will depend on
the category of the designed thermal envi-
ronment. The same category, i.e. the same
level of draught risk, can be achieved by
different combinations of velocity, turbu-
lence intensity and air temperature. This is
shown in Table 6.1.
As discussed in Chapter 6.2, if the diffus-
ers are located close to each other or the
room is very long and narrow in the direc-
tion of flow a plane flow will be generated
(Figure 6.5B). In this case the flow veloc-
ity does not decrease with the distance
from the diffuser (see Chapter 6.5) and it
is impossible to define an adjacent zone.
Therefore, it is more correct to relate the
required comfort to the draught rating
model discussed in Chapter 3.3.
In practice, the number of diffusers and the
diffuser design, room dimensions and fur-
nishing may give a semi radial flow with a
higher velocity and longer adjacent zone
than expected for the given diffuser type in
the radial flow case, see Chapter 6.4.5
“Flow between obstacles” and (Melikov
and Langkilde 1990, Nielsen et al. 2004).
Avoiding draught is the major challenge in
developing low velocity air diffusers. To
reduce the size of the adjacent zone, the
number of diffusers in the room or diffuser
face area should be increased. This also
generates a more homogenous indoor envi-
ronment in the occupied zone. Different
diffuser types may be used. Typically,
those that supply air in only one direction
will generate a longer adjacent zone than
those distributing supply air from semi-cir-
cular or circular diffusers. One way to re-
duce the draught in the occupied zone is to
direct the supply air sideways to the wall
outside the occupied zone. Figure 6.9
shows two typical adjacent zones estab-
lished with forward discharged flow and
with sideway discharged flow.
Table 6.1. Adjacent zone of the same category can be achieved by different combinations of local
air temperature and velocity of the supplied flow.
Draught rate requirement (1 Local airflow temperature[°C] Local airflow velocity[m/s]
Category A: 10 %
19 0.11
21 0.12
23 0.14
Category B: 20 %
19 0.19
21 0.21
23 0.25
Category C: 30 %
19 0.27
21 0.30
23 0.35
1) Draught rate calculated according to ISO-7730 indoor environment category by assuming typical turbulent intensity of
20 % in displacement ventilation flow.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
47
Figure 6.9. Adjacent zones for wall-mounted
diffusers. a) Forward discharge, b) Sideways
discharge. Diffuser data: Height: H = 0,9 m,
Width: B = 0,6 m. Supply air flow: qs =
0,04 m³/s. Under-temperature: Δθs = 6 K.
6.4.2 Velocity distribution
Figure 6.5A shows how the cold stratified
layer of supply air flows into the occupied
zone as a radial air movement that covers
the whole floor in the room.
Nordtest (2002) differentiates between the
acceleration region near the diffuser and the
velocity decay region outside the accelera-
tion region (Figure 6.10). The flow near
displacement diffusers has been studied and
a model for the velocity distribution has
been developed (Magnier-Bergeron et al.
2017).
A typical height of the discharge flow is
about 20 cm in the velocity decay region.
The maximum velocity in the flow is lo-
cated at a height of approximately 2 to 5 cm
above – see Figure 6.10. This is also re-
ported by Melikov and Langkilde (1990).
Figure 6.10. Velocity distribution in front of a
diffuser, when the supply air is colder than the
room air (Jacobsen and Nielsen 1992).
Typical depth
20 cm
z
~ 2 - 5 cm
z
v
Acceleration region
Velocity decay region
vx
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
48
Measurements of this air movement show
that the entrainment in the horizontal flow
is very small and the height of the flow is
constant along the whole velocity decay re-
gion. The height of the stratified layer is a
function of the Archimedes number (Niel-
sen 2000). The height of the diffuser is im-
portant because the supplied cold flow is in-
fluenced by initial vertical acceleration due
to gravity. The Archimedes number is de-
fined as:
2
)(
f
soz
v
HgAr
(6.1)
or as the Archimedes ratio
2
)(
s
soz
q
(6.2)
where
β = volume expansion coefficient, 1/θoz K-1;
g = acceleration of gravity = 9,81 m/s²;
H= height of diffuser;
θoz – θs = under-temperature, i.e. differ-
ence between the temperature at a
height of 1,1 m inside the room and
the supply temperature;
vf = face velocity (qs/as);
qs = supply air flow rate.
A radial stratified flow with constant height
will have a velocity distribution that is in-
versely proportional to the distance from
the diffuser or from a point (virtual origin)
very close to the diffuser. Measurements of
the flow from diffusers confirm the theory
of this development (Nielsen 1992b, 2000
and Skåret 2000). Figure 6.11 shows an ex-
ample of measurements of maximum ve-
locity in the stratified flow along the floor
from a wall-mounted diffuser.
Figure 6.11. Maximum velocity close to the
floor versus distance x from the diffuser.
The cold air has a high initial acceleration
due to the buoyancy effect, and the highest
velocity is, in this case, obtained at a dis-
tance of 0,6 m from the diffuser. The meas-
urements indicate that the velocity vx is pro-
portional to 1/x for distances larger than
~1 m from the diffuser.
Therefore, it is possible to find the maxi-
mum velocity vx at any distance from the
diffuser when the adjacent zone ln, is known
from measurements. The maximum veloc-
ity is given from
[m/s]20x
l,v n
x (6.3)
where x is the distance from the diffuser.
The velocity vx will for example, be equal
to 0,075 m/s at a distance of 4 m if the
length ln is 1,5 m. The maximum velocity
will be located 2 to 5 cm above the floor
when the temperature difference is large,
but it will have a higher location if the tem-
perature difference is small. It is assumed
in equation (6.3) that ln is within the veloc-
ity decay region (Figure 6.11).
0,2 0,4 0,6 1,0 2,0 4,0 6,00,04
0,06
0,080,10
0,20
0,40
0,60
x [m]
v [m/s]x
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
49
It is also possible to find the velocity dis-
tribution in the occupied zone as a function
of the volume flow rate and the tempera-
ture difference. The maximum velocity is
given by
vx = qs KDr⋅(1/x) [m/s] (6.4)
where KDr is a function of the flow rate and
the under-temperature of the supply flow
(function of Archimedes number). The
equation is valid for the velocity decay re-
gion, x > 1−1,5 m.
The length of the adjacent zone, ln, can be
found from equation (6.5):
Drsn Kql 5 [m] (6.5)
KDr has to be measured for each individual
diffuser. Figure 6.12 shows the variation of
KDr for different types of diffusers. The fig-
ure indicates that the first generation of dif-
fusers – located in the upper part of the
shaded area – had a radial distribution of
the flow and a high level of the KDr value.
Some diffusers even had a forward dis-
charge of the flow at a low Archimedes
number, which in this situation will give a
further increase in KDr, but the gravity ef-
fect turns the flow into a radial pattern at
higher Archimedes numbers. The newest
generation of diffusers has a distribution
with high velocity parallel to the wall and a
lower velocity perpendicular to the wall
(sideways discharge). This will give the
low KDr values shown in the lower part of
Figure 6.12.
The upper part of the shaded area in the fig-
ure is therefore typical of diffusers with ra-
dial/axial distribution of velocity, Fig-
ure 6.1 and 6.2, and the lower part is typical
of a “flat” velocity distribution, Figure 6.3.
Some diffusers may also have initial high
mixing to reduce the velocity level along
the floor.
The Archimedes number may be reduced to
zero in some situations (for example, in the
situation where a CAV-system reduces θoz-
θs to zero by the temperature control). Fig-
ure 6.12 shows that KDr obtains a given
level and equation (6.4) describes the flow
as an isothermal radial wall jet, which will
be formed in the occupied zone.
Figure 6.12 indicates that KDr is a func-
tion of the square root of the Archimedes
number, or √𝜃𝑜𝑧 − 𝜃𝑠 / 𝑞𝑠. The maxi-
mum velocity vx will therefore be a linear
function of the square root of the Archi-
medes number.
Figure 6.12. KDr-values for different types of
wall-mounted diffusers for displacement venti-
lation.
.10-3 [ ]
16
14
12
10
8
6
4
2
01614121086420
θoz – θs
qs2
Ks2
m6
KDr (m-1)
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
50
The KDr value is expressed by the following
equation (Nielsen 2000).
o
mDr
eb,K 90 [m-1] (6.6)
where
e = factor that represents the initial in-
crease in the flow rate, which is due
to entrainment in the downward ac-
celerating flow close to the opening,
bm = factor that adjusts the flow in the di-
rection of the x-axis to the flow pro-
file generated by the diffuser, see
Nielsen (1994a, 2000),
αo = the angular width of the radial flow
close to the diffuser
δ = the thickness of the stratified flow de-
fined as the height from the floor to
the level where the velocity is 0,5 vx.
The variables are indicated in Figure 6.13.
Both e and bm are functions of the Archime-
des number.
Figure 6.13. The diffuser has a forward dis-
charge, bm > 1, and αo is smaller than π. The
distance xo can be ignored in many practical
cases.
An all-round conventional diffuser has a
KDr value of ~ 7 corresponding to δ ~ 0,1 m,
αo = π and bm ~ 1. Many of the early diffus-
ers had a radial flow distribution with a rel-
atively high level in the symmetry plane as,
for example, a bm value of 1,5. This will
give a KDr -value of 11, which is in good
agreement with the values given in Figure
6.12. A further increase in the velocity level
will be obtained by a design where αo is
smaller than π, which also is typical of an
early diffuser design.
A design with sideways discharge, see Fig-
ure 6.3, can for example be expressed by a
bm value of ~ 0,85, which gives a KDr value
of ~ 6, which is typical of the new genera-
tion of diffusers.
New diffusers with adjustable nozzles be-
hind the front cover can be set to provide
strong mixing and generate such a high
level of turbulence in front of the diffuser
that the high entrainment removes a large
part of the temperature difference in the
flow. The flow will therefore have a re-
stricted velocity in the direction away from
the diffuser. The restricted temperature dif-
ference in the flow means that δ will be
large and KDr will therefore be small. A
small KDr means a low velocity vx and, in
this case, that disturbance from other
sources in the room could dominate the
flow in the occupied zone.
A Nordtest method (2002) gives a test pro-
cedure for diffusers and an expression for
the velocity distribution that can be estab-
lished from measurements.
Example:
Calculation of the adjacent zone for a
wall-mounted diffuser. In practice, the
crucial question is: “How long will the
adjacent zone be for a given supply air
flow rate?” The following example
shows the calculation of the adjacent
zone for a wall-mounted diffuser with
adjustable nozzles behind the front cover
(Figure 6.14).
x
v
qs e q
s
b > 1m
0
0
x
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
51
Figure 6.14. The wall-mounted diffuser of the
example. H = 0,45 m, B = 0,54 m.
The KDr-value for the diffuser with a given
standard set up of the nozzles is given by:
78751182
,q
,Ks
sozDr
[m-1] (6.7)
This KDr-function could be evaluated from
laboratory measurements. From equation
(6.7) and (6.5) the values in Table 6.2 can
be calculated.
Table 6.2. Length of adjacent zone for
θoz – θs = 3 K.
qs [m³/s] KDr [m-1] ln [m]
0,02 8,63 0,86
0,03 8,14 1,22
0,04 7,97 1,59
0,06 7,85 2,35
The obtained length of the adjacent zone
may be too long for some applications. It is
possible to adjust the nozzles in the diffuser
to obtain a flow with more sideways dis-
charge. This adjustment will reduce the
KDr-values to a lower level, and decrease
the length of the adjacent zone.
6.4.3 Air distribution from free standing
and corner mounted diffusers
A frequently used low velocity diffuser is
the circular freestanding unit. The supply
duct can be either from below (as shown) or
from above.
Figure 6.15. Circular, freestanding unit.
This diffuser can be regarded as two semi-
circular diffusers standing beside each
other, i.e. it can supply twice as much air as
a semi-circular diffuser for a given length
of the adjacent zone. The KDr factor can be
expressed in the same way as for wall
mounted diffusers and the adjacent zone is
given by equation (6.5) and the velocity
along the floor by equation (6.3).
A corner-mounted diffuser can be regarded
as half a wall mounted diffuser or a semi-cir-
cular diffuser with regard to the length of the
adjacent zone. However, some manufactur-
ers make the discharge flow direct along the
walls in order to avoid draught along the
floor in the occupied space in the room.
Figure 6.16. Corner-mounted diffuser (Half of
a semi-circular diffuser).
B = 0,54 m
H =
0,45 m
l 0,2
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
52
6.4.4 Documentation for diffusers giv-
ing radial flow
Documentation that is suitable for use by
computer calculation methods should have
the diffuser constants given as a function of
the under-temperature and the supply air
volume flow:
KDr = f {(θoz – θs), qs} (6.8)
The length of the adjacent zone could then
be calculated from formula (6.5) and the
maximum velocity in the occupied zone
can be calculated from (6.4).
6.4.5 Flow between obstacles
The flow in the vicinity of the floor may be
influenced by furniture and by other obsta-
cles in the occupied zone. The maximum ve-
locity in the flow is located close to the floor
(between 2 to 5 cm above the floor), and a
large part of the air movement will therefore
take place in this region (Melikov and Lang-
kilde 1990, Nielsen 1992a). Furniture with
some air gap at floor level will only have a
small influence on the air movement while
obstacles placed directly on the floor will
block the flow. An opening between these
types of obstacles will work as a new supply
opening because the flow in the room is
stratified. Figure 6.17 shows an example
from a room with short movable walls.
Experiments have shown that the flow from
an opening between obstacles can be de-
scribed as a semi-radial flow like the air
movement from a wall-mounted supply
opening, Nielsen (1992b, 2000). The veloc-
ity decay can be described by the equation:
xK
q
vob
ob
x 1 (6.9)
vx is maximum velocity in distance x from
the opening and qob is the excess air sup-
plied on the other side. vx is measured in the
symmetry plane.
Figure 6.17. Office with short movable walls.
Flow through openings between obstacles and
definitions of the variables.
Figure 6.18 shows the measurements of
Kob in equation (6.9). The structure of equa-
tion (6.9) and the distribution of Kob -values
are equivalent to the structure of equation
(6.4) and the structure of KDr -values. The
temperature difference θoz − θob is the dif-
ference between the temperature at a height
1,1 m in front of the opening and the lowest
temperature in the opening between the ob-
stacles.
vx
x
θozθob
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
53
The width of the opening is varied from
0,1 m to 1,5 m in the experiment in Figure
6.18 and measurements show that the width
has limited significance.
Figure 6.18. Kob versus flow rate and tempera-
ture difference, Nielsen (1992b, 2000).
6.5 Air distribution from wall-mounted diffusers giving plane flow in the occupied zone
The flow from a number of diffusers placed
close to each other on the wall will merge
into a plane, stratified flow. See
Figure 6.5B. The velocity does not de-
crease with the distance from the diffusers
in this area, and can be expressed by the fol-
lowing equation, see Nielsen (1994b):
Dpl,sx Kqv (6.10)
The flow rate qs,l should, in this case, be
given as flow rate per m width of the main
air movement. The KDp value is a function
of the flow rate and temperature difference
(Archimedes number) and it is dependent
on the type of diffuser and on the installa-
tion of the diffusers (distance between dif-
fusers). The equation shows that the veloc-
ity is independent of the axial distance x,
but it must be assumed that the Archimedes
number has to have a certain level. The
same type of plane flow will also be gener-
ated in a narrow and deep room with a sin-
gle diffuser located at the end wall.
It is clear that it is not possible to work with
the adjacent zone theory for sizing except if
the velocity is high close to the diffuser. A
diffuser selection just requires that vx is
smaller than a given value, which will be
the case in the whole occupied zone up to
the location of the heat loads.
Even a single diffuser can generate a plane
flow in the occupied zone if the room is
deep and the diffuser has a high sideways
discharge. Figure 6.19 shows, as an exam-
ple, the flow from a wall mounted diffuser
and the direction of the velocity in the oc-
cupied zone.
Figure 6.19. Velocity distribution in a room
with a low velocity diffuser with high sideways
discharge.
.10-3 [ ]
16
14
12
10
8
6
4
2
086420
θoz – θob
qob2
°C s2
m6
Kob [m-1]
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
54
Figure 6.20 shows typical KDp values for
diffusers located close to each other (blue
and black) and for a diffuser with high side-
ways discharge (green), Nielsen et al.
(2004).
The Archimedes coefficient is given by
3
210
l,s
sozw
qar
(6.11)
Figure 6.20. Examples of KDp values for plane
flow versus an Archimedes coefficient arw.
6.6 Air distribution from floor-mounted diffusers
A floor-mounted diffuser supplies air verti-
cally from the floor, and they are often de-
signed to generate a flow with a swirl. In
this way, room air is entrained efficiently
into the primary air, which implies that the
velocity decreases very rapidly. Also, the
temperatures are mixed very rapidly. See
Figure 6.21.
Figure 6.21. Floor-mounted diffuser with
swirling flow.
An advantage of floor-mounted diffusers is
the large entrainment of room air into the
primary air. This makes them well suited
for large temperature differences, and they
are often used in rooms with high thermal
loads. The supply area is small and the sup-
ply velocity should have a sufficient mo-
mentum in the vertical direction (with a
supply velocity of 2 ~ 4 m/s). When using
floor-mounted diffusers, care should be
taken to apply the right air volume flow.
Too much air might be discharged up into
the upper layer, so that it creates a mixing
air distribution. On the other hand, too little
air will create too low momentum and in-
sufficient mixing with the room air, which
creates a cold air layer along the floor.
Figure 6.22 shows the vertical throw, zm,
from a floor mounted diffuser, Fitzner
(1989). This throw should not be confused
with the stratification height which is present
in a room with displacement ventilation.
Figure 6.22. Maximum height of flow from a
floor mounted diffuser and stratification height
in a room with displacement ventilation.
12
10
8
6
4
2
03020100
KD
p [
m-1]
arw [Ks2/m4]
zst
zm
21°C
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
6. AIR DIFFUSERS FOR DISPLACEMENT VENTILATION
55
The height zm can be found from, Koestel
(1954):
50
)(333
,
osoz
sa
o
m
a
vK,
a
z
(6.12)
The velocity decay in the upward flow for
a floor-mounted diffuser with swirl can be
described by the same equation as for a
free, circular jet:
z
aK
v
v oa
o
z 2
(6.13)
where
vz = the maximum velocity at the dis-
tance z above the floor;
vs = supply velocity for floor mounted
diffusers = qs /ao;
ao= the supply area of the diffuser;
Ka= wall jet diffuser constant.
The velocity decay for two different floor-
mounted diffusers is shown in Figure 6.23.
It can be seen that the velocity decays much
more rapidly for a jet with swirl than for a
jet without swirl. The value of Ka for the
free jet (without swirl) is 6,8, and the Ka-
value for the jet with swirl is 0,42. See Niel-
sen et al. (1988). This expresses how the
swirl will generate a high entrainment and
very fast velocity decay. The diffuser is of-
ten used in a group of four within an area of
0,6 m x 0,6 m. The velocity level will in
this case be higher than the velocity level
from a single diffuser but both arrange-
ments will have the same velocity level at a
height of 0,8 m (Bauman 2003).
Figure 6.23. Dimensionless velocity decay vz/ vo
in a free jet and in a jet with swirl versus the
height z above the floor.
Wall-mounted and floor-mounted diffusers
can handle a load in the room of ~ 50 W/m².
Carpet diffusers, which are designed as an
upper part of a double floor covered by a
carpet, can handle loads larger than
100 W/m². Carpet diffusers have a surface
(floor) temperature θf equal to θs, which in-
fluences the vertical temperature gradient
and make the gradient larger than in a dis-
placement ventilated room with one of the
other types of diffusers. Carpet diffusers do
not create any velocity in the room due to
the extreme low supply velocity qs/Afloor.
The velocities in the room are generated by
the heat sources in the room and the design
procedure is mostly about finding the max-
imum thermal load giving an acceptable ve-
locity level (Nielsen 2011).
0,01
0,02
0,030,04
0,060,080,10
0,2
0,30,4
0,60,81,0
1 2 4 6 10 20 40
z
Free jet
Jet with swirl
ao
vz / vo
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
56
7 Design of displacement ventilation
7.1 Design criteria
Air quality based design is typically used in
industrial applications where the contami-
nant stratification plays an important role.
Chapter 4.4 provides the equations for air
quality based design.
Temperature based design is the most com-
mon method in commercial buildings.
Chapter 5 provides methods of how to eval-
uate vertical room air temperature distribu-
tion.
A calculation method should be used that is
suitably accurate for the particular heat
loads. Dynamic energy simulation should
be used in order to accurately estimate the
cooling load. Simplified steady-state calcu-
lations typically overestimate the actual de-
mand and cannot take into account the ef-
fect of thermal mass on room air tempera-
ture. In demanding design cases a full-scale
mock-up is recommended together with
CFD.
7.2 Design of air distribution
Design of air distribution is crucial in dis-
placement ventilation to guarantee draught
free conditions and good air quality across
the whole occupied zone. The principles for
the calculation of the required air flow rate
are presented in Chapter 5. The required to-
tal air flow rate is the basis for the selection
of supply units. In Chapter 6, the perfor-
mance of different types of displacement
ventilation air diffusers is discussed. Spe-
cifically, the adjacent zone, close to the
units, should be analysed to prevent
draught.
7.2.1 A design chart for the room air
distribution systems
Figure 7.1 provides a chart that describes
the main considerations required when de-
signing air distribution for rooms that re-
quire cooling (Nielsen 2007). The chart is
based on the minimum and maximum al-
lowable flow rate qs supplied to the room,
and also on the maximum temperature dif-
ference between return air and supply air.
Figure 7.1 indicates the required outdoor
air minimum flow rate into the room to ob-
tain a given air quality. The minimum flow
rate can, for example, be that given by
standard EN 15251 (2007). It can also be
seen that draughts will occur at a particular
flow rate which is dependent on the temper-
ature difference θe−θs and on the type, and
location, of diffusers.
Figure 7.1. A design chart that indicates the re-
strictions on the flow rate qs and on the tempera-
ture difference θe-θs between return and supply.
The temperature difference, θe−θs, between
the return and supply air is also limited as
indicated in Figure 7.1. An excessive tem-
perature difference may either cause
draught in the occupied zone or create an
extreme temperature gradient in the room.
θe – θs
qs
Draught or/andtemperaturedifference
Draught
Low
air
qua
lity
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
7. DESIGN OF DISPLACEMENT VENTILATION
57
Figure 7.2 shows an example on a design
chart for an office with two persons and a
give displacement diffuser. The curve
shows the combination of θe−θs and qs
which encloses an area that fulfils thermal
comfort requirements.
In Figure 7.2, a maximum velocity of
0,2 m/s is accepted where the stratified
flow enters the occupied zone (adjacent
zone equal to 1 m) and it gives the re-
striction indicated by the blue line (equa-
tions 6.4 and 6.5). The red vertical line
shows the minimum flow rate based on air
quality. The blue horizontal dotted line
shows the restriction on the temperature
difference to the room which limits the ver-
tical temperature gradient in the room to
2,5 K/m. Thus, the area enclosed by the
curve, the horizontal line and the vertical
line (the white area) defines the permitted
ranges for variation of θe-θs and qs in a room
with this type of displacement ventilation.
Figure 7.2. Design chart for the air distribution
where lines show the different limitations for the
air flow rate (m³/s) and the temperature differ-
ence, which ensure thermal comfort in the room
(white area) (Nielsen 2007)
7.2.2 Location and number of units
When the total required supply air flow rate
is calculated the diffuser type, the number,
and locations can be selected taking into ac-
count the following considerations:
• dimension of the room
• the location of heat loads and/or pollutant
sources
• possible locations of units
• internal obstacles of the flow
• restriction of installation
• architectural aspects
Typical standard diffuser types are: wall
mounted, corner mounted, freestanding and
floor mounted. The diffusers require a cer-
tain wall and floor area. With regard to the
type and location of diffusers the following
aspects should be taken into account:
• the supplied air should be uniformly dis-
tributed in the room by a sufficient num-
ber of units;
• special attention needs to be given to the
adjacent zone around the diffuser so that
it is as small as possible.
Knowing the total airflow rate, it is possible
to estimate the number of required supply
units. Table 7.1 (Halton 2000) presents a
preselection of the supply unit.
As a rule of thumb in a large open space
layout, the maximum distance between
supply units is 30 m (Figure 7.3). If the dis-
tance between the supply units is more than
30 m, an additional row of supply units be-
tween these units is needed.
Figure 7.3. A rule of thumb of maximum dis-
tance between the supply units.
Exhausts in the ceiling
max 30 m
Low velocity units
on the floor level
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
58
Table 7.1. Air flow rates and covered floor areas with different nominal sizes of displacement units.
Nominal size [mm] Airflow rate per unit [m³/s[ Floor area per supply unit [m²]
100 Up to 0,030 10-15
125 0,020-0,030 10-20
160 0,030-0,080 10-30
200 0,070-0,150 10-50
250 0,100-0,200 15-60
315 0,200-0,400 20-70
400 0,250-0,500 30-100
500 0,400-0,800 40-150
630 0,600-1,300 50-170
If there are special requirements for interior
design, supply units can be recessed in the
structure. If needed, the supply units can be
covered with special decorative panels. In
this case the free supply area should be de-
signed to guarantee the normal perfor-
mance of the air supply units. The units can
be painted to meet particular decoration
needs.
The selected supply units may serve as a
visible architectural element of the interior
design, for example, by using free-standing
units installed on the floor within the space.
The use of free-standing units makes design
of uniform air distribution for large spaces
relatively simple.
7.2.3 Selection of supply unit
In practice, supply units will often be se-
lected with product related software or prod-
uct specific design graphs. Figure 7.4 shows
an example of product selection. With se-
lected air flow rate and supply air tempera-
ture, adjacent zone and possible draught risk
being the main selection criteria. The pres-
sure difference over the supply unit and the
sound pressure level (Lp) are also important
considerations in the design process.
7.2.4 Ducting systems
For ducting, there are three possible solu-
tions: ducting through the ceiling and walls;
Figure 7.4. Example of product data (adjacent zone, pressure drop and sound level) in displacement
unit selection.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
7. DESIGN OF DISPLACEMENT VENTILATION
59
ducting through the floor; and employing
an under-floor plenum. The exhaust of the
room air is located above the occupied zone
- preferably close to ceiling level.
Ducting through ceilings and walls
When ducting through the ceiling the typi-
cal location for supply units is close to the
walls where the units are freely mounted on
the floor or embedded in the wall structure.
Ducts are installed either visibly on the wall
or hidden inside the structure.
Air supply units can also be integrated in a
column, which creates an ideal ventilation
solution for the central part of large open
area. The supply units can be designed to
look like columns, which can suit the inte-
rior environment (Figure 7.5). The height
of the diffuser is selected based on the de-
sign concept.
Ducting through floor
When the units are connected to the duct-
work through the floor, it is possible to se-
lect the suitable location of the supply unit
quite flexibly.
The supply unit can be a visible element as
part of the interior design. The industrial
design of the unit could be specifically tai-
lored to meet the needs of the interior dec-
oration. Figure 7.6 shows an example of
visual displacement units that are ducted
through the floor.
Supply plenum
Air supplied through a under floor plenum
system is an excellent solution for applica-
tions such as concert halls and theatres
(Figure 7.7). With supply plenum system
typically, the height of the raised floor is
0,30–0,45 m. To reduce the air leakage and
noise generation of the supply units, the rel-
ative pressure is maintained between 10 Pa
and 30 Pa. With relatively low pressure
drop (<15 to 20 Pa), it is possible to meet
the high acoustic demands that are required
in concert halls.
Figure 7.5. Ceiling ducted displacement units
that are integrated with pillar structure.
Figure 7.6. Floor ducted displacement units
that are architectural visual elements.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
60
Figure 7.7. Supply plenum with under floor supply units in a concert hall.
In office applications, where standard
raised floor elements are used, as a rule-of-
thumb, the expected leakage of supply air is
10–30 % depending on the quality of the
structure. The maximum size of the under-
floor plenum can be about 300 m². The
maximum distance between air inlet and
the point of discharge is 15–18 m.
Due to heat transfer between supply air and
plenum structure, the supply air is warmed
in the plenum. This makes difficult to pro-
vide a fast response, controlled, room air
temperature.
7.3 Integration with separate heating and cooling systems
7.3.1 Integration with separate heat-
ing systems
Radiant panels
Using radiant panels is a good method for
room heating with displacement ventilation
(Figure 7.8). The panels should preferably
be located below the coldest elements in the
room, i.e. the windows and the outer walls.
The larger part of the heat emission is the
radiation. The minor part is convection,
which will counteract the cold down-
draught from windows and cold walls.
Convectors
Convectors go well with displacement ven-
tilation when the heat is distributed along
the cold walls/windows (Figure 7.9). One
concentrated convector may cause mixing
of the room air.
Figure 7.8. Radiant panel – well suited for dis-
placement ventilation.
Figure 7.9. The convector works well with dis-
placement when located below the cold walls or
windows.
Radiator
Convector
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
7. DESIGN OF DISPLACEMENT VENTILATION
61
Heating by ceiling panels
Heating by ceiling panels is very suitable for
displacement ventilation (Figure 7.10). In
normal conditions, without heating demand,
the ceiling is 3 K – 4 K warmer than the
floor, yielding a heat transfer from ceiling to
floor of about 20 W/m². Thus, a slight in-
crease in ceiling temperature will provide
sufficient room heating. The convection part
of the heat from the ceiling panels will coun-
teract the heat loss through the ceiling.
Heated ceiling panels stabilise the thermal
stratification, and thus benefit displacement
ventilation.
Floor heating
In normal conditions, floor heating is well
suited for displacement ventilation (Fig-
ure 7.11). Part of the heat transmission from
the floor is radiation towards the cold sur-
faces of the room. The convective heat trans-
fer will heat the supply air that spreads
across the floor.
If the floor is too warm, it will heat the air
and make it rise so that it causes mixing, at
least in the lower part of the room. How-
ever, practice has shown that with a floor
temperature below approximately 25 °C
and the supply air being some 2 K or cooler
than the room air, the supply air spreads
along the floor (Causone et al. 2010).
The “50 % rule” can no longer be used for the
rough calculation of the vertical air tempera-
ture gradient in a room when a radiant floor is
used together with displacement ventilation.
Figure 7.10. Heating by ceiling panels.
Figure 7.11. Floor heating with displacement
ventilation.
A new method of calculation is proposed,
using an “80 %-rule” as the limit, for a floor
heating capacity of about 60 W/m². In Fig-
ure 7.12, the effect of the specific floor heat-
ing capacity on the vertical room air stratifi-
cation is shown (Causone et al. 2010).
Figure 7.12. Correlation between the dimensionless room air temperature profile in the occupied
zone and the floor heating capacity (Causone et al. 2010).
Heating panels
Heated floor
1,0
0,8
0,6
0,4
0,2
0,00,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0 1,1
(Ty – Ts) = (Te – Ts)
y
H
Floor heating capacity: 60 W/m² 50%
65%
70%
75%
80%Floor heating capacity: 0 W/m²
Floor heating capacity: 35 W/m²
Floor heating capacity: 40 W/m²
Floor heating capacity: 50 W/m²
Dimensionless temperature profile in occupied zone and floor heating capacity
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
62
7.3.2 Integration with separate cool-
ing systems
The combination of chilled ceiling and dis-
placement has been proved to meet the ther-
mal comfort requirements of ISO7730
(2005) when designing for sedentary work-
ers in offices (Loveday et al. 2002, Hodder
et al. 1998, Alamdari et al. 1998). Also,
passive chilled beams can be integrated in a
similar way to displacement ventilation, but
that has not been studied extensively.
The combined system of a chilled ceiling,
displacement ventilation and desiccant de-
humidification is proposed for space condi-
tioning in hot and humid climates to im-
prove energy efficiency (Hao et al. 2007).
Choosing displacement ventilation as an air
distribution method does not by itself result
in a stratification strategy, if the whole
room air conditioning system is not de-
signed for that purpose. One example of
that is a system consisting of displacement
ventilation and chilled ceilings. Low veloc-
ity air supply and cooled ceiling systems
behave like mixing systems when the
cooled ceiling provides a substantial part of
the cooling (Tan et al. 1998 Rees and Haves
2013 and Schiavon et al. 2015).
In Figure 7.13 the vertical air temperature
distribution as a function of that de-
scribes the ratio of the cooled ceiling output
to the total cooling output, is presented
(Tan et al. 1998). The stratification effect
decreases gradually as the relative cooling
load of the ceiling, , increases. When is
less than about 0.6 there still is some ther-
mal stratification in the room. A similar
type of behaviour has also been found with
the contaminant stratification.
Figure 7.13. Vertical air temperature distribu-
tion in a room with chilled ceiling. Tempera-
tures relative to temperature at 0,1 metre
above the floor (Tan et al. 1998)
Figure 7.14 shows the contamination ratio
(contaminant in the occupied zone /ex-
haust) as a function of the relative cooling
load from the chilled ceiling (Krühne
1995). Figure 7.14 shows that when the
relative cooling of ceiling is increased the
contaminant in the occupied zone also in-
creases.
Figure 7.14. Contamination ratio in occupied
space versus cooling effect from ceiling panels.
Hei
gh
t ab
ov
efl
oo
r le
vel
,z
[m]
0,0
1,0
1,5
2,5
0,8 1,0 1,2 1,4
2,0
0,5
η = ratio of the cooled
ceiling cooling output
to the total cooling
output (Tan 1998)
Relative air temperature(relative to temp. at 0,1 m above the floor)
Chilled ceiling
η = 0,4
η = 0
η = 0,6
η = 0,5
0,0 0,2 0,4 0,6 0,8 1,0
0,0
0,2
0,4
0,6
0,8
1,0
1,2
/A = 12 W/m²
/A = 20 W/m²
/A = 30 W/m²
/A = 50 W/m²
/A = 65 - 93 W/m²
/A = 50 W/m²
Heat surplus per
unit floor area
Relative cooling load of cooled ceiling, Co
nta
min
atio
n r
atio
in
occ
up
ied
zo
ne,
c /
co
ze
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
7. DESIGN OF DISPLACEMENT VENTILATION
63
When chilled ceiling panels are used, the
supply air temperature should be kept con-
stant and the panels should be used to con-
trol the room temperature (Figure 7.15).
The best alternative is variable air volume
systems controlled by an air quality sensor,
and having the cooling panels control the
operative air temperature in the room.
Figure 7.15. Temperature control in a room
with chilled ceiling.
Radiant floor cooling works well with dis-
placement ventilation and it improves ther-
mal comfort in large spaces (Simmonds et
al. 2000). A radiant floor is an effective
sensible heat removal terminal due to its di-
rect absorption of solar radiation (Zhao et
al. 2016).
Figure 7.15 presents the air temperature
stratification in room when radiant floor
cooling and displacement ventilation is
used (Causone et al. 2010).
The radiant floor effects the temperature
gradient and higher vertical air temperature
differences are expected. The “50 %-rule”
is not valid in the occupied area. It is also
evident that increasing the air flow rate, and
thus raising the floor temperature, the ver-
tical air temperature differences decrease.
It should be noted that in many cases the
vertical air temperature differences be-
tween head and ankles could be higher than
the limits imposed by ISO standard due to
the action of the floor cooling. This phe-
nomenon is probably less important in
buildings with high solar loads. In these
buildings, lower vertical air temperature
differences occur, and thus the use of floor
cooling should not create any local thermal
discomfort (Causone et al. 2010).
Figure 7.16. Vertical air temperature profile at three locations (S1-S3) in the room of 16,8 m² with
airflow rate 0,050 m³/s and supply air temperature of 20 °C. Radiant floor cooling capacity is
31 W/m² (Causone et al. 2010).
Chilled ceiling controlsthe room temperature
Temperature sensor
Air qualitysensor
=Constantθs
2,5
2,0
1,5
1,0
0,5
0,020 21 22 23 24 25 26
Air temperature [°C]
Heig
ht
[m]
27 28 29
50%
30
S1
S2
S3
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
64
It is reported that radiant floor cooling does
not disturb convection flows of heat loads
and thus gives ideal additional cooling for
displacement applications (Babiak et al.
2009). In that case, as applied in the Bang-
kok airport terminal building, the surface
cooling system provides a maximum cool-
ing load of 70–80 W/m² when it works with
a constant supply water temperature of
13 °C and a return temperature of 19 °C.
7.4 Control of indoor conditions
Control of displacement ventilation does
not differ much from the control of mixing
ventilation. The main difference is the loca-
tion of the sensors for air quality and air
temperature.
The location of the temperature and air
quality sensors should be carefully consid-
ered. The location of the temperature/air
quality sensors depends on the height of the
room and on the ventilation system.
Constant Air Volume System
In some applications, supply air tempera-
ture and air flow rate have been kept con-
stant with acceptable results. In these cases,
the thermal mass of the building is utilized
and a slight increase in the room air temper-
ature is accepted during the occupied pe-
riod. In those applications, the occupied pe-
riod is quite short, for example a 1,5 to 3-
hour concert.
In a constant air volume system, the supply
air temperature should be controlled to give
a constant room air temperature at a certain
height in the occupied zone. In these cases,
the result depends on the height of the tem-
perature sensor, as illustrated in Fig-
ure 7.17. It should be noted that the supply
air temperature should not vary too much to
avoid creating draughts along the floor.
Figure 7.17. Idealised temperature distribu-
tion curves for varying heat loads and different
levels of the temperature sensors.
An example of a control curve for a venti-
lation system with cooling of the supply air
is shown in Figure 7.18. When the temper-
ature in the occupied zone, oz, exceeds a
certain maximum value, oz 2, the supply air
temperature is kept at its minimum value,
say 18 °C (depending on the performance
of the diffuser). When oz decreases below
a certain minimum value, oz 1, the supply
air temperature is kept at its maximum
value, say 20 °C.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
7. DESIGN OF DISPLACEMENT VENTILATION
65
Figure 7.18. Supply temperature, s, controlled by air temperature in the occupied zone, oz.
Variable Air Volume System
In the case of variable air flow rate, the air
volume flow is controlled according to the
air quality or according to the temperature
in the occupied zone.
To prevent cold draughts along the floor,
the temperature sensor for displacement
system should be located at 0,6 m above the
floor for rooms with wall-mounted or free-
standing diffusers and the air quality sensor
at the height of breathing zone of a sitting
person (1,1 m) (Figure 7.15) (Fitzner
2001).
In tall rooms the air quality sensor should
be located at the top of the occupied zone,
because this is the zone most prone to infe-
rior air quality (Figure 7.19).
Figure 7.19. Location of temperature and air
quality sensors in a room with larger ceiling
height.
Displacement ventilation is well suited for
variable air volume systems. When the air
flow rate through a low velocity diffuser is
reduced, the adjacent zone also decreases
(Figure 7.20).
Figure 7.20. Low velocity diffusers are well suited for variable air volume flow (VAV).
Temperature in the occupied
zone,
NB:
Supply temp > room temp
Not recommended!
22°C
min
(~18°C?)
18°C
max
(~20°C?)
(~21°C?)
(~23°C?)
s
s
oz 1 oz 2
oz
ozs
20°C 22°C
Supply air temperature, s
Occupied zone
Temperature sensor Air quality sensor
Reduced air volume flow = Reduced adjacent zone
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
66
When the dominant contamination source
are people, CO2 is a natural control param-
eter. The control logic can be illustrated by
the curve in Figure 7.21. The air quality
limits shown in Figure 7.21 are examples,
and must be chosen according to the target
values set in each specific case.
When there is no dominant or measurable
contaminant source, the temperature may
be the best controlling parameter (Fig-
ure 7.22).
Figure 7.21. Control logic for supply air flow
rate, controlled by air quality (for example,
CO2-concentration).
The same logic applies to tall rooms as to
rooms with normal ceiling heights. Addi-
tionally, the difference in temperature be-
tween the lowest and the highest level in the
occupied zone has to be taken into account.
When the vertical temperature difference
becomes too large, increasing the ventila-
tion rate can reduce it.
Figure 7.22. Control logic for supply air flow
rate, controlled by the temperature in the occu-
pied zone.
Ventilation rate, q
Max
Min
CO -
concentrationMin
(~ 600 ppm)
Max
(~ 1000 ppm)
s
2
Max
Min
Room air temperature,
Min
(~ 20°C)
Max
(~ 23°C?)
NB: location of the sensor
Ventilation rate, qs
oz
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
67
8 Case studies
8.1 Air distribution with four typical air supply methods in a classroom
The performance of four typical air distribu-
tion methods in winter and summer condi-
tions with different occupancy ratio was
studied by physical measurements and
smoke visualization in a mock-up classroom
(Kosonen and Mustakallio 2010) and by
CFD- simulations (Mustakallio and Koso-
nen 2011).
The measured mock-up room (6,0 m x
4,4 m x 3,3 m (H)) was half of an actual
classroom (floor area 6 m x 10 m). The
mock-up chamber was located inside an-
other chamber. During the test, the sur-
rounding temperature was controlled
based on demand to reach the room air
temperature setpoint. The simulated win-
dow size was 4,4 m x 1,4 m (H). The room
air distribution was identified at three dif-
ferent load conditions: summer conditions
with maximum occupancy (cooling load
of 54 W/m²) and partly occupied (cooling
load of 40 W/m²) and winter conditions
with partly occupied room (heating de-
mand of 38 W/m²). The room was venti-
lated at 0,006 m³/s per person in all cases.
In the winter condition, a radiator was in-
troduced underneath the window to pre-
vent the risk of draught due to downward
buoyancy flow from the cold surface of the
window. The heat balance and breakdown
of the loads in the measurement cases are
presented in Table 8.1. Utilizing dynamic
energy simulations, room air temperatures
in winter and summer were set to corre-
spond to average conditions in Scandina-
vian classrooms. During the full-scale
room measurements heat loads were offset
by heat transfer through the structures to
attain the room air temperature required.
Table 8.1. Heat balance and the breakdown of the loads in the mock-up classroom section.
Heat loads and heat losses of the simulated
classroom (half size of the actual classroom)
Summer Full Occupancy
Summer
Half- Occupancy
Winter
Half-Occupancy
Room air temperature 26 °C 24 °C 21 °C
Occupants - 58 W/person (total heat load) 15 (870 W) 7 (406 W) 7 (406 W)
Lighting 15 W/m² 360 W 360 W 360 W
Solar load or heat loss from window
(surface temperature of window)
197 W
(30 °C)
296 W
(30 °C)
-448 W
(11 °C)
Power of a radiator underneath window 0 W 0 W 250 W
Total heat loads 1427 W 1062 W 1016 W
Supply airflow rate 0,090 m³/s
(supply temperature)
-972 W
(17 °C)
-756 W
(17 °C)
-324 W
(18 °C)
Heat loss through structures -455 W -306 W -244 W
Total heat losses -1427 W -1062W -1016 W
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
68
The performance of four typical air distri-
bution methods was studied: a corridor-
wall grille, a ceiling diffuser in the middle
of the ceiling, a perforated duct diffuser in
the middle of the ceiling, and two displace-
ment ventilation units in the floor corners
(Figure 8.1).
The supply units were selected based on the
throw pattern analysis. The supply airflow
rate was 0,090 m³/s (0,006 m³/s per person)
in all cases (half classroom). The supply air
temperatures were 17 °C and 18 °C in sum-
mer and winter cases respectively. The room
air temperatures were 26 °C and 24 °C in
summer case with full and half occupancy
respectively. In winter conditions, the room
air temperature was set to be 21 °C.
Figure 8.1. Air distribution schemes: A) Wall
grille, B) Displacement ventilation, C) Multi-
nozzle ceiling diffuser and D) Perforated duct
diffuser.
Air velocity and temperatures were meas-
ured at 24 pole locations and at 7 heights
(0,1, 0,5, 0,9, 1,3. 1,8, 2,4 and 3,1 m above
the floor) at each location. a total of 168
points. The dimensioned simulated class-
rooms are shown in Figure 8.2.
Smoke and CFD visualizations of air distri-
bution in the fully occupied summer cases
are shown in Figure 8.3. In summer condi-
tions, thermal plumes did not have a signif-
icant effect of the performance of a wall-
grille: the momentum flux of a wall-grille
was strong. With a wall grille, the jet
reaches the other side of room. Also, air is
spread effectively over the whole occupied
zone with the low velocity units, whereas in
summer time air supplied from the ceiling
diffuser tends to be carried along thermal
plumes from heat sources. A perforated
duct diffuser had a tendency to create un-
stable flow conditions and varied loads can
unexpectedly change the throw pattern.
High velocities (over 0,3 m/s) were meas-
ured in the occupied zone in all condi-
tions with a wall-grille. The highest ve-
locities (above 0,2 m/s) were measured
near the window (at a distance of 0,25 m).
In all conditions velocities higher than
0,2 m/s were also measured near the
floor, and at 0,1 m height, as far as 3,6 m
from the window.
The displacement ventilation concept was
not sensitive to load variation and air veloc-
ities were low (< 0,15 m/s) except at meas-
urement points close to the corner-installed
supply unit. With a ceiling diffuser, air ve-
locities were, in all cases, between 0,19 and
0,23 m/s. With a perforated duct diffuser,
relatively high velocity (0,15 – 0,2 m/s)
was measured near the floor (at 0,1 m
height). In the two summer conditions the
velocity was above 0,2 m/s (up to 0,31 m/s
with full occupancy) close to the floor for
locations 3 m, and 4,8 m from the window,
i.e. the influence of heat load increased air
velocities. This indicates more unstable
performance with a perforated duct diffuser
when higher heat loads are introduced in
the classroom.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
8. CASE STUDIES
69
Figure 8.2. The geometry of the measured half (left) and simulated classroom (right).
Figure 8.3. Visualizations in a half classroom of air distribution using smoke and CFD in cooling
mode with full occupancy. Supply air units: A) a wall-grille, B) displacement ventilation with low
velocity units, C) a ceiling diffuser and D) a perforated duct diffuser.
The air distribution with the corridor wall-
grille gave high velocities in all load con-
ditions. In winter conditions, air velocities
were particularly increased close to the
window.
In principle, the throw length could be op-
timized for winter conditions and so de-
liver more moderate velocities at work-
spaces close to the window (for example
by selecting a larger wall-grille). How-
ever, this increases the risk of draught in
summer conditions.
The supply air jet from the ceiling diffuser
tended to be carried along with thermal
plumes from the heat loads during summer.
In winter when there was no window plume
effect and so the air distribution was more
uniform. Ceiling diffusers can provide an ap-
propriate solution in varied load conditions.
With the perforated duct diffuser, the perfor-
mance was quite unstable and sensitive to
the subsequent removal of high heat loads.
In such conditions, the supply air could un-
expectedly drop causing an increased risk of
draught in certain work spaces.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
70
When applying mixing ventilation, the
types of heat loads have a significant effect
on air distribution. Therefore, when such an
air distribution strategy is designed, the sys-
tem performance should be analysed under
different conditions. In the design phase, it
is not possible to analyse the interaction of
convection flows and jets without using
CFD- simulation or laboratory mock-ups.
Displacement ventilation was least sensi-
tive for different heat load conditions of all
studied concepts. Using a ceiling diffuser,
air velocities were reasonable low in all
cases. A wall grille gave high velocities in
both summer and winter conditions. With a
perforated duct diffuser, air distribution
was quite unstable causing increased
draught risk in some load conditions. The
performance of the wall-grille and the per-
forated duct diffuser were particularly sen-
sitive to the strength and location of heat
loads.
8.2 Comparison of calculated and measured vertical temperature gradients for displacement air distribution
During the full-scale experiments some
typical convection flow elements as well as
combinations of flow elements were meas-
ured (Kosonen et al. 2016). The internal
heat loads consisted of heated cylinders
representing people, heated cube-shaped
boxes representing computers, fluorescent
ceiling lighting units, heated foil panels on
the wall representing window solar load,
heated foil panels on the floor representing
direct solar load, heated foil panels in the
ceiling representing skylight solar load.
The test setup consisted of displacement
diffusers and ceiling exhaust in a well-insu-
lated room (employing 100 mm polysty-
rene) with 20,7 m² (4,6 m x 4,6 m) floor
area and room heights of 5,2 m and 3,3 m
Figure 8.4).
In Chapter 5, the principles of temperature
based design models were described. In this
chapter, three of the models are applied and
compared with measurements. Those mod-
els are: Mundt (1996), Nielsen (1995 and
2003) and Mateus and da Graça (2015).
Figure 8.4. Test facility.
The measured and calculated vertical pro-
files of room air temperature are compared
for a space with 12 occupants and one with
a warm window (Figure 8.5). Figure 8.6,
shows a comparison of measured and cal-
culated vertical temperature profiles for
combinations of heat load from occupants,
warm window and floor.
The vertical air temperature gradient with a
warm window and ceiling gains are quite
linear whereas temperature profiles are far
from linear with the heat provided by the
computer, people and a warm floor. With
those heat loads, the major part of the ver-
tical air temperature gradient occurs across
the occupied zone.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
8. CASE STUDIES
71
Figure 8.5. Measured and calculated room air
temperature profiles a) occupants and b) warm
window in a 5,2 m high test room.
The agreement between prediction and ex-
periment of the three-node model is quite
good when the heat loads that are located in
the low level of the occupied zone, such as
computers, people and the warm floor.
When the heat loads are located at high
level (warm ceiling) or a linear type (warm
window), the modelled results did not cor-
relate well with the measurements.
Mateus and Da Graça’s model works fine
with occupants, computers and floor heat
loads. However, accuracy with the warm
window and warm ceiling were not so good.
The linear two node models (Mundt) works
well with a warm window and ceiling, but
the linear model cannot accurately describe
heat loads that exist in the occupied zone.
The three-node model (Nielsen) predicts
different slopes for the temperature profile
between the nodes and gives better accu-
racy than the linear two node models.
Figure 8.6. Measured and calculated room air
temperature profiles with the combinations of
heat loads in a 3,3 m high test room.
0
1
2
3
4
5
18,5 19,5 20,5 21,5 22,5 23,5
Hei
gh
t (m
)
Temperature (°C)Measured
Mundt
Nielsen
Mateus and da Graça
Φ12occupants= 900 W, θs = 18,7°C, qs = 0,15 m³/s
a)
0
1
2
3
4
5
19,5 20,0 20,5 21,0 21,5
Hei
gh
t (m
)
Temperature (°C)Measured
Mundt
Mateus and da Graça
Nielsen
Φwindow_3.5 = 520 W, θs = 19,1°C, qv = 0,15 m³/s
b)
0
0,5
1
1,5
2
2,5
3
3,5
20,0 22,0 24,0 26,0 28,0
Hei
gh
t (m
)
Temperature (°C)Measured
Mundt
Mateus and da Graça
Nielsen
Φ10occupants = 750 W, Φfloor= 260 W,
Φlight= 232 W, Φwin_3.5m= 520 W,
θs = 18,1 °C, qv = 0,1 m³/s
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
72
The accuracy of the three-node model is
quite good when applied with heat loads that
are located in the low level of the occupied
zone, like computers, people and warm
floor. All in all, the three nodal models are
quite useful in engineering calculation
where the supply air flow rate of displace-
ment ventilation is defined and energy con-
sumption of the whole building is calculated.
8.3 Field measurements for a multipurpose arena
8.3.1 Air distribution concept
This case study presents the use of displace-
ment ventilation in Malmö-arena in which the
seating capacity was up to 13 000 individuals.
The dimensions were 100 m (L) x 90 m (W)
x 30 m (H) (Figure 8.7). The arena is de-
signed to offer a wide range of entertainment
events for people ranging from a hockey
game to various entertainment events. The
arena comprises an ice rink, seating area, en-
closed suites and surrounding service areas
for example, restaurants and shops.
Physical measurements were taken in the
seating sectors, the ice rink and in the centre
of the space volume. In addition, computa-
tional fluid dynamics (CFD) simulations
were undertaken to provide a generic view
of air distribution (Lestinen et al. 2012).
Displacement ventilation was employed for
the lower-seating area and zoned ventilation
used for the upper-seating area. The dis-
placement supply air was distributed from
under the retractable stands (on movable
stands) beside the ice rink. The supply air
flowed through openings below the seats.
The exhaust air was taken at the ceiling
level (Figure 8.8). The overall ventilation
system contained four air-handling units
(301-304) and two air-recirculating units
(305-306) that were operating during
events providing up to 70 m³/s airflow rate.
At the upper level high momentum down-
ward jet flows enhance an effective mixing
of indoor air.
The smoke tests indicated an upward flow
along the lower-seating area and a down-
ward flow along the upper-seating area
from where the air was moving into the
middle of the space volume.
Figure 8.7. Malmö arena.
Figure 8.8. Ventilation system in arena.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
8. CASE STUDIES
73
8.3.1 Measurements and CFD- simu-
lations
The room air sensors were installed onto
measuring masts at heights of 1,5 m,
2,5 m and 3,5 m over the seating-row and
the variables were recorded over a 3 min
average. (Lestinen et al. 2016).
CFD-simulations were used to investigate
the flow patterns in the arena enclosure.
The grid sizes were between 0,1 m and
0,6 m for the whole arena model. In the
arena model, the total number of grid
nodes was 19,68x106 with 102,78x106 ele-
ments.
8.3.2 Performance during an ice
hockey game
During the measurement period, there were
4 000 spectators. The experiments indi-
cated that the arena indoor air temperature
increased about 2 K during the game when
the airflow rate increased up to near 70 m³/s
when CO2 reached 900 ppm with a 15 °C –
16 °C supply air temperature.
The room air temperature during the game
was 12 °C – 17 °C at the lower-seating area
and 15 °C – 17 °C at the upper-seating area.
Experiments show the relatively low tem-
perature stratification (less than 2 K) and
the well-mixed conditions in the arena
(Figure 8.9). The corresponding air speeds
were below 0,35 m/s in the seating areas.
Figure 8.9. The simulated room air temperatures in three cross-sections (panel a). The measured
and predicted room air temperature profiles in the centre of the seating area in a hockey game
(panel b). The scheme of air movement in the arena (panel c).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
74
9 Research findings
9.1 A CFD Benchmark test for manikins in displacement flow
It may be important to test and adjust Com-
putational Fluid Dynamics (CFD) software
by comparing results with a benchmark
test. The benchmark test can be used as a
test of a new program, but can also be used
for the development of a virtual person to
be situated in a ventilated room. This chap-
ter shows a benchmark test that considers
the three-dimensional flow around a person
in displacement ventilation
This benchmark test is defined on the web-
page: www.cfd-benchmarks.com
For many years thermal manikins have
been used in full-scale indoor environment
experiments. CFD provides an alternative
to full-scale measurements. Research cen-
tres around the world have therefore devel-
oped different configurations (subroutines)
to represent a Computer Simulated Person
(CSP).
The CSPs can be very different in respect
to size, form (employing a rectangular grid
or body-fitted grid), heat emission details,
etc. The variations may reflect the different
software possibilities, but may also be de-
termined by different standards from coun-
try to country. Some examples of predic-
tions made with CSPs are given by Mura-
kami et al. 1995), Brohus and Nielsen
(1996b) and Topp et al. (2002).
The idea behind a benchmark test is to com-
pare individual concepts under the same
boundary conditions. The tests of different
manikins may also improve the design of a
CFD manikin and so lead to new standards.
The CSP in the displacement ventilation
benchmark is standing, facing the diffuser
in a displacement ventilated room, as illus-
trated in Figure 9.1. The ceiling, floor, side
wall and end wall should be simulated as
solid surfaces. Full details are given on the
web page, www.cfd-benchmarks.com or in
Nielsen et al. (2003).
The thermal manikin benchmark tests have,
as at 2016, been used in more than 30 dif-
ferent papers, theses and articles.
Figure 9.1. A person exposed to a flow field in
a displacement ventilated room.
9.2 Full-scale tests and CFD- simulations of indoor climate conditions
Full-scale tests and CFD-simulations have
been commonly used to ensure that indoor
climate conditions meet the displacement
ventilation system design criteria.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
75
Figure 9.2. Illustration of 5,13 m high test room with objects: A) 2 supply air diffusers used in tests,
B) exhaust, C) 4 heated foil panels (window), D) 10 heated cylinders (occupants), E) lighting units
and P1-P3) locations of vertical temperature gradient measurement. Ceiling height of 3,29 m test
room with two heated foil panel windows marked.
The required cooling power is determined
so as to maintain the occupied zone of a
room at design conditions. CFD-simula-
tions are especially useful when designing
ventilation for large and complex rooms. It
is essential to validate the accuracy of
widely used CFD-modelling methods
against full-scale test measurements
(Deevy et al. 2008). This example presents
indoor climate predictions with CFD and a
full-scale test for a test room with two room
heights. Different CFD models are used
(Mustakallio et al. 2012).
9.2.1 Full scale test
The test setup consisted of two displace-
ment diffusers, with perforated front faces,
and ceiling exhaust in a well-insulated
room with 20,8 m² floor area. A room with
two ceiling heights 5,13 m and 3,29 m was
studied. The internal heat loads consisted of
10 heated cylinders representing people,
heated foil panels on one wall representing
solar load from a window surface, and flu-
orescent lighting units. Full-scale tests were
undertaken in steady state conditions. The
full-scale test setup is shown in the illustra-
tion in Figure 9.2 and the photograph in
Figure 9.3.
Vertical temperature profiles were measured
at three locations (P1-P3 in Figure 9.2) at
eight heights. The inner wall surface, ambi-
ent air, supply air and exhaust air tempera-
tures were measured. Supply and exhaust air
flow rate measurements were undertaken.
Supply and exhaust air flow rates were bal-
anced for each measurement. The ductwork
in the test room was insulated and all sur-
faces of the test room were covered with
0,1m polystyrene boards and plastic foil to
minimise the effect of surrounding condi-
tions on the vertical temperature stratifica-
tion. However, heat flux through the walls
affected the measurements especially in the
case of the tests with the higher ceiling. This
was noted by measuring the total amount of
electrical power supplied for the internal
heat loads, the supply air flow and tempera-
ture, and comparing the calculated exhaust
air temperature and the measured exhaust air
temperature. The obtained heat flux can be
defined for the room surfaces during the
CFD-simulations.
Part of the difference between calculated
and measured values was also caused by
measurement inaccuracy, but this effect was
assumed to be small and neglected when de-
fining the corresponding CFD model.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
76
Figure 9.3. Photos from the full-scale test rooms. Rectangular heated boxes on the tables simulate
computers, floor standing and heated vertical cylinders simulate occupants. The middle of the three
floor standing displacement diffusers seen on the left and right photos was not used.
9.2.2 CFD-simulation
CFD-simulations were carried out with
Ansys CFX 14.0.A grid comprising of
1,9/1,6 million unstructured elements or
430/370 thousand nodes (room height of
5,13 m/room height of 3,29 m) was used.
Inflation layers were used near surfaces.
Grid independency was tested by dou-
bling grid size. A SST turbulence model
with automatic wall treatment was used
for the simulated cases, but for compari-
son same cases were calculated with
standard k-e and RNG k-e turbulence
models.
Buoyancy was modelled with Boussinesq
approximation and compared with ideal
gas model predictions. Cases were solved
with a high (2nd) order discretization
scheme, except for the turbulence equation
which was solved with a 1st order scheme.
The effect of additionally having a high or-
der scheme for turbulence and also includ-
ing buoyancy turbulence source terms
were compared to the initial case. Radia-
tion was modelled with a discrete transfer
model. CFD-simulations were solved as a
steady state case so as to reach good con-
vergence.
The supply air inlet was specified as the
whole area of the displacement diffuser
with an additional momentum source in the
front of perforated front face to account for
the effect of the free area of the perforated
plate. Realistic CFD models of displace-
ment diffusers are needed to evaluate ther-
mal comfort in the near zone of the diffuser.
9.2.3 Results
The vertical temperature distribution was
measured and compared to the CFD-simu-
lation results. The measured temperature
distribution in three locations is presented
in Figure 9.4. The distribution was nearly
the same in all locations. The first and last
readings in Figure 9.4 show the supply and
exhaust air temperature.
Comparison of CFD simulation results with
different turbulence models (Figure 9.5)
showed a significant difference to the
measured temperature in the occupied
zone. The difference was about 1 K to
1,5 K at 1 m to 1,5 m height. The effect of
changing the buoyancy model from Bous-
sinesq to the ideal gas model was negligi-
ble, as also was the effect of high order dis-
cretization for turbulence.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
77
Figure 9.4. Measured vertical temperature distribution in A) 5,13 m and B) 3,29 m high test rooms.
Figure 9.5. Comparison of CFD-results and measurement in A) 5,13 m room with k-e, RNG k-e and
SST turbulence models, with Boussinesq (Bsqm) and with ideal gas models for buoyancy (Igm), and
with higher order turbulence discretization scheme (TurbHdsch), and B) 3,29 m room with k-e, RNG
k-e and SST models.
The predicted vertical temperature stratifi-
cation of the CFD-simulations had similar
form to the measured temperature stratifi-
cation. The SST turbulence model made
the best prediction of the temperature dis-
tribution. The temperature in the lower
part of the room was still significantly
lower in the CFD-simulation than in the
full-scale test for both test room heights.
This could be partly corrected by using
displacement diffuser CFD models that
improve the mix of supply air with the
room air in the occupied zone.
9.3 Test on the performance of displacement ventilation– proper simulation of occupants
The design of displacement ventilation of-
ten includes one or more of the following
elements: full scale laboratory tests, field
measurements and CFD simulations. Be-
cause displacement air distribution is a re-
sult of the thermal flows in spaces, it is im-
portant to properly simulate the heat
sources. Occupants are important heat
sources in rooms. With development of low
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
78
powered office equipment and lighting,
windows with controlled surface tempera-
tures, etc. occupants will produce major
part of the heat load in rooms. Therefore,
the buoyancy flow, i.e. the thermal plume,
generated by the human body is important.
Simulators of occupants with simple geom-
etry, such as a cylinder and a rectangular
box, generate significantly more concen-
trated plumes compared to simulators with
a complex body shape, such as thermal
manikins. As a result, the volume flux in
the thermal plumes above a cylinder and
above a thermal manikin can differ by 40 %
(Zukowska et al. 2012b). The differences
may affect the air distribution in rooms with
displacement air distribution. However, it is
easier and cheaper to simulate occupants
with simplified geometries. The question
rises “how far it is possible to simplify the
geometry of the human body” without af-
fecting the displacement air distribution
and validity of the obtained results.
The graph on Figure 9.6 presents normal-
ized vertical distribution of CO2 concentra-
tion based on measurements performed in a
full-scale test room (4,7 m x 5,4 m x 2,6 m)
with displacement ventilation with a supply
air temperature 21,6 °C and a total flow rate
0,080 m³/s (Zukowska et al. 2008). Two oc-
cupants seated at two identical workstations
were simulated first by two thermal mani-
kins accurately resembling human body
shape and then by two heated cylinders (as
shown in the photograph in Figure 9.6).
The manikins and the cylinders had the
same surface area of 1,63 m² and the same
heat generation of 73 W. CO2 supplied
from the top of their “head” was used for
simulating human body bio-effluents. CO2
concentration was measured at 16 heights
in 9 locations and at 20 points in a horizon-
tal plane 0,2 m below the ceiling.
The values of the CO2 concentration ratio
above 100 % indicate that the concentration
at some points of the room was higher than
in the exhaust air (Figure 9.6). It is clearly
seen that for the lower heights the concen-
tration ratio values obtained with the ther-
mal manikins are larger than the values ob-
tained with the cylinders. The complex ge-
ometry of the manikin body leads to more
intense air mixing around, and pollution
diffusion from, the upper zone to the lower
zone. In the case of the cylinders as occu-
pant simulators, the concentration profile in
the lower zone is steeper.
In Figure 9.6, the concentration ratio equal
to 50 %, i.e. between the lower and upper
zones, is used to show the difference in the
vertical CO2 distribution in the room be-
tween the cases with thermal manikins and
with cylinders. The height obtained for the
case using manikins is approximately 1,3 m
and for the cylinders it is 1,7 m. This is be-
cause the manikin generates a thermal
plume with volume flux greater than the
cylinder, and therefore equals the supply
airflow rate at a lower height than in the
case with the cylinder.
Thermal flows from the manikins cause
more mixing in the lower zone and there-
fore more pollution is diffused from the up-
per zone to the lower zone. The different
shapes of the occupant simulators cause
different contaminations distributions in
the upper zone – higher CO2 concentration
ratios for the cylinders (Figure 9.6).
Thus, simulation of occupants by objects
with simplified geometry, such as cylin-
ders, is insufficient for obtaining accurate
results when studying airflow in rooms
with displacement ventilation. Simulation
of the complex shape of the human body is
highly recommended. However, use of
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
79
thermal manikins during physical measure-
ments is expensive. Implementing and ap-
plying complex body shapes during CFD
predictions is rather time consuming. Fig-
ure 9.7 shows a heated dummy with sim-
plified geometry but with the same body
surface area and heat production as the ther-
mal manikin also shown in the figure.
Comprehensive measurements reveal that
the two human body simulators (thermal
manikin and heated dummy) generate ther-
mal plumes with similar characteristics
(Zukowska et al. 2012b). Thus, the heated
dummy can be used to obtain reliable re-
sults in case of displacement air distribu-
tion. However, it should be made clear that
heated dummy is not appropriate to study
the airflow in the vicinity of human body.
In this case a thermal manikin with a com-
plex body shape is to be used. Small non-
uniformity in the velocity field
(±0,005 m/s) and in the temperature field
(±0,02 K) of the surrounding environment
affects the development of the thermal
plume above a sitting person and causes
skewness of the plume cross-section (Zu-
kowska et al. 2010a).
Figure 9.6. Vertical CO2 concentration distribution with manikins and with cylinders. C, Cs and Ce
are respectively CO2 concentration at the measured point, at the supply and at the exhaust.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
80
Figure 9.7. Heated dummy with simplified body shape and thermal manikin with complex body
shape generate thermal plums with similar characteristics.
9.4 Airborne cross infection risk in a room with displacement ventilation
We often address the ventilation of a room
at a macro scale level. Macro scale is the tra-
ditional level of description of air distribu-
tion in rooms as, for example, in standards
where it is expressed that contaminant re-
moval effectiveness in a room with displace-
ment flow should have the level of c ~ 1,2.
A person in a stratified flow is assumed to
be exposed to the same level of contami-
nation from another person independent of
the position in the room so long as the peo-
ple are not too close to each other and are
standing with their faces at the same
height, Figure 9.8A. The distance should
be larger than 1,2 m in case of breathing,
Nielsen et al. (2008).
When the people are close to each other,
< 1,2 m, the exposure can rise to a high
level independent of the general contami-
nant level of the occupied zone. In other
words, if the air distribution system is de-
signed to make an efficient ventilation of
the room, there will still be a microenvi-
ronment around people close to each other,
which can’t be influenced efficiently by
the general air distribution system, see
Figure 9.8B.
Figure 9.8. A) Displacement flow, same con-
centration everywhere along the breathing
height. B) Illustration of the microenvironment
with a local high exposure.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
81
The microenvironment around two people
is illustrated in Figure 9.9. The exhalation
from the source person can be divided in
two parts. One part of the exhalation flows
into the macroenvironment, and the gen-
eral room air distribution system dilutes,
stratifies and transports this part out of the
room and creates a concentration distribu-
tion around the target person, coz. The
other part of the exhalation from the
source person flows directly to the target
person’s breathing zone, or to this person’s
thermal boundary layer.
Figure 9.9. A source manikin (right) and a tar-
get manikin (left). The contaminant flow be-
tween the two manikins is indicated by smoke.
The target person is exposed to a level of
cexp, and this exposure therefore consists of
an indirect exposure from the macroenvi-
ronment, coz, and a direct exposure from the
source person’s exhalation. The concentra-
tion coz can be measured direct at the stand-
ing target person’s chest. This concentra-
tion is also the target person’s inhalation
concentration if this person is not influ-
enced by a direct exposure, because the in-
halation normally originates from the ther-
mal boundary layer (Brohus and Nielsen
1996a, Bjørn and Nielsen 2002).
The concentrations cexp and coz can be given
in a dimensionless form when the concen-
tration is divided by the return flow concen-
tration.
A personal exposure index εexp can be de-
fined as (Brohus and Nielsen 1996a, Mundt
et al. 2004): εexp = ce /cexp . The exposure is
alternatively defined as cexp /ce.
9.4.1 Exposure in stratified flow at
the macro scale level
The contaminant distribution in a room
with displacement ventilation may, in some
situations, be stratified into layers through
the room. The air exhaled from a person
may, for example, be concentrated in a
layer. Figure 9.10 shows full scale experi-
ments with stratification of exhalation in a
hospital ward. The two illustrated situations
are where the source patient is lying on
his/her back and then on his/her side.
Figure 9.10. Two patients in a hospital ward.
A) The source patient is lying on their back
and the exhalation is flowing to the upper zone
resulting in high contaminant removal effec-
tiveness. B) The source patient is lying on their
side and the exhalation is stratified in a layer
in the breathing height of the target patient
with a high concentration, even across the full
width of the ward.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
82
Figure 9.10 shows how the exhaled air
from a person can either pass to the upper
zone with high contaminant removal effec-
tiveness (εexp ~ 70) when the source patient
is lying on their back or the exhaled air can
stratify in the ward in the inhalation height
with a very low contaminant removal effec-
tiveness (εexp ~ 0,7) when the source patient
is lying on their side (Qian et al. 2006).
The increased cross infection risk that can
take place in the macro environment in a dis-
placement ventilated room due to stratifica-
tion of the exhalation is a serious problem in
rooms occupied with people having air
borne disease. Displacement ventilation can
therefore not be recommended in this case
(Bjørn and Nielsen. 2002 and Li et al. 2011).
9.4.2 Exposure in stratified flow in
the microenvironment
Figure 9.11 shows measurements in the mi-
croenvironment between two people (sim-
ulated with thermal manikins) who are
standing in four different positions,
namely: face to face, face to back, face to
side and sitting source person. The expo-
sure of the target person to the air exhaled
by the source person is given as cexp/ce,
where cexp is the concentration in the target
manikin’s inhalation, and ce is the concen-
tration at the exhaust. When the distance
between the manikins is 110 cm, the target
manikin inhales a concentration which is
equal to the background concentration in
the room. The two people do not have a
common microenvironment with respect to
cross-infection considerations. The concen-
tration cexp/ce is ~0,5 for face to the side and
face to the back, which is typical of dis-
placement ventilation where the inhalation
contains air from the lower zone in the
room (Brohus and Nielsen 1996b). cexp/ce is
~1,0 for the face to face situation, and the
higher value indicates that a small fraction
of direct exposure takes place at a distance
of 110 cm.
Figure 9.11. Experiments with cross infection risk
in a room with displacement ventilation. A) Illus-
tration of the four positions of the two people; B)
Exposure versus distance between people.
There is a remarkable increase in the direct
exposure when the distance between the
people is less than 80 cm for the cases face
to face and face to side of the target person.
The exposure increases up to 12 times the
concentration in a fully mixed situation, in
the face to face situation, and up to 7 times
in the face to the side of the target person
situation, when the distance is 35 cm. With
respect to the protection against cross-in-
fection this is a serious setback for systems
generating a vertical temperature gradient
(Nielsen et al. 2012 and Olmedo et al.
2012).
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
83
Protection from cross-infection seems to be
high in the face to back situation. The ex-
posure cexp/ce does only reach 0,75 at a dis-
tance of 35 cm, which is below a fully
mixed case. It should be noted that the peo-
ple have the same height. A difference in
height with a tall target person may give a
larger exposure.
Many parameters may influence the cross-
infection risk between people situated close
to each other in a ventilated room and they
can be summarized in the following:
Distances between the people, positions and
orientations of the people, breathing process
(breathing through the mouth or through the
nose, opening of mouth, coughing, speak-
ing), difference in the height of the people,
activity levels of the people, number of peo-
ple, temperature and vertical temperature
gradient in the microenvironment around the
people, air velocity (speed and direction) in
the microenvironment around the people,
and turbulence level of the air flow in the mi-
croenvironment around the people.
Figure 9.11 gives the results for displace-
ment ventilation in the room when the total
heat load is 500 W and the air change rate
is 5,6 h-1. The exhalation frequency is 19
exh/min for the source person and 15,5 for
the target person. The exhaled flow is
11 L/min for the source person and
10 L/min for the target person. The source
person produces a total heat realise of 94 W
and the target person 102 W.
9.5 Displacement ventilation design based on occupants’ response
In rooms with displacement air distribution
thermal comfort concerns are focused on
draught at the feet/lower leg and discomfort
due to vertical temperature difference be-
tween neck level (1,1 m height) and ankle
level (0,1 m height above floor). Both air
velocity (highest near the floor) and air
temperature (lowest near the floor) are im-
portant for avoiding draught discomfort and
they are a function of the supply air temper-
ature and flow rate. At fixed heat load the
same room air temperature (1,1 m above
the floor) can be obtained by different com-
binations of supply air temperature and
flow rate, i.e. small temperature difference
between the room air and supply air and
high supply flow rate or vice versa.
A small temperature difference between
supply and room air temperature and high
supply flow rate will lead to elevated veloc-
ity at floor level and thus increased risk of
draught (draught risk is discussed in Chapter
3). However, this will reduce the risk of local
thermal discomfort due to vertical tempera-
ture difference, because the vertical temper-
ature difference will be small. Large temper-
ature difference and small flow rate will lead
to reduction of the velocity but will increase
the vertical temperature difference.
Perceived air quality must also be consid-
ered during design because it is directly re-
lated to room air temperature, supply air
temperature and flow rate. Inhaling warm
and polluted air will negatively impact per-
ceived air quality (ASHRAE Guide 11,
2011, see also Chapter 3). In rooms with dis-
placement ventilation an increase of the sup-
ply flow rate will move the stratification
height to a higher level which is expected to
improve perceived air quality (inhaled air
will be cleaner and cooler) and vice versa in
the case when the flow rate is decreased but
the temperature difference is increased.
Thus, different approaches in the design of
displacement ventilation can be adopted
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
84
with advantages and disadvantages. With-
out considering the energy implications,
the questions are: What is more beneficial
for occupants, high supply flow rate and
small difference between room tempera-
ture and supply air temperature or low sup-
ply flow rate and high temperature differ-
ence? How does the selection of these two
parameters depend on the room air temper-
ature? This has been studied by Dalewski
et al. (2014) and is briefly explained in the
following.
Design details:
• full-scale room (3,6 x 4,8 x 2,6 m³) fur-
nished to simulate an office with two
workstations (Fig. 9.12);
• each workstation consisted of a desk, an
adjustable chair, a desk lamp (20 W) and
a laptop PC (50 W);
• one semi-circular floor standing air sup-
ply diffuser for displacement ventilation
(DV);
• one ceiling air supply diffuser for mixing
ventilation (MV);
• one ceiling exhaust diffuser;
• room air temperature maintained at 1,1 m
height;
• operation modes: five combinations of
room air temperature, difference be-
tween room temperature and supply air
temperature (ΔTs) and supply flow rate
were applied in the case of displacement
ventilation and one combination of these
parameters was studied in the case of
mixing ventilation and used for compar-
ison (Table 9.1):
• Thirty-two subjects were exposed to each
of the six conditions randomly; two sub-
jects at the time seated at the workstations
performing office work on computer;
• Subjects responded to questionnaires on
thermal comfort, perceived air quality, air
movement sensation, etc.
Figure 9.12. Test room set-up.
Table 9.1. Operating modes (conditions): DV – displacement ventilation; MV – mixing ventilation.
Condition Room air
temperature at 1,1 m
[°C]
ΔTs
[K]
Air flow supplied by
DV or MV
[m³/(s·person)]
Ventilation
system
tested
23 °C 3 K DV 23
3 0,045
DV
23 °C 3 K MV 3 MV
26 °C 3 K DV 26
3 0,045 DV
26 °C 5 K DV 5 0,027 DV
29 °C 3 K DV 29
3 0,045 DV
29 °C 6 K DV 6 0,023 DV
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
85
Results
Physical measurements
The velocity field measurements revealed
that the workstations were not situated
within the adjacent zone defined as the area
near the supply air diffuser where air veloc-
ity exceeds 0,2 m/s at a height of 0,1 m
above the floor. The adjacent zone extended
up to 1,4 m from the DV diffuser when ΔTs
was 3 K (supply air flow of 0,090 m³/s), and
was shorter, up to 1,0 m, when ΔTs was 5 K
or 6 K (supply airflow of 0,054 m³/s or
0,046 m³/s respectively). In the case of mix-
ing ventilation, air velocity at 0,1 m and
1,1 m above the floor was below 0,2 m/s.
The vertical temperature gradient existed in
the case of DV, indicating stratification in
the room (Figure 9.13). The vertical tem-
perature difference between 1,1 and 0,1 m
was 2,0 K, when ΔTs was 3 K. This com-
plies with environmental category A ac-
cording to EN15251 (2007). Vertical tem-
perature differences increased to 3,0 K and
3,5 K, as ΔTs was changed to 5 K and 6 K
respectively. In the case of MV, no vertical
temperature gradient was found, indicating
that the air in the room was well mixed.
Subjective response
The percentage of subjects dissatisfied with
perceived air quality (PAQ) was lowest at
23 °C and was the same with displacement
and mixing ventilation (Figure 9.14). This
result was expected because there was not a
strong pollution source in the room. The in-
crease of the room temperature and ΔTs and
decrease of the supply flow rate caused an in-
crease in the percentage dissatisfied subjects.
The operating mode of the displacement
ventilation had different impact on the peo-
ple’s thermal comfort compared to the im-
pact on PAQ. At room temperatures of 23
and 26 °C, the thermal environment was ac-
ceptable for most of the subjects (only 2 %
complained of thermal discomfort) regard-
less the air distribution (DV or MV), ΔTs
and the supplied flow rate (Figure 9.15). At
29 °C more than 15 % of the subjects were
dissatisfied with the thermal environment.
More subjects (28 %) were dissatisfied at
ΔTs = 3 K and high flow rate (0,090 m³/s)
than at ΔTs =6 K and flow rate of
0,046 m³/s, 18 % of the subjects.
Figure 9.13. Measured vertical temperature
profiles.
Figure 9.14. Percentage dissatisfied people
with the PAQ at the studied operating modes of
room ventilation.
Figure 9.15. Percentage dissatisfied people
with the thermal environment at the studied op-
erating modes of room ventilation.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
86
The impact of the operating mode on the air
movement perception of the people was
also different (Figure 9.16). At 23 °C only
few people (1–2 % of the subjects) were
dissatisfied with the air movement gener-
ated by displacement or mixing ventilation.
In the case of DV, the percentage subjects
dissatisfied with the air movement (too low
air movement) increased with the increase
of the room air temperature. The change of
ΔTs and supply flow rate had a different im-
pact on thermal sensation: more subjects
(44 %) were dissatisfied and requested
more air movement at ΔTs =6 K and low
supply flow rate (0,046 m³/s) than at ΔTs
=3 K and at a high flow rate of 0,090 m³/s,
32 % of the subjects.
Figure 9.16. Percentage dissatisfied people
with the air movement sensation at the studied
operating modes of room ventilation.
Useful outcomes for occupant based
design of displacement ventilation
• The rate at which clean outdoor air is
supplied is more important for PAQ
than decrease of its temperature: The
stratification height will be lowered when
the supply flow rate is decreased and this
will have negative impact on the inhaled
and perceived air quality;
• Increase of room air temperature
above 26 °C has a negative impact on
occupants’ comfort: Increase of the
room air temperature reduces the strength
of the free convection flow around the hu-
man body and thus its ability to entrain
room air. The PAQ will not be felt of high
quality even when the supply flow rate is
increased (small temperature difference)
because the warm air of the free convec-
tion flow around the human body is in-
haled. Complaints due to a warm environ-
ment and lack of air movement will in-
crease. When DV is used at room air tem-
perature above 26 °C supply of cooler air
will help to improve occupants’ thermal
sensation compared to increasing the sup-
ply flow rate;
• No substantial difference between MV
and DV in terms of the thermal com-
fort and PAQ: At a comfortable air tem-
perature and without highly polluted
room air DV and MV perform similarly
with regard to occupants’ thermal com-
fort and PAQ. Request for more air
movement may be reported with DV. No
differences in subjects’ response exists
between MV and DV alone at 23 °C in
terms of the thermal comfort, PAQ or
SBS (sick building symptom) symptoms.
However, a need for more air movement
is reported with DV;
• Use of DV in a warm environment with
additionally provided local convective
cooling is inefficient: The standards rec-
ommend an energy saving strategy by
maintaining relative high room tempera-
ture and improving occupants’ comfort
by locally applied air movement with el-
evated velocity. This strategy is not a fea-
sible application in the case of DV. In this
case the airflow interaction will cause
mixing of the room air and will destroy
the stratification.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
87
9.6 Convective boundary layer around human body
People are important heat sources in occu-
pied spaces. In a comfortable and uniform
indoor thermal environment, the skin and
clothing surface temperatures are higher
than the indoor air temperature. The air in
contact with the skin and clothing becomes
warmer than the surrounding air and a tem-
perature gradient with a resulting density
gradient is established in the layer of air in
the vicinity of the human body. The effect
is a buoyancy force which induces upward
airflow known as free (natural) convection
flow.
The free convection flow starts with a con-
vective boundary layer around the body
which is transformed in a thermal plume
above the body. The convective boundary
layer transports pollution generated by the
human body and in its surroundings to the
breathing zone and therefore is important
for occupants’ exposure and inhaled air
quality. The pollution is transported further
by the thermal plume and is mixed with the
background room air. The importance of
the thermal plume for the performance of
displacement ventilation was already dis-
cussed in Chapter 4 and Chapter 9 (Case
study 9.3). In the following the convective
boundary layer (CBL) is discussed in the
light of its practical importance.
The CBL is slow, laminar and thin over the
lower parts of the body but becomes faster,
turbulent and thick at the height of the head.
Velocity and temperature distribution in the
CBL is similar to that in a free convection
flow over a heated vertical surface: near to
the surface the velocity increases from zero
to a maximum followed by a decrease; the
temperature decreases with increasing dis-
tance from the surface until it reaches the
surrounding temperature (Figure 9.17a).
The thickness of the velocity and tempera-
ture boundary layers varies from less than
5 mm at the lower legs up to 150 mm and
more at the head height (Clark and Toy
1975, Homma and Yakiyama 1988, Özcan
et al. 2005, Licina et al. 2014, Voelker et al.
2014).
a) b)
Figure 9.17. a) Temperature field of the CBL
around seated person – the colours show the
difference between the local temperature in the
CBL and the room air temperature (Homma and
Yakiyama 1988); b) Profiles of velocity meas-
ured with laser Doppler anemometer at the
front/centre of a nude and clothed seated ther-
mal manikin with a female body shape. Calm
environment at 19 °C (Melikov 2015).
At a comfortable room air temperature, the
maximum velocity in the CBL may be as
high as 0,25 – 0,30 m/s. It decreases when
the difference of the body surface tempera-
ture and the surrounding air temperature
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
88
decreases, i.e. when the room temperature
increases (Licina et al. 2014). The velocity
and temperature distribution in the CBL, as
well as the thickness of the boundary layer
is influenced by the body posture, clothing
thermal resistance and its design, presence
of obstacles in the vicinity of the body (for
example, a desk that greatly reduces the
strength of the natural convection flow),
etc. (Licina et al. 2014, 2015).
Breathing also influences the natural con-
vection flow (Özcan et al. 2003, 2005,
Bivolarova et al. 2017). The CBL at the
breathing zone of a person sitting with
slightly open legs is result of interaction of
the convection flow which starts to develop
at the groins with the thermal flows gener-
ated by the thighs and the legs. Such flow
interaction is not present with a standing
posture. Leaning backwards induces a peak
velocity in the CBL that is substantially
(40 %) higher than when leaning forward
(Licina et al. 2014).
Clothing weakens the CBL and reduces its
thickness because its surface temperature is
lower than the skin temperature, though
still higher than the surrounding air temper-
ature (Figure 9.17a). At the same room
temperature covering the body with cloth-
ing will reduce the velocity by half (Me-
likov 2015). Clothing insulation is non-uni-
formly distributed over the body surface
and thus introduces non-uniformity in the
generated CBL. For a seated person the
chair isolates part of the body from the sur-
rounding air and locally blocks the estab-
lishment of the CBL. Changes in the CBL
will have impact on the heat exchange be-
tween the body and the surrounding and
thus on occupant’s thermal comfort (Licina
et al. 2016, Melikov 2015).
The CBL entrains and transports air and
pollution from the surrounding upward. In
rooms with displacement ventilation the
largest proportion of the inhaled air origi-
nates from the CBL (Clark and Cox 1973,
Zhu et al. 2005). The CBL transports pollu-
tion generated in the human body micro-en-
vironment (bio-effluents, secondary prod-
ucts of chemical reaction between ozone
and skin oil, pollution generated from the
clothing, etc.) to the breathing zone (Licina
et al. 2015, Bivolarova et al. 2017). It en-
trains and transports pollution generated in-
doors or infiltrated from outdoors when it
has arrived in the proximity of the human
body (Rim and Novoselac 2009). The CBL
interacts with the transient flow of breath-
ing. The interaction of the CBL with the
flow of exhalation (mouth or nose) is com-
plex and important for exposure to pollu-
tion generated by the body itself (Bivo-
larova et al. 2017).
The performance of displacement ventila-
tion with regard to providing clean air to the
breathing zone depends on the location of
the pollution source. When the pollution
source is located at the floor or lower level
in the room the CBL will bring it upward to
the breathing zone (Rim and Novoselac
2009, Licina et al. 2015a, 2015b). In this
case the best performance of displacement
ventilation will be the same as mixing ven-
tilation (Brohus and Nielsen 1994, Cermak
et al. 2006).
The reduction of exposure to pollution from
a point source located near the feet can be
achieved by control of the CBL. Active and
passive control can be applied (Bolashikov
et al. 2010). The passive control is based on
breaking the CBL at the lower chest level
with a movable board as a part of the desk
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
9. RESEARCH FINDINGS
89
design (Figure 9.18b). The active control
method is based on local suction of the
CBL below the desk (Figure 9.18c). In
both cases the transport of pollution from
the lower level is terminated and new CBL
is developed in front of the body above the
desk level. Closing the gap between the ta-
ble and the abdomen blocks the CBL that
has developed over the legs and reduces the
maximum velocity of the CBL (developed
above the board) at the level of the mouth
from 0,17 m/s to 0,11 m/s (Licina et al.
2014).
Figure 9.18. Control of the CBL in front of sit-
ting person: a) without control, b) passive con-
trol; c) active control (Bolashikov et al. 2010).
Figure 9.19. Effect of local radiant cooling on
the development of the CBL: a) strong CBL; b)
weak CBL when the local clothing surface tem-
perature is equal to the room air temperature;
c) Downward CBL establishes locally when the
clothing surface temperature is lower the room
temperature (Melikov 2015).
The combination of the DV with other
methods for generating high quality indoor
environment should be considered care-
fully because it may have negative effect on
its performance. For example, the genera-
tion of the CBL will cease locally for areas
where due to local radiant cooling the cloth-
ing surface temperature has decreased to
the level of the surrounding air temperature
(Melikov 2015). The CBL will therefore be
weakened (Figure 9.19 middle). When the
clothing surface temperature becomes
lower than the surrounding air temperature,
local downward flows opposing and dis-
turbing the main upward flow of the CBL
may occur (Figure 9.19 right). As a result,
the transport of clean air from near the floor
to the breathing zone will stop and more of
the surrounding polluted air will be inhaled.
The CBL and clothing are in continuous
contact. Clothing made of deodorant mate-
rials can be used to clean and disinfect the
air of the CBL and thus to improve inhaled
air quality (Melikov 2015).
The thermal plume above a person has an
upward velocity of approx. 0,25 m/s (25 cm
above head), and this flow often prevents
draught at head height. Figure 9.20 shows
the thermal boundary layer around a stand-
ing thermal manikin with size and heat pro-
duction as an “average” person. The free
convection flow around the manikin is vis-
ualized by smoke. The interaction between
the thermal plume generated by the mani-
kin and a downward flow with different ve-
locities is demonstrated with six photo-
graphs. The boundary layer is preserved
with a downward velocity lower than
0,25 m/s (Nielsen 2009). A similar result is
reported by Licina et al. (2015c). The direc-
tion and magnitude of the surrounding air-
flows considerably influence the airflow
distribution around the human body.
Downward flow with velocity of 0,175 m/s
does not influence the convective flow in
the breathing zone, while flow at 0,30 m/s
collides with the CBL at the nose level re-
ducing the peak velocity from 0,185 to
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
90
0,10 m/s. Transverse horizontal flow from
the front disturbs the CBL at the breathing
zone even at 0,175 m/s. In case of a mani-
kin sitting on a chair, airflow from below
(assisting the CBL) with velocity of be-
tween 0,30 and 0,45 m/s reduces the peak
velocity in the breathing zone and changes
the flow pattern around the body, com-
pared to the assisting flow of 0,175 m/s or
quiescent conditions. In this case, the air-
flow interaction is strongly affected by the
presence of the chair.
Figure 9.20. A thermal manikin located in a downward air flow. The boundary layer around the
manikin at head height is preserved up to a downward velocity of 0,25 m/s.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
91
10 References
Alamdari, F. 1998. Displacement ventilation and cooling ceilings. In Proceedings of 6th conference on air
distribution in rooms. Stockholm, Sweden, vol. 1, pp. 197-204. Alamdari F, Butler DJG, Grigg PF, Shaw MR. 1998. Chilled ceilings and displacement ventilation.
Renewable Energy, 15: 300-305.
ASHRAE Guideline 11. 2011. Interactions affecting the achievement of acceptable indoor environments.
American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., ASHRAE, Atlanta.
ANSI/ASHRAE. 2013. ANSI/ASHRAE Standard 55-2013. Thermal environmental conditions for human
occupancy. American Society of Heating, Refrigerating and Air Conditioning Engineers, Inc., ASHRAE,
Atlanta.
Arens, A.D. 2000. Evaluation of Displacement Ventilation for Use in High-ceiling Facilities, Master Thesis,
Massachusetts Institute of Technology, Department of Mechanical Engineering, Cambridge USA.
Babiak, J., Olesen B.W, and Petras D. 2009. Low temperature heating and high temperature cooling. REHVA
Guidebook 7.
Bach, H et al. 1993. Gezielte Belüftung der Arbeitsbereiche in Produktionshallen zum Abbau der
Schadstoffbelstung. Forschungsbericht HLK-1-92 ISSN 0943-013 X.
Batchelor, G.K. 1954. Heat convection and buoyancy effects in fluids. Quart. J. Roy. Met. Soc., 80: 339-358.
Bauman, F. 2003. Underfloor Air Distribution (UFAD) Design Guide. ASHRAE Research Project RP-1064.
Bivolarova, M., Kierat, W., Zavrl, E., Popiolek, Z., and Melikov, A. 2017. Exposure to human body bio-
effluents: impact of airflow interaction at the breathing zone. In proceedings of Healthy Buildings 2017
Europe, July 2-5, 2017, Lublin, Poland, Paper ID 0231 ISBN: 978-83-7947-232-1.
Bjørn, E., and Nielsen, P.V. 2002. Dispersal of exhaled air and personal exposure in displacement ventilated
rooms. Indoor Air, 12 (3): 147-164.
Bolashikov, Z.D., Melikov A.K., and Kranek, M. 2010. Improved Performance of Personalized Ventilation
by Control of the Convection Flow around an Occupant’s Body. ASHRAE Transactions, 115 (2): 421-
431.
Brohus, H and Nielsen, PV 1994, Contaminant Distribution around Persons in Rooms Ventilated by
Displacement Ventilation. in Proceedings of ROOMVENT'94: Fourth International Conference on Air
Distribution in Rooms, Cracow, Poland, June 15-17., vol.1, 294-312.
Brohus, H., and Nielsen, P. V. 1996a. Personal Exposure in Displacement Ventilated Rooms. Indoor Air, 6
(3): 157-167.
Brohus, H., and Nielsen, P. V. 1996b. CFD Models of Persons Evaluated by Full-Scale Wind Channel
Experiments. In: Proceedings of the 5th International Conference on Air Distribution in Rooms,
ROOMVENT’96. Yokohama, Japan, July 17-19, Murakami, S. (ed.), 2: 137-144.
Brohus, H., and Ryberg, H. 1999. The Kappa Model– a Simple Model for the Approximate Determination of
Vertical Temperature Distribution in Rooms. Department of Building Technology and Structural
Engineering, Aalborg University.
Brouns, C., and Waters, B. 1991. A guide to contaminant removal effectiveness. Technical note, AIVC, 28:2.
Carrer, P., Fanetti, A., Bischof, W., Hatrmann, T., Kephalopoulos, S., Pelkonen, S., Seppänen, O.,
and Wargocki, P. 2012. Health and ventilation: review of the scientific literature. 2012. Report WP4.
Second Programme of Community Action in the Field of Health 2008-2013 European Commission -
Directorate General for Health and Consumers September 20.
Causone, F., Baldin, F., Olesen, B.W., and Corgnati, S. 2010. Floor heating and cooling combined with
displacement ventilation: possibilities and limitations. Energy and Buildings, 42: 2338-2352.
Cermak, R., and Melikov, A. 2006. Air quality and thermal comfort in an office with underfloor, mixing and
displacement ventilation. International Journal of Ventilation, 5 (3): 323-332.
Cermak, R., Melikov, A.K., Forejt, L., and Kovar, O. 2006. Performance of personalized ventilation in
conjunction with mixing and displacement ventilation. HVAC&R Research, 12 (2): 295-311.
Chen, Q., Glicksman, L., Yuan, X., Hu, S., Hu, Y., and Yang, X. 1999. Performance evaluation and
development of design guidelines for displacement ventilation. Final Report for ASHRAE RP949,
Massachusetts Institute of Technology, Cambridge.
Chen, Q., and Glicksman, L. 2003. System performance evaluation and design guidelines for displacement
ventilation. RP-949, ISBN 1-931862-42-7 ASHRAE, Inc. Atlanta, GA.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
92
Clark, R.P., and Toy, N. 1975. Forced Convection Around the Human Head. Journal of Physiology, 244: 295-
302.
Clark, R.P., and Cox, R.N. 1973. The generation of aerosols from the human body; in Airborne Transmission
and Airborne Infection: Concepts and Methods. Willy New York, Chapter 95: 413-426.
Dalewski, M., Melikov, A.K., and Vesely, M. 2014. Performance of ductless personalized ventilation in
conjunction with displacement ventilation: Physical environment and human response. Building and
Environment, 81: 354-364.
da Graça, G. C. 2003. Simplified models for heat transfer in rooms [Ph.D. thesis]. San Diego: Department of
Engineering Physics, University of California.
Deevy, M., Sinai, Y., Everitt, P., Voigt, L., and Gobeau, N. 2008. Modelling the effect of an occupant on
displacement ventilation with computational fluid dynamics, Energy and Buildings, 40: 255-264.
DOE 2015. EnergyPlus engineering reference. The Reference to EnergyPlus Calculations. DOE, United
States.
Etheridge, D., and Sandberg, M. 1996. Building Ventilation - Theory and Measurement. Wiley.
EN 15251. 2007. Criteria for the indoor environment including thermal, indoor air quality, light and noise.
European Committee for Standardization. Brussels.
Eriksson, L., Grozman, G., Grozman, P., Sahlin, P., Vorre, M., and Ålenius, L. 2012. CFD-free, efficient,
micro indoor climate prediction in buildings. Inproceedings of the First building simulation and
optimization conference. Loughborough, UK, 10-11 September.
Fang, L., Wyon, D.P., Clausen, G., and Fanger, P.O. 2004. Impact of indoor air temperature and humidity in
an office on perceived air quality, SBS symptoms and performance. Indoor Air, 14 (s7): 74–81.
Fitzner, K. 1989 Förderprofil einer Wärmequelle bei verschiedenen Temperaturgradienten und der Einfluss
auf die Raumströmung bei Quellüftung. Ki Klima-Kälte-Heizung, Nr. 10.
Fitzner, K. 1996. Displacement ventilation and cooled ceilings, results of laboratory tests and practical
installations. In Proceedings of Indoor Air’1996, Nagoya., vol.1, pp. 41-50.
Fitzner, K. 2001. Private communications.
Griffith, B. T. 2002 Incorporating nodal and zonal room air models into building energy calculation
procedures [Doctoral dissertation]. Massachusetts Institute of Technology.
Hao X, Zhang G, Chen Y, Zou S, Moschandreas D.J 2007, A combined system of chilled ceiling,
displacement ventilation and desiccant dehumidification, Build. Environ. 42 3298-3308.
Hagström, K., Sandberg, E., Koskela, H., and Hautalampi, T. 2000. Classification for the room air
conditioning strategies. Building and Environment, 35: 699-707.
Halton Oy. 2000. Displacement Ventilation Design Guide.
Halvoňová, B., and Melikov, A. 2010. Performance of “ductless” personalized ventilation in conjunction with
displacement ventilation: Impact of disturbances due to walking person(s). Building and Environment, 45
(2): 427-436.
Hensen, J.L.M., and Hamelinck, M.J.H. 1995. Energy simulation of displacement ventilation in offices.
Building Services Engineering Research and Technology, 16(2): 77-81.
Hodder, S.G., Loveday, D.L., Parsons, K.C., and Taki, A.H. 1998. Thermal comfort in chilled ceiling and
displacement ventilation environments: vertical radiant temperature asymmetry effects. Energy and
Buildings, 27: 167-173
Homma, H., and Yakiyama, M. 1988. Examination of free convection around occupant’s body caused by its
metabolic heat. ASHRAE Transactions, 94(1):104–24.
ISO. 1998. ISO 7726, International Standard: Ergonomics of the thermal environment - Instruments for
measuring physical quantities. International Organization for Standardization.
ISO. 2005. ISO 7730, Moderate thermal environment- Determination of the PMV and PPD indices and
specification of the conditions for thermal comfort. International Organization for Standardization.
Jacobsen, T.V., and Nielsen, P.V. 1992. Velocity and temperature distribution in flow from an air inlet device
with displacement ventilation. In proceedings of Roomvent’92, p.21, ISBN 87-982652-6-2, vol. 3, pp. 23-
32.
Jaluria, Y. 1980. Natural convection, heat and mass transfer. Pergamon Press.
Jin, Y. 1993. Particle transport in turbulent buoyant plumes rising in a stably stratified environment. Ph.D.
Thesis. Dept. of Building Services Engineering, KTH, Stockholm.
Koestel, A. 1954. Computing Temperatures and Velocities in Vertical Jets of Hot or Cold Air. Heating,
Piping & Air Conditioning, June.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
10. REFERENCES
93
Kosonen, R., and Mustakallio, P. 2010. Ventilation in classroom: a case-study of the performance of different
air distribution methods. In proceedings of Clima 2010 10th. REHVA World Congress. Sustainable
Energy Use in Buildings. (CLIMA 2010). 9-12 May, Antalya, Turkey.
Kosonen, R., Lastovets, N., Mustakallio, P., da Graca, G.C., Mateus, N.M., and Rosenqvist, M. 2016. The
effect of typical buoyant flow elements and heat load combinations on room air temperature profile with
displacement ventilation. Building and Environment, 108: 207-219.
Krühne, H. 1995. Experimentelle und theoretische Untersuchungen zur Quelluftströmung, Dissertation TU
Berlin.
Krühne, H., Fitzner, K. 1995. Luftqualität in der Atemzone von Personen mit Quelluftströmung. Luft- und
Kältetechnik 12.
Lestinen, S., Koskela, H., Nyyssölä, H., Sundman, T. L., Laine, T., and Siikonen, T. 2012. CFD-simulation
and measurement of indoor environment in a multipurpose arena. In Proceedings of Ventilation 2012,
The 10th International Conference on Industrial Ventilation. 17-19 September, INRS, Paris, France, paper
V077, 6 pages.
Lestinen, S., Koskela, H., Jokisalo, J., Kilpeläinen, S., and Kosonen, R. 2016. The use of displacement and
zoning ventilation in a multipurpose arena. International Journal of Ventilation, 5 (2): 151-166.
Li, Y., Sandberg, M., and Fuchs, L. 1992. Vertical Temperature Profiles in Rooms Ventilated by
Displacement: Full‐Scale Measurement and Nodal Modelling. Indoor Air, 2(4): 225-243.
Li., Y., Nielsen, P.V., and Sandberg, M. 2011. Displacement Ventilation in Hospital Environments. ASHRAE
Journal, 53 (6): 86-88.
Licina, D., Melikov, A., Pantelic, J., Sekhar, C., and Tham, K.W. 2014. Experimental investigation of the
human convective boundary layer in a quiescent indoor environment, Building and Environment, 75: 79-
91.
Licina, D., Melikov, A., Sekhar, C., and Tham, K.W. 2015. Air temperature investigation in
microenvironment around a human body. Building and Environment, 92: 39-47.
Licina, D., Melikov, A., Pantelic, J., Sekhar, C. and Tham, K.W. 2015a. Human convection flow in spaces
with and without ventilation: Personal exposure to floor released particles and cough released droplets,
Indoor Air, 25(6):672-82.
Licina, D., Melikov, A., Sekhar, C., and Tham, K.W. 2015b. Transport of Gaseous Pollutants by Convective
Boundary Layer around a Human Body. Science and Technology for the Built Environment, 21 (8):
1175-1186.
Licina, D., Melikov, A.K., Sekhar, C., Tham, K.W. 2015c. Human convective boundary layer and its
interaction with room ventilation flow, Indoor Air, doi:10.1111/ina.12120, 25(1): 21–35, 2015.
Licina, D., Melikov, A., Sekhar, C., and Tham, K.W. 2016. Airflow characteristics and pollution distribution
around a thermal manikin - Impact of specific personal and indoor environmental factors. ASHRAE
Transactions, American Society of Heating, Refrigerating and Air-Conditioning Engineers, 122: 366-379.
Livchak A and Nall D 2001. Displacement ventilation- application for hot and humid climate. Clima 2001
Napoli World Congress. 15-18 September.
Loveday, D.L., Parson, K.C., Taki, A.H., and Hodder, S.G. 2002. Displacement ventilation environment with
chilled ceilings: thermal comfort design within the context of the BS EN IS07730 versus adaptive debate.
Energy and Buildings, 34 (6): 573-579.
Magnier-Bergeron, L., Derome, D., and Zmeureanu, R. 2017. Three-dimensional model of air speed in the
secondary zone of displacement ventilation jet. Building and Environment, 114: 483-494.
Mateus, N. M., da Graça, G. C. 2015. A validated three-node model for displacement ventilation. Building
and Environment, 84: 50-59.
Melikov, A.K., and Langkilde, G. 1990. Displacement ventilation - Airflow in the near zone. In Proceedings
of ROOMVENT'90, Oslo, Norway, June 13-15, Session B1- 3, Paper 23.
Melikov, A.K., Pitchurov, G., Naidenov, K., and Langkilde, G. 2005. Field study of occupants thermal
comfort in rooms with displacement ventilation. Indoor Air, 15 (3): 205-214.
Melikov, A.K. and Kaczmarczyk, J. 2012. Air movement and perceived air quality, Building and
Environment, 47: 400-409.
Melikov, A.K. 2015. Human body micro-environment: The benefits of controlling airflow interaction.
Building and Environment, 91: 70-77.
Mierzwinski, S. 1981. Air motion and temperature distribution above a human body in result of natural
convection. A4-serien no. 45, Inst. för Uppv.- o Vent. teknik, KTH, Stockholm.
Morton, B.R., Taylor, G., and Turner, J.S. 1956. Turbulent gravitational convection from maintained and
instantaneous sources. In Proceeding of Royal Soc., 234A: 1.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
94
Mundt, E. 1990. Convection flows above common heat sources in rooms with displacement ventilation In
Proceedings of the Second International Conference on Air Distribution in Rooms, ROOMVENT’90,
Oslo, Norway
Mundt, E. 1992. Convection Flows in Rooms with Temperature Gradients - Theory and Measurements. In
Proceedings of the Third International Conference on Air Distribution in Rooms, ROOMVENT’92,
Aalborg, Denmark, 3, vol. 3, pp. 69-86.
Mundt, E. 1994. Contamination Distribution in Displacement Ventilation - Influence of Disturbances.
Building and Environment, Vol 29, No. 3. pp. 311-317.
Mundt, E. 1996. The Performance of Displacement Ventilation Systems- Experimental and Theoretical
studies. Ph.D. Thesis, Bulletin no 38, Building Services Engineering, KTH, Stockholm.
Mundt, E. 2001. Non-buoyant pollutant sources and particles in displacement ventilation. Building and
Environment, 36 (7): 829-836.
Mundt, E., Mathisen, H. M., Nielsen, P. V. and Moser, A. 2004. Ventilation Effectiveness. REHVA
Guidebook No. 2, p. 75.
Murakami, S., Kato, S., and Zeng, J. 1995. Development of a Computational Thermal Manikin – CFD
Analysis of Thermal Environment around Human Body. In Proceedings of Tsinghua HVAC-’95, Beijing,
2: 349 – 354.
Mustakallio, P. and Kosonen, R. 2011. Indoor air quality in classroom with different air distribution systems.
In proceedings of Indoor Air June 5-10. Austin Texas USA, paper 952.
Mustakallio, P., Rosenqvist, M., Sinai, Y., and Kosonen, R. 2012. Full-scale test and CFD-simulation of
indoor climate conditions in displacement ventilation case with different room heights and CFD models.
In Proceedings of the 10th International Conference on Ventilation, Paris, France. 17-19 September.
Nielsen, P.V., Hoff, L., and Pedersen, L.G. 1988. Displacement Ventilation by Different Types of Diffusers.
In proceedings of the 9th AIVC Conference on Effective Ventilation, Gent, Belgium, Sept 12-15.,
Nielsen, P.V. 1992a. Air Distribution Systems: Room Air Movement and Ventilation Effectiveness.
International Symposium on Room Air Convection and Ventilation Effectiveness, ISRACVE, Tokyo,
July.
Nielsen, P.V. 1992b. Velocity Distribution in the Flow from a Wall-Mounted Diffuser in Rooms with
Displacement Ventilation. In Proceedings of ROOMVENT'92, The Third Int. Conf. on Air Distribution in
Rooms, Aalborg, Denmark, September, vol. 3, pp. 1-19.
Nielsen,P.V.1993. Displacement Ventilation- Theory and Design. Aalborg University, Aalborg, Denmark,
ISSN 0902-8002 U9306.
Nielsen, P.V. 1994a Velocity Distribution in a Room with Displacement Ventilation and Low-Level
Diffusers. International Report, IEA Annex 20, Aalborg University. ISSN 0902-R9403.
Nielsen, PV 1994b, 'Stratified Flow in a Room with Displacement Ventilation and Wall-Mounted Air
Terminal Device' ASHRAE Transactions, vol 100, Part 1, pp. 1163-1169.
Nielsen, PV 1995, Vertical Temperature Distribution in a Room with Displacement Ventilation., in IEA
Annex 26: Energy Efficient Ventilation of Large Enclosures, Rome 1995. Aalborg. ISSN 0902-7513
R9509.
Nielsen, P.V. 1996. Temperature Distribution in a Displacement Ventilated Room. In Proceedings of the 5th
International Conference on Air Distribution in rooms, ROOMVENT’96, Yokohama, Japan, July 17-19,
1996, 3: 323-330. ISBN: 4-924557-01-3.
Nielsen, P. V. 2000. Velocity Distribution in a Room Ventilated by Displacement Ventilation and Wall-
Mounted Air Terminal Devices. Energy and Building, 31 (3): 179-187.
Nielsen, P. V. Murakami, S., Kato, S., Topp, C. and Yang, J.-H. 2003. Benchmark Tests for a Computer
Simulated Person. Aalborg University, Indoor Environmental Engineering, ISSN 1395-7953 R0307.
Nielsen, PV 2003, Temperature and Air Velocity Distribution in Rooms Ventilated by Displacement
Ventilation. in KHKNTKEM (eds.) (red.), Proceedings of the 7th International Symposium on
Ventilation for Contaminant Control, Sapporo, Japan, pp.: 691-696.
Nielsen, PV, Nickel, J & Baron, DJG 2004, Plane Stratified Flow in a Room Ventilated by Displacement
Ventilation. in MCG Da Silva (red.), Proceedings of ROOMVENT 2004, 9th International Conference on
Air Distribution in Rooms, September 5-8, 2004, Coimbra, Portugal, pp. 249-250.
Nielsen, P.V. 2007. Analysis and design of room air distribution systems. American Society of Heating,
Refrigerating and Air-Conditioning Engineers, Inc., HVAC&R Research, 13 (6):987-997.
Nielsen, P. V., Hyldgaard, C. E., Melikov, A. K., Andersen, H., and Soennichsen, M. 2007. Personal
exposure between people in a room ventilated by textile terminals – with and without personalized
ventilation. HVAC&R Research, 13 (4): 635-643.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
10. REFERENCES
95
Nielsen, P.V., Winther, F.V., Buus, M., and Thilageswaran, M. 2008. Contaminant Flow in the
Microenvironment Between People Under Different Ventilation Conditions. ASHRAE Transactions, Vol
114, Part 2, pp. 632-640.
Nielsen, P. V. 2009. Control of Airborne Infectious Diseases in Ventilated Space. Journal of the Royal
Society Interface, Vol 6(6), pp.747-756.
Nielsen P.V., Li Y., Buus M., and Winther F.V. 2010. Risk of cross-infection in a hospital ward with
downward ventilation. Building and Environment, 45: 2008-2014.
Nielsen, P.V. 2011. The "Family Tree" of Air Distribution Systems. Proceedings of ROOMVENT 2011, 12th
International Conference on Air Distribution in Rooms, June, Trondheim, Norway. TAPIR Akademisk
Forlag. Norway. ISBN 978-82-519-2812-0.
Nielsen, P.V., Olmedo, I., Adana, M.R.D., Grzelecki, P., and Jensen, R.L. 2012. Airborne Cross-Infection
Risk Between two People Standing in surroundings with a Vertical Temperature Gradient. HVAC&R
Research, 18 (4):552–561.
Nordtest 2002. Nordtest Method: Air Terminal Devices: Aerodynamic Testing and Rating at Low Velocity.
NT VVS 083:A, ISSN 0283-7226, Finland.
Novoselac, A., Burley, B. J., and Srebric, J. 2006. Development of new and validation of existing convection
correlations for rooms with displacement ventilation systems. Energy and buildings, 38(3): 163-173.
Olmedo, I., Nielsen, P.V., Adana, M.R.D., Jensen, R.L., and Grzelecki, P. 2012. Distribution of Exhaled
Contaminants and Personal Exposure in a Room using Three Different Air Distribution Strategies. Indoor
Air, 22 (1): 64–76.
Popiolek, Z. 1981. Problems of testing and mathematical modeling of plumes above human body and other
extensive heat sources. A4-serien no. 54, Inst. för Uppv.- o Vent. teknik, KTH, Stockholm.
Qian, H., Li, Y., Nielsen, P.V., Hyldgård, C.-E., Wai Wong, T., and Chwang, A.T.Y. 2006. Dispersion of
exhaled droplet nuclei in a two-bed hospital ward with three different ventilation systems. Indoor Air, 16
(2):111-128.
Rees, S. J. 1998. Modelling of Displacement Ventilation and Chilled Ceiling Systems using Nodal Models.
Ph.D. Thesis, Loughborough University.
Rees, S. J., and Haves, P. 2001. A nodal model for displacement ventilation and chilled ceiling systems in
office spaces. Building and Environment, 36(6): 753-762.
Rees, S. J., and Haves, P. 2013. An experimental study of air flow and temperature distribution in a room
with displacement ventilation and a chilled ceiling. Building and Environment, 59: 358-368.
Rim, D. and Novoselac A. 2009. Transport of particulate and gaseous pollutants in the vicinity of a human
body. Building and Environment, 44: 1840–1849.
Sahlin, P. 1996. Modelling and Simulation Methods for Modular Continuous Systems in Buildings. Doctoral
Dissertation. Department of Building Sciences, Division of Building Services Engineering, Royal
Institute of Technology, Stockholm, Sweden. ISSN 0284-141X.
Schiavon, S., Bauman, F., Tully, B., and Rimmer, J. 2015. Chilled ceiling and displacement ventilation
system: laboratory study with high cooling load. Science and Technology for the Built Environment, 21
(7): 944-956.
Schmidt, W. 1941. Turbulente Ausbreitung eines Stromes erhitzer Luft ZAMM. Bd. 21 #5.
Simmonds, P., Gaw, W., Holst, S., Reuss, S. 2000. Using radiant cooled floors to condition large spaces and
maintain comfort conditions. ASHRAE Transactions, 106 (1): 695-701.
Skistad, H. 1994. Displacement Ventilation. Research Studies Press, John Wiley & Sons, Ltd., West Sussex.
UK.
Skåret, E. 2000. Ventilasjonsteknisk hand-bok. Håndbok 48 Norges Bygg-forskningsinstitutt. ISBN 82-536-
0714-8.
Stymne, H., Sandberg, M., and Mattsson, M. 1991. Dispersion pattern of contaminants in a displacement
ventilated room- implications for demand control. In proceedings of 12th AIVC conference, Ottawa,
Canada.
Sutcliffe, H. 1990. A guide to air change efficiency. Technical note AIVC 28, AIVC.
Tan, H., Murata, T., Aoki, K., Kurabuchi, T. 1998. Cooled ceilings / displacement ventilation hybrid air
conditioning system - Design Criteria. In Proceedings of Roomvent '98, Stockholm, vol. 1, pp. 77-84.
Topp, C., Nielsen, P. V., and Sørensen, D. N. 2002. Application of Computer-Simulated Persons in Indoor
Environmental Modeling. ASHRAE Transactions, 108 (2). pp.1084-1089.
Turner, J.S. 1973. Buoyancy effects in fluids. Cambridge University Press. ISBN 0 521 08623 X.
Voelker, C., Maempel, S., Kornadt, O. 2014. Measuring the human body's microclimate using a thermal
manikin. Indoor Air, 24: 567–579.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Displacement Ventilation Guidebook
96
Wildeboer, J. and Müller, D. 2006. Lüftungseffektivität als Qualitätskriterium für Quelllüftungssysteme - die
Auswirkung des Luftaustauschwirkungsgrades auf die Höhe der Frischluftschicht. KI - Luft- und
Kältetechnik, C. F. Müller Verlag Hüthig, 42 (10): 439-443.
Yamanaka, T., Kotani, H., and Xu, M. 2007. Zonal Models to Predict Vertical Contaminant Distribution in
Room with Displacement Ventilation Accounting for Convection Flows along Walls. In: Proceedings of
Roomvent 2007, session C01 – Room Air Distribution Systems, Paper 1279.
Zhao, K., Liu, X., and Jiang, Y. 2016. Application of radiant floor cooling in large space buildings- a review.,
Renew. Sust. Energy Rev. 55: 1083-1096.
Zhu, S.W., Kato, S., Murakami, S., and Hayashi, T. 2005. Study on inhalation region by means of CFD
analysis and experiment. Building and Environment, 40(10):1329-36.
Zeldovitch, Y.B. 1937. Fundamental Principles for Free Convective Plumes. Journal of the Experimental and
Technical Physics, 7 (12).
Zukowska, D., Melikov, A., and Popiolek, Z. 2008. Impact of Thermal Plumes Generated by Occupant
Simulators with Different Complexity of Body Geometry on Airflow Pattern in Rooms. In: Proceedings
of the 7th International Thermal Manikin and Modeling Meeting – 7I3M, Coimbra, Portugal, Paper 8.
Zukowska, D., Popiolek Z., and Melikov, A., 2010a. Impact of boundary conditions on the development of
the thermal plume above a sitting human body. In: Proceedings of CLIMA 2010, Antalya, Turkey, Paper
R7-TS55-PP05.
Zukowska, D., Popiolek, Z., and Melikov, A. 2010b. Determination of integral characteristics of an
asymmetrical thermal plume from air speed/velocity and temperature measurements. Experimental
Thermal and Fluid Science, 34: 1205-1216.
Zukowska, D. 2011a. Airflow interaction in rooms - Convective plumes generated by occupants. PhD Theses.
Department of Civil Engineering, The Technical University of Denmark, p 58.
Zukowska, D., Melikov, A., Popiolek, Z., and Spletsteser, J. 2011b. Impact of facially applied air movement
on the development of the thermal plume above a sitting occupant. In: Proceedings of the 12th
International Conference on Air Distribution in Rooms, Roomvent’2011, Trondheim, Norway, Paper No.
161.
Zukowska, D., Melikov, A., and Popiolek, Z. 2012a. Impact of personal factors and furniture arrangement on
the thermal plume above a sitting occupant. Building and Environment, 49: 104 - 116.
Zukowska, D., Melikov, A., and Popiolek, Z. 2012b. Impact of geometry of a sedentary occupant simulator
on the generated thermal plume: Experimental investigation. HVAC&R Research, 18 (4): 795-811.
Özcan, O., Meyer, K.E., and Melikov, A. 2003. Turbulent and Stationary Convective Flow Field Around the
Head of a Human. In Proceedings of International Symposium on Turbulence, Heat and Mass Transfer -
THMT-03, October 12-17, Antalya, Turkey. Turbulence, Heat and Mass Transfer 4, K. Hanjali ´c, Y.
Nagano and M. Tummers (Editors) 2003 Begell House, Inc.
Özcan, O., Mayer, K.E., and Melikov, A.K. 2005. A visual description of the convective flow field around the
head of a human. Journal of Visualization, 8 (1): 23-31.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
Single user license only, copying and networking prohibited. All rights reserved by REHVA.
REHVA Guidebooks:No 1 Displacement Ventilation in Non-industrial Premises (Out of print)No 2 Ventilation EffectivenessNo 3 Electrostatic Precipitators for Industrial ApplicationsNo 4 Ventilation and Smoking (Out of print)No 5 Chilled Beam CoolingNo 6 Indoor Climate and Productivity in OfficesNo 7 Low Temperature Heating And High Temperature CoolingNo 8 Cleanliness of Ventilation SystemsNo 9 Hygiene Requirement for Ventilation and Air-conditioningNo 10 Computational Fluid Dynamics in Ventilation DesignNo 11 Air Filtration in HVAC SystemsNo 12 Solar Shading – How to integrate solar shading in sustainable buildingsNo 13 Indoor Environment and Energy Efficiency in Schools – Part 1 PrinciplesNo 14 Indoor Climate Quality AssessmentNo 15 Energy Efficient Heating and Ventilation of Large HallsNo 16 HVAC in Sustainable Office Buildings – A bridge between owners and engineersNo 17 Design of energy efficient ventilation and air-conditioning systemsNo 18 Legionellosis Prevention in Building Water and HVAC SystemsNo 19 Mixing Ventilation – Guide on mixing air distribution designNo 20 Advanced system design and operation of GEOTABS buildingsNo 21 Active and Passive Beam Application Design Guide – For Global ApplicationNo 22 Introduction to Building Automation, Controls and Technical Building ManagementNo 23 Displacement Ventilation
REHVA Reports:No 1 REHVA Workshops at Clima 2005 - LausanneNo 2 REHVA Workshops at Clima 2007 - HelsinkiNo 3 REHVA Workshops at Clima 2010 - AntalyaNo 4 REHVA nZEB ReportNo 5 REHVA Workshops at Clima 2013 - PragueNo 6 REHVA Workshops at Clima 2016 - Aalborg
rehvaFederation of European Heating, Ventilation and Air Conditioning Associations
REHVA OfficeWashington Street 40, 1050 Brussels – BelgiumTel: +32-2-5141171 Fax: +32-2-5129062Orders: www.rehva.eu [email protected]
Single user license only, copying and networking prohibited. All rights reserved by REHVA.