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.

Displacement

Ventilation

Risto Kosonen (ed.)

Arsen Melikov

Elisabeth Mundt

Panu Mustakallio

Peter V. Nielsen

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


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


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


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


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.


http://www.rehva.eu/member-associations/belgium.html

http://www.rehva.eu/member-associations/croatia.html

http://www.rehva.eu/member-associations/czech-republic.html

http://www.rehva.eu/member-associations/denmark.html

http://www.rehva.eu/member-associations/estonia.html

http://www.rehva.eu/member-associations/finland.html

http://www.rehva.eu/member-associations/france.html

http://www.rehva.eu/member-associations/germany.html

http://www.rehva.eu/member-associations/hungary-ete.html

http://www.rehva.eu/member-associations/italy.html

http://www.rehva.eu/member-associations/latvia.html

http://www.rehva.eu/member-associations/lithuania.html

http://www.rehva.eu/member-associations/moldavia.html

http://www.rehva.eu/member-associations/netherlands.html

http://www.rehva.eu/member-associations/norway.html

http://www.rehva.eu/member-associations/poland.html

http://www.rehva.eu/member-associations/portugal.html

http://www.rehva.eu/member-associations/romania.html

http://www.rehva.eu/member-associations/russia.html

http://www.rehva.eu/member-associations/serbia.html

http://www.rehva.eu/member-associations/slovakia.html

http://www.rehva.eu/member-associations/slovenia.html

http://www.rehva.eu/member-associations/spain.html

http://www.rehva.eu/member-associations/sweden.html

http://www.rehva.eu/member-associations/switzerland.html

http://www.rehva.eu/member-associations/turkey.html

http://www.rehva.eu/member-associations/united-kingdom.html

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.


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,


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


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.


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 [-]


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]


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


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.


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


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



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.



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).


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



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).



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).


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.



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


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



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



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%



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



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



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



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



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



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



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



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



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



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).



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



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



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



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.


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


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).



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



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



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.



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.



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).


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



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.


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


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.


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).



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



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).



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.



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



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



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)



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



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



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



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]



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



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


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


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



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.



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.



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



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



Dimensionless temperature profile in occupied zone and floor heating capacity



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



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



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.



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



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


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



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.


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.



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.


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

)


Mundt


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

)


Mundt


Nielsen

Φ10occupants = 750 W, Φfloor= 260 W,

Φlight= 232 W, Φwin_3.5m= 520 W,

θs = 18,1 °C, qv = 0,1 m³/s



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.


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).


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.


http://www.cfd-benchmarks.com/

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.



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.



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



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



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.



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.



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.



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).



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


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



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



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.


with the thermal environment at the studied op-

erating modes of room ventilation.



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.


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.



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



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



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



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.


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.


http://www.aivc.org/resource/displacement-ventilation-and-cooled-ceilings

http://www.aivc.org/resource/displacement-ventilation-and-cooled-ceilings

http://vbn.aau.dk/en/persons/henrik-brohus(c7c64297-e1c1-4b88-a402-85fde5c8ee22).html

http://vbn.aau.dk/en/persons/peter-v-nielsen(196620ef-551a-4bda-8a8a-252ad68f185b).html

http://vbn.aau.dk/en/publications/contaminant-distribution-around-persons-in-rooms-ventilated-by-displacement-ventilation(2c055fe0-9c2e-11db-8ed6-000ea68e967b).html

http://vbn.aau.dk/en/publications/contaminant-distribution-around-persons-in-rooms-ventilated-by-displacement-ventilation(2c055fe0-9c2e-11db-8ed6-000ea68e967b).html


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.


https://research.aalto.fi/en/journals/building-and-environment(c6543c1d-e9c5-4fd6-bc0a-7111e1899314)/publications.html

https://research.aalto.fi/en/journals/building-and-environment(c6543c1d-e9c5-4fd6-bc0a-7111e1899314)/publications.html

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.



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.


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.


http://vbn.aau.dk/da/persons/peter-v-nielsen(196620ef-551a-4bda-8a8a-252ad68f185b).html

http://vbn.aau.dk/da/persons/carlerik-hyldgaard(cf3c5c11-1b10-4af8-8a60-351555f72ffb).html

http://vbn.aau.dk/da/publications/dispersion-of-exhaled-droplet-nuclei-in-a-twobed-hospital-ward-with-three-different-ventilation-systems(88f3c950-d13d-11db-b8b8-000ea68e967b).html

http://vbn.aau.dk/da/publications/dispersion-of-exhaled-droplet-nuclei-in-a-twobed-hospital-ward-with-three-different-ventilation-systems(88f3c950-d13d-11db-b8b8-000ea68e967b).html

http://vbn.aau.dk/da/journals/indoor-air(f94ec27d-b806-43d6-9b27-dabc7fdd445b).html


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.


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]


Displacement Ventilation - Biblioteka...2017/06/23 · Displacement Ventilation rehva Federation of...

Documents

Transcript of Displacement Ventilation - Biblioteka...2017/06/23 · Displacement Ventilation rehva Federation of...