Lectures On Building Physics

Prof Gudni Jhannesson
Div. of Building Technology

Dept. of Civil and Architectural Engineering
School of Architecture and Built
Environment
S 100 44 Stockholm
Tel: +46 (0)8 790 8670
Fax: +46 (0)8 411 8432
E-mail gudni.johannesson@byv.kth.se
Lectures on Building Physics
Hotel Hamar. Borgarnes Iceland.

Architect: Peter Ottosson
Heat and Moisture Transfer

Lecture notes
Building Materials and Building Physics
Prof. Gudni Jhannesson - Lectures on Building Physics

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Autumn 2005

2005-10-24
-------------------------------------------------------------------------------------------------------------------

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Preface
As building physics is a rather young discipline a suitable course literature
is not very easily found that suits the level and aims of the building physic
part of the present course in building physics and building materials within
the program for Civil Engineering and Urban Management (S) at KTH. The
following lecture notes have been worked out over the last years to cover the
theoretical content of the different lectures on building physics within the
course. This edition has been in some part rewritten and given some
features of a more finished publication.
In this edition some examples have been added to the text. As a
complementary material, solved examples in Mathcad that cover
applications of most theory given have been worked out. Mathcad which is
now available to all students at KTH is a powerful tool to solve physical
problems but at the same time the used code closely follows the way of
expressing theoretical expressions in general which makes it possible to
read solved problems without to much knowledge about the program. It is at
this stage not demanded that the students do their own programming in
Mathcad but it is my believe that it will, for those who try it, be a powerful
tool to train on applications of the theory, and it is a good link to bring the
learned skill along to future tasks in advanced courses and later on in real
professional life.
These lecture notes, together with the solved examples and exercises
handed out, define the contents of building physics within the course.
October 2005
Gudni Jhannesson
Professor in Building Technology

2005-10-24
-------------------------------------------------------------------------------------------------------------------

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Table of contents
PREFACE
1
LIST OF SYMBOLS
1.1
4
10
Indexes
13
BUILDING PHYSICS IN CIVIL ENGINEERING
15
2.1
What building physics is mainly about
15
2.2
Examples of heat, air and moisture transfer
15
2.3
Combined processes
16
2.4
New mathematical tools make life easier
16
2.5
Building physics and the environment
17
INTRODUCTION TO BUILDING HEAT TRANSFER
19
3.1
The basic elements of building heat transfer
19
3.2
Steady state calculations
22
3.3
Two-dimensional heat conduction
25
HEAT CONDUCTION EQUATION ANALYTICAL SOLUTIONS
29
4.1
Simplifications
29
4.2
Solution for a one dimensional slab with harmonic boundary temperatures
30
4.3
Steady state heat flow - simplifications
34
NUMERICAL METHODS
39
5.1
The control element - two dimensional heat flow
39
5.2
Expression of boundary conditions
42
5.3
Resulting heat balance
42
5.4
Choice of time steps and element size - stability
43

2005-10-24
------------------------------------------------------------------------------------------------------------------5.5
Simplifications for special cases
44
5.6
Finite element methods
45
CONVECTIVE HEAT TRANSFER
47
6.1
Basics of convective heat transfer
47
6.2
Expressions for different flow types
50
6.3
Heat balance in a duct with air flow
50
6.4
Expressions for surface heat transfer coefficients
52
6.5
Air flow through building components
57
RADIATIVE HEAT TRANSFER
61
7.1
Exemplification of applications
61
7.2
Basic theory
62
7.3
Long wave radiation exchange
64
7.4
Calculation of view factors
66
7.5
The Radiosity matrix
70
7.6
Combined radiation and convective heat transfer
73
7.7
Solar radiation in building design
73
7.8
The elements of solar radiation
74
7.9
Heat balance on exterior surfaces
75
HEAT BALANCE PROBLEMS
77
8.1
General heat balance problems
77
8.2
Heat balance for a building
77
8.3
Window heat balance
79
8.4
The heat balance for a ventilated air gap
84
EXAMPLES - HEAT TRANSFER APPLICATIONS
87
10
MOISTURE IN BUILDINGS - INTRODUCTION
97

2005-10-24
------------------------------------------------------------------------------------------------------------------10.1
What do we mean by moisture?
97
10.2
Exterior timber framed wall
97
10.3
Boundary conditions
99
10.4
Moisture in building materials
102
10.5
Rainwater
103
10.6
Ground water
104
11
MOISTURE TRANSFER IN CONSTRUCTIONS - DIFFUSION
105
11.1
Moisture transfer - introduction
105
11.2
Examples of applications
105
11.3
The physics of moisture flow in porous materials
105
11.4
Diffusion
106
11.5
Diffusion in porous materials
107
11.6
Moisture surface resistances
108
11.7
Distribution of relative humidity
109
12
MOISTURE TRANSFER IN CONSTRUCTIONS - CAPILLARY
TRANSFER
111
12.1
Water in porous materials
111
12.2
Capillary flow
112
12.3
Water absorption
114
12.4
Capillary moisture transfer within materials
115
13
MOISTURE TRANSFER IN CONSTRUCTIONS - CONVECTION VENTILATED CONSTRUCTIONS
117
13.1
Water flow through porous materials
117
13.2
Water flow through a crack.
117
13.3
Moisture flow by air convection
118
13.4
Heat and moisture balance in a ventilated air gap.
119

2005-10-24
-------------------------------------------------------------------------------------------------------------------
14
APPLIED MOISTURE TRANSFER - DAMAGES
121
14.1
Attic ventilation
121
14.2
Solar condensation
123
14.3
Smell of fungi, the search for possible reasons
123
14.4
The wooden foot rail
124
14.5
Wooden floors on concrete slabs
125
14.6
The crawl space paradox
125
14.7
High relative humidity indoors
126
14.8
Water leakage
126
15
APPENDIX I. DATA FOR CALCULATIONS
127
Table A1 Thermophysical properties of materials.
127
Table A2. Hygric properties of materials.
129
Table A:4 Water vapor content at saturation, vsat in g/m3. The table is calculated by the
approximate formula (13:13).
131
Table A:5 Climate examples. Monthly averages for two locations.
132
Table A:6 Sorption curves for some materials
132
16
EXAMPLES FOR MOISTURE TRANSFER
135
16.1
Air humidity
135
16.2
Taking a shower in Trondheim
135
16.3
Moisture distribution in a multilayer construction
137
16.4
Capillary action
139
16.5
Ventilated air gaps
141

2005-10-24
-------------------------------------------------------------------------------------------------------------------
1 List of Symbols
Area
Thermal diffusivity for a material
Solar asimut
Apsorptance for solar radiation
Complex matrix element
m2
Absorptance for long wave radiation

m2K /W
B0
c
C
C
Volume expansion coefficient

horizontal angle
Specific permeability of the material
Specific heat capacity, capacitivity
Thermal capacity
K-1
rad
m2
J/kgK
J/K
W/m2K
A
a
a
a
A
Point thermal transmittance of a thermal bridge

CCF Cloud cover factor
D
m2/s
-
W/K
-
Declination from a plane through equator

Energy
rad
W/m2K
J
Emissivity
F
Fab
Heat flow rate

Configuration factor or view factor from a to b
J/s
W/m2K
rad
F
F1
F2
G
g
Gr
h
H
h
H
I
i
The latitude of a location

Total window transmittance for solar radiation
Partial window transmittance (Short-wave radiation)
Partial window transmittance (Surface heat flow)
Acceleration of gravity, 9.81
Grashof number
Coefficient of surface heat transfer
solar altitude
Height above a reference level,
Dynamic viscosity of a fluid
Solar radiation
Angle of incidence
Temperature
E
E
W/m2K
m/s2
W/m2K
W/m2K
rad
m
Ns/m2
W/m2
rad
C
10

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Heat production per unit volume
The total radiosity from an opaque surface
W/ m3
W/m2
Dynamic length coefficient

rad1/2/m
l
l
L
f
m
M
M
Nu
p
Pr
Q
q
R
r
R
Re
Thermal conductivity
W/mK
Thermal conductance
Alt: s
Characteristic length
Length
Air flow
Friction coefficient.
Mass
Total excitance of a real surface
Cloudiness factor
Form factor
Total hemispheral excitance, black body
W/K
m
m
m3/s
kg
W/m2
W/m2
m2/s
Kinematic viscosity
Nusselt number
Static pressure,
Prandtl number
Heat, energy
Density of heat flow rate
Density
Thermal resistance
m2K/W
ground reflectance
Gas constant, 8314.3
Reynolds number
Pa
J
2
W/m
kg/m3
J/(kMol.K)
-
Stefan Boltzmann constant = 5.67 10-8

W/m2K4.
Temperature
t
U
u
V
v
x
reflectance
Day number
Thermal transmittance by area
Air velocity
Volume
wall asimute
Length coordinate
Pressure loss coefficient
Length coordinate
W/m2K
m/s
m3
m
m
Linear thermal transmittance of a thermal bridge
W/mK
11

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Length coordinate
12

2005-10-24
-------------------------------------------------------------------------------------------------------------------
1.1
a
c
d
D
e
equ
H
H
i
N
r
s
se
si
V
Indexes
Ambient
Convection (often includes conduction)
Diffuse
Direct
Exterior
Equivalent
Horizontal
Hydraulic
Inner, interior
Normal
Radiation
Surface
Exterior surface
Inner surface
Vertical
13

2005-10-24
-------------------------------------------------------------------------------------------------------------------
14

2005-10-24
-------------------------------------------------------------------------------------------------------------------
2 Building physics in civil engineering

Building physics is the science of climate protection in buildings and in built
structures. We build and heat houses to create a comfortable indoor climate,
which can be maintained within specified limits, regardless of the variations
in outdoor climate. To gain this we need structures that can withstand the
external and internal forces of wind, snow and life loads but also to provide
the structures with qualities such as thermal insulation, air and rain
tightness, solar radiation control, sound insulation and biological protection.
Furthermore, our task is not only to provide for a good indoor environment,
but also to provide for an environment within the building structures and
for the building materials that does not enhance the decay of the structures
due to corrosion, mould growth and rot, cracking due to thermal or humidity
related stresses and so on. This even applies to constructions that are not
parts of a climatic shield such as bridge and road constructions etc. where
the decay and lost performance in many cases is more due to environmental
factors than to the loads over time. Thus the construction of a built
structure for optimum performance and durability has to be based on
extensive knowledge on both structural mechanics and building physics.
2.1
What building physics is mainly about
When we describe the physical state within a building or a building

structure we usually refer to the temperature, the air pressure and the
moisture content. By moisture we mean water in different phases. The
physical state can either be the result of our observations and calculations
or give the initial and boundary conditions for further calculations. The
combination of temperature and moisture content can for instance give us
the risk estimate for fungi growth and rot in wooden construction parts and
the distribution of temperatures in a rigid construction part can be used to
estimate thermal stresses. We are also interested in how heat, air and
moisture are transferred within our structures and systems. The potentials
for these processes can be expressed as the physical parameters, the
temperature, the air pressure and the moisture content respectively.
2.2
Examples of heat, air and moisture transfer
A common example of the heat transfer process is the heat flow from the
warm indoor environment through an insulated wall to the cold outdoor
environment. This can involve heat conduction through parallel material
layers as well as the treatment of two or three-dimensional material layers
15

2005-10-24
-------------------------------------------------------------------------------------------------------------------
as well as the influence of air movements in the construction.

During wintertime the warm indoor air has lower density than the cold
surrounding outdoor air. Thus by Archimedes law, the lightweight indoor
air creates a force on the inside of the roof and the upper parts of the
building relating to the weight difference between the outdoor and indoor
air creating a pressure difference across the roof construction. The pressure
difference in its turn will generate air flow through the roof through
different leakage paths which may consist of air gaps, holes, cracks and
porous materials.
The drying out of a fresh concrete floor deck is a moisture transfer process
governed by the high moisture content within the floor slab and the lower
moisture content at the surface of the floor slab, which is practically in
equilibrium with the surrounding dry air.
2.3
Combined processes
Even if this course gives the transfer processes as separate chapters the
processes we deal with in reality usually are combinations of these
processes. As discussed above a temperature difference generates a pressure
difference, which generates air movement, which in turn may affect the
temperature difference. When moisture from the air condenses on a cold
surface, heat is released and the temperature of the surface rises. Since we
are dealing with processes that can be non-linear, the complexity of the
general solution of a problem has often been beyond what can be treated in
normal engineering work and the problems have often been studied in an
oversimplified way.
2.4
New mathematical tools make life easier
With the new mathematical tools

available for personal computers
such as the latest versions of
Mathcad, Maple and Matlab the
application of building physics
has come into a new and
revolutionary era. If the user
knows the governing transfer
equations, boundary conditions
and construction and material
parameters and links the
transfer equations into balances that provide for the conservation of energy
and mass, the solutions will be given by the computer. The transfer
16

2005-10-24
-------------------------------------------------------------------------------------------------------------------
equations can be non-continuous and the transfer parameters can be nonlinear i.e. varying with the physical state such as the temperature or the
moisture content.
2.5
Building physics and the environment
When addressing the environmental issue,

important parameters are the performance and the
durability of the construction. One cubic meter of
insulation in a well performing insulated
construction may under its lifetime reduce the
heating demand for the building with 5 cubic meters
of oil compared to a poorly insulated construction.
The main environmental issue is therefore not only
the environmental qualities of the materials used, but also that the
materials used will serve their purpose in an optimum way, regarding
technical and economical as well as environmental factors. With increased
durability and service life of the construction, the environmental impact per
year of use is reduced. Bad design of insulated constructions also creates
favorable conditions for biological growth which may endanger indoor air
quality and human health and in general have negative consequences for
the indoor environment and comfort.
17

2005-10-24
-------------------------------------------------------------------------------------------------------------------
18

2005-10-24
-------------------------------------------------------------------------------------------------------------------
3 Introduction to Building Heat Transfer

3.1
3.1.1
The basic elements of building heat transfer
Temperatures
0 C
273.15 K
Temperatures are in this text either expressed as the thermodynamic
temperature T, K, or the Celsius temperature , C. They are related as

follows.
= T-273.15
T = +273.15
(3.1)
(3.2)
Your thermodynamic body temperature is normally 310.15 K and can rise

up to about 315 K when you get very ill.
3.1.2
Heat - a form of energy
For a certain amount of energy stored or released in the form of heat we use
the quantity Q, J (joules). Before the unit for heat was calorie, which is the
quantity of heat needed to heat one gram of water by one degree on the
Celsius scale. One calorie is 4.184 J. In American literature the unit BTU
(British thermal unit) is still used. One BTU is 1054.35026448 J, to be
exact.
For the amount of heat produced or transferred per time unit we use the
term heat-flow rate , W = J/s. The unit J/s is also called watt.
The following processes can release approximately equal amount of heat per
time unit:
- Electrical radiator of 500 watts
- Burning 0.05 liter of oil per hour
- Burning 0.3 kg of wood per hour
2
- Solar radiation absorbed on a tilted surface of 0.5 m around noon on a
clear day in June in Stockholm
- 4 persons working in a factory
- 2 cows at rest.
19

2005-10-24
-------------------------------------------------------------------------------------------------------------------
3.1.3
Heat storage - heat capacity
If, for a body or a whole system, the temperature is raised by dT as a result

of adding a small quantity of heat dQ then the heat capacity is defined as
C = dQ/dT
J/K
(3.3)
If the body is made of homogenous material the specific heat capacity c is

defined as heat capacity divided by the mass m
c = C/m
J/(kg.K)
(3.4)
or with known density , kg/m3, and volume, V, m3

c = C/(V.)
J/m3K
(3.5)
Correspondingly the heat capacity for a given volume of material with

specific heat capacity c is given by
C = V. . c
J/K
(3.6)
In literature on thermodynamics the product .c is often expressed with the

letter q, J/Km3. It is important to note this since the same letter is also
used for the density of heat flow rate i.e. heat flow rate per unit area.
The amount of heat stored in a body or a system with heat capacity C,
always has to be expressed in relation to some reference temperature Tref
to which the system is supposed to be cooled during the process of utilizing
the stored heat.
The following systems can store approximately equal quantity of heat, or
energy, which can be transformed to heat.
3
- 1 m water tank at 50 K above reference

temperature
- 100 m2 concrete deck, 0.2 m thick at 5 K above
reference temperature
- 200 m3 of water stored at 10 m above reference
level
- 70 ordinary car batteries
- 5 liters of gasoline
20

2005-10-24
-------------------------------------------------------------------------------------------------------------------
3.1.4
Heat conduction
If we have two systems or bodies at different temperatures T1 and T2 which

are in some way thermally connected, heat will flow from the warmer to the
colder
= (T1 - T2).,
(3.7)
, (Big lambda), is the thermal conductance, W/K.

If within a body of an isotropic material there exists a temperature
gradient, grad T, the density of heat flow rate q can be calculated as
q = - . grad T, W/m2
(3.8)
This is often referred to as
Fouriers law.
grad T
is the thermal conductivity of

the material, W/mK.
T(x,y,z) = konstant
If T only depends on x, equ (3.8)

becomes
(3.9)
TK
dT
The heat flow rate through a

surface with area A given by
x = constant with a uniform
temperature gradient then
becomes
dx
lm
= A
T
x
The thermal properties of a material are highly dependent on the structure

and density of the material. What is referred to as the thermal conductivity
of a porous material is often a combination of conduction, radiation
exchange, convection and conduction in water in the pore structure.
Examples of thermal and hygric properties of building materials are given
in Appendix I.
21

2005-10-24
-------------------------------------------------------------------------------------------------------------------
3.2
Steady state calculations
In heat transfer calculations it is convenient to make a distinction between

steady state and non-steady state heat transfer. Steady state means that
the temperatures of the system do not vary with time. From the definition of
the heat capacity
dQ = C dT
(3.11)
it is evident that in steady state heat is not being stored in or removed from
any part of the system since this implies a change in temperature.
For a homogenous wall slab, with a temperature gradient in the direction
normal to the surface, the consequence is, that if no heat is being stored at
any point in the wall, the temperature gradient has to be constant. This also
implies that the temperature is linearly distributed between the surfaces.
dT/dx = constant = (T2-T1)/d
T2
(3.12)
The density of heat flow rate q, W/m2, can

accordingly be expressed as
T1
q =
(T2 T1 )
= (T1 T2 )
d
d
(3.13)
This can also be written as
(T T )
q= 1 2
R
R=
W/m2
(3.14)
(3.15)
m K/W
is known as the thermal resistance of the wall slab.
22

2005-10-24
-------------------------------------------------------------------------------------------------------------------
3.2.1
Steady state heat flow and temperature distribution in a multilayer wall

with no internal heat sources.
qn
q1
R1
q=
From the steady state

condition it follows that
the heat flow is constant
through the construction
Tn
Tn-1
T2
T1
R2
q1 = q2 = q3 = q4 =. . . .
.. = qn
Rn-1
Tn )
(T T ) (T T )
(T
3 = .......... = n - 1
1
2 = 2
R
R
R
1
2
n -1
(3.17)
or
(T1-T2)=q.R1 , (Tk-Tk+1)=q.Rk
n 1
n 1
k =1
k =1
(3.18)
n 1
(T1 Tn ) = (Tk Tk + 1 ) = (q R k ) = q (R k )
R tot =
q=
n 1
(R
k =1
(3.19)
k =1
(3.20)
T1 Tn
Rtot
(3.21)
Assume that we have a known temperature T1 on the left side of a

construction and the heat flow q through the construction. The temperature
difference across a layer k is given by
(Tk-Tk+1) = q.Rk
(3.22)
Obviously the temperature in on the left boundary of a layer k can then be

calculated as the sum of the temperature on the left side of the construction
and the temperature differences across all layers on the left side of the
boundary
23

2005-10-24
------------------------------------------------------------------------------------------------------------------k 1
k 1
j =1
j =1
Tk = T1 q R j = T1 q R j
q=
nk>1
(3.23-24)
T1 Tn
Rtot
(3.25)
k 1
Tk = T1
3.2.2
(T1 Tn ) R j
j =1
Surface resistances
Ts
q = hs
(3.26)
R tot
The heat transfer from a construction

surface to the surroundings with a
given temperature is taking place by
radiation to surrounding surfaces and
due to heat conduction and air
T a movements close to the surface as will
be treated in coming lectures. The
coefficient of surface heat transfer hs ,
2
W/m K, is defined as
(Ts - Ta )
(3.27)
In practical applications, these complicated processes are often

approximated by a fictive material layer between the surface, Ts, and an
ambient temperature, Ta, which is often chosen as the air temperature.
q=
(Ts - Ta )
Rs
(3.28)
For simplified calculations in normal building applications these fictive

resistances can be chosen on the inside towards a heated room with normal
indoor climate
Rsi = 0.13 m2K/W
(3.29)
and on the outside towards average external temperature and wind

conditions
24

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Rse = 0.04 m2K/W
(3.30)
In reality however, the surface resistances vary greatly, due to

miscellaneous factors, which will be further discussed in the lectures on
radiation and convection in buildings.
3.2.3
Definition of the U-value
The thermal transmittance or U-value, W/m2K, for a construction is defined

as the ratio between the density of heat flow rate q, W/m2, through the
construction and the temperature difference between the ambient
temperatures on both sides
U=
q
(T T )
i
e
(3.31)
For a construction with n layers the U-value then becomes
U=
(3.32)
R si + R j + R se
j=1
3.3 Two-dimensional heat conduction

We have so far learned that
1. The relation between the temperature gradient, thermal conductivity of a
material and the heat flow rate known as Fouriers law.
2. The relation between a quantity of heat added to a given volume, the
density and specific heat capacity of a given material and the resulting
increase in temperature.
We are going to use these relations to establish a differential equality
between the heat flow to an infinitesimal volume during an infinitesimal
period of time and the increase in temperature. The aim is to establish a
differential equation which describes the heat conduction process in a
homogenous material
Assume a homogeneous material with material properties given as thermal
conductivity , density and specific heat capacity c.
25

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The temperature field within the material is given as

T=T(x,y,z,t)
(3.33)
the heat flow rate at each point (x,y,z) is given as
q = - grad T
(3.34)
The heat production per unit volume is given as
= (x,y,z,t)
(W/m3)
(3.35)
The heat production can for instance be due to a chemical reaction as in

concrete being cured or due to absorbed solar radiation in a transparent
insulation material. Let us look at a small element in Cartesian
coordinates. Assume for simplicity that all variables are constant in the z
direction that is
dT/dz = 0 and dz = 1
(3.36)
= (x,y,t)
(3.37)
The regarded element has the
dimensions dx and dy and
one unit length in the z
direction. Since the
temperature gradient in the z
direction equals zero the heat
exchange between the element
and the surrounding material
goes through surfaces 1 to 4
and the areas of the sides of
the elements are given by dx
y+dy
c
1
4
x
x+dx
and dy respectively
During a small time step dt the quantity of heat added to the volume is
dQ = dt (1 - 2 - 3 + 4 + .dx.dy)
26
(3.38)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
By using Fouriers law we can express the heat flow rates
1
2
3
4
= - dy (dT/dx)x
(3.39)
= - dx (dT/dy)y+dy
(3.40)
= - dy (dT/dx)x+dx
(3.41)
= - dx (dT/dy)y
(3.42)
Expansion by using the first two terms of Taylor

(dT/dx)x+dx = dT/dx +(d2T/dx2) xdx
(3.43)
(dT/dy)y+dy = dT/dy +(d2T/dy2) ydy
(3.44)
and substituting into equation (3.38) we get

dT d 2 T
dT d 2 T
dT
dT
.
.
+ dx
+ ( 2 )dy + dy
+ ( 2 )dx - dx
dQ = dt - dy
+ dx dy
dx
dy
dy
dx
dy
dx
(3.45)

d 2T
d 2T
dQ = dt dxdy ( 2 ) + dxdy ( 2 ) + . dx . dy
dy
dx
(3.46)
From the definition of thermal capacity

dQ=.c.dx.dy.dT
(3.47)
T
d 2T
)
+
dxdy
(
) ) +.dx.dy)
dy 2
dx 2
.c.dxdy. .dT = dt ((dxdy ( d
dT d 2 T
d 2T
(
)
(
=
+
)+
dt c dy 2
dx 2 c
(3.48)
(3.49)
The thermal diffusivity a for a material is defined as

a= /(.c)
m /s
(3.50)
and if dt, dx and dy are made infinitesimally small
27

2005-10-24
-------------------------------------------------------------------------------------------------------------------
T
2T 2T
= a( 2 + 2 ) + .
t
x
y
c
(3.51)
which is the general equation for heat conduction or in more general

mathematical terms
T/t= aT+
.c
= a(T+/)
(3.52)
Equation (3.52) is valid in three dimensions as well.
28

2005-10-24
-------------------------------------------------------------------------------------------------------------------
4 Heat Conduction Equation Analytical

Solutions
The conduction equation derived in chapter 3 in its general form is not
easily solved analytically. The choice between numerical and analytical
solutions is generally made on the basis of the relative cost effectiveness of
the method chosen. The increasing availability of computers has lead into
the direction of more frequent use of numerical methods. However new
mathematical programs for personal computers now provide convenient
tools for analytical or hybrid solutions.
4.1
Simplifications
By stating some limitations the equation can be simplified to a form where

trivial analytical solutions can be found.
4.1.1
No heat production
Generation of heat within building materials is in most cases not relevant.

The generation of heat can then be put equal to zero.
=0
(4.1)
The equation then becomes
T/t = aT
4.1.2
(4.2)
No heat production with one dimensional heat flow
If we further assume that the temperatures are constant in the y and z

directions it follows that
T/y =T/z = 0
(4.3)
and the equation is reduced to
T/t = a.2T/x2
(4.4)
29

2005-10-24
-------------------------------------------------------------------------------------------------------------------
4.2
Solution for a one dimensional slab with harmonic boundary temperatures
T
q
0
0
x=0
A temperature variation that can be expressed as a sinus or cosinus

function with time is called harmonic. By limiting the temperature
variations at the boundaries to harmonic functions we can make use of their
important properties that the derivative is equal to the function multiplied
with a complex constant. Since heat conduction in building materials at
normal temperature levels can be regarded as a linear process, all
temperature variations within the system will be harmonic as well, but with
different amplitude and phase lag. Since non-harmonic functions can be
transformed to Fourier series, i.e. a sum of harmonic functions, problems
with arbitrary boundary conditions can be solved in this way. A harmonic
function always has the mean value zero. A rational technique is
- to solve the problems for the mean boundary conditions as a steady state
problem
- do the Fourier transform on the variations of the boundary conditions
around the mean value
- solve the response for the different frequency components as shown below
- get the final result by superposition.
An important precondition for rational calculation work is that a harmonic
function at a given frequency can be expressed as a complex number
relating the actual function to a basic oscillation.
30

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The figure below shows how a harmonic oscillation can be expressed in the
complex plane.
Tamp
t +
u
0 +
t
T = Tamp.ei(t+) =(u+i v) .eit
(4.5)
The time derivative of T then becomes
T/t = iT
(4.6)
And the heat equation can be written as

iT = a(2T/x2)
(4.7)
having a solution in the form

T= Csinh((1+i)x) + Dcosh((1+i)x)
(4.8)
(4.9)
2a
This can be verified looking at the properties of the hyperbolic function and
the complex i.
sinh(u)=(eu - e-u)/2
i=
1
2
cosh(u)=(eu + e-u)/2
(1 + i)
(4.10)
(4.11)
The properties of the basic functions at x=0

31

2005-10-24
-------------------------------------------------------------------------------------------------------------------
sinh(0)=0
cosh(0)=1
(4.12)
together with a known boundary temperature at x = 0.

T=T(0)=T0
(4.13)
gives the unknown coefficient D
D=T
(4.14)
If we now apply Fouriers law we get the equation for the heat flow
q(0)=(-dT/dx)x=0=-C(1+i)cosh((1+i).0) + ........ sinh(0)
(4.15)
which gives C in terms of the heat flow
C=(-q0)/((1+i))
(4.16)
and we can express the temperature at any location x in terms of the

temperature and the heat flow at the surface.
T ( x ) = T0 cosh ((1 + i ) x ) q 0 .
.
sinh ((1 + i ) x)
(1 + i )
(4.17)
An Important simplification is when we have a semi infinite body with x=0

at the surface. At an infinite distance from the surface the temperature
variations at the surface have vanished, or
x then T(x) 0
q0
(4.18)
(1 + i) T0 cosh((1 + i) x)
sinh((1 + i) x
(4.19)
x Then sinh(x) cosh(x)
(see equ. 4.10)
(4.20)
and we can now for the semi-infinite body establish a relation between
temperature and heat flow at the surface.
q = (1+i).T
0
0
(4.21)
And by substitution of equ. (4.21) into equ (4.17) we get

T(x)=T0{cosh((1+i).x)-sinh((1+i).x)}
32
(4.22)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
T(x)=T0e-(1+i)x
(4.23)
Expression (4.23) can as an example, be used to estimate the amplitude of

the annual temperature variation as a function of the depth from the
surface. From equation (4.17) the heat flows at the boundaries of a finite
material layer can be related to the boundary temperatures. This can be
formalized as a matrix equation relating the temperatures and heat flows at
the boundaries by a two by two complex matrix.
~ A
q~1 = C1
1 1
B ~
1 0
D q~
1 0
(4.24)
A 1 = cosh((1+ i)1d1)
B1 =
(4.25)
sinh((1+ i) 1d1 )
1(1 + i)1
(4.26)
C1 = 1(1 + i)1 sinh((1+ i) 1d1 )
(4.27)
D1 = cosh((1+ i) 1d1 )
(4.28)
4.2.1
Solution for a multilayer construction
qso
qsi
so
si
~si A
q~ = C
si
~
B v~
~so
D q
so
A resulting matrix for a

multilayer construction with n
layers can then be calculated by
a simple multiplication of the
2x2 matrices for the single
layers. A matrix relation
between heat flows and
temperatures on both surfaces
can in this way be established.
(4.29)
33

2005-10-24
-------------------------------------------------------------------------------------------------------------------
A
C
B A
= n
D C n
B n A n 1 B n 1 A 1
....
D n C n 1 D n 1 C1
B1
D1
(4.30)
Purely resistive layers such as surface layers and air gaps can represented
by a matrix as well in order to calculate the heat flows at the ambient
temperatures.
B k 1 R k
=
D k 0
1
A k
C
k
(4.31)
Equ (4.29) can be rearranged to give the heat flow variations as a function
of the temperatures on both sides
q~si E
q~ = G
so
F ~si
H ~so
E=D/B
4.3
(4.32)
F=C-DA/B
G=1/B
H=-A/B
(4.33)
Steady state heat flow - simplifications
By steady state heat flow we mean that the boundary conditions and
generated heat are constant with time or
T/t = 0
(4.34)
The general equation for heat conduction is then reduced to
T + / = 0
(4.35)
This equation is for instance useful when calculating the resulting

temperatures in newly cast concrete slabs during the curing process.
4.3.1
Steady state one dimensional heat flow with heat generation
If we further assume that the temperature is constant in the y and z

directions we get
2T/x2 + / = 0
(4.36)
34

2005-10-24
-------------------------------------------------------------------------------------------------------------------
This equation can as an example be applied to a cooling fin or a relatively

thin wall with thickness d, high thermal conductivity and linear surface
heat transfer coefficient hs = 1/Rs on both sides.
h
0
d
a
x=0
x=L
The heat production can be expressed as the heat flow from the ambient
temperature to the cooling fin per volume
(x) = 2.hs.(Ta-T(x))/d
(4.37)
2T(x)/x2 - 2.hs.(T(x)-Ta)/(d.) = 0
(4.38)
By expressing
(x)
=T(x) - Ta and
hs . 2
d .
(4.39)
and making the substitution, equ (4.38) becomes

2/x2 - 2 = 0
(4.40)
having a solution in the form
= Ce
+ De
(4.41)
We assume that at x=0 the temperature is from where it follows that
= C + D
which gives C = - D
35
(4.42)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
and, that the heat flow in the x direction at the farther end in x=L is equal
to 0 which gives the other condition
L
L
L
L
( x )L = Ce + De = (0 D)e + De = 0
De
bL
+ De
bL
= 0 e
bL
(4.43)
(4.44)
Rearranging we get
D = 0
e bL
bL
bL
e +e
(4.45)
e L
e L
)
=
(
)
0
e L + e L
e L + eL
(4.46)
= 0
e L
e L
x
x
e
+
e
0
L
L
L
L
(e + e )
(e + e )
(4.47)
= 0
e L e x + e L ex
(e L + e L )
(4.48)
C = 0 (1
When x <
L
this expression can with acceptable accuracy be given as
3
= 0 e x
(4.49)
If we assume that the wall has a unit length in the direction perpendicular
to the paper the heat flow from the base to the thin wall at x=0 can be
expressed as
= d( x)0
(4.50)
e L e L
= d0
L + e L
(4.51)
= d0 tanh(L)
(4.52)
In building physics equation (4.52) is useful to study the heat transfer along
surface layers and attached wall or floor slabs toward a thermal bridge in
an insulated wall. At values for L above 2.0, tanh(L) approaches 1 and
equation (4.52) approaches
36

2005-10-24
-------------------------------------------------------------------------------------------------------------------
= d 0
4.3.2
(4.53)
One dimensional steady state heat flow without heat sources
Since the temperatures are constant in time and in the y and z direction the
general equation for heat conduction reduces to
d2T(x)/dx2 = 0
(4.54)
Integrating on both sides

dT(x)/dx = konstant
(4.55)
and we see that the trivial solution is

T
T(x) = T1+(x/d)(T2-T1)
See chapter 3.
1
d
37
(4.56)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
38

2005-10-24
-------------------------------------------------------------------------------------------------------------------
5 Numerical Methods
5.1
The control element - two dimensional heat flow
We are going to us the same relations as in chapter 3 to establish a

differential equality between the heat flow to an infinitesimal volume
during an infinitesimal period of time and the increase in temperature. The
aim is to establish a difference equation, which describes the heat
conduction process in a homogenous material, on the border between two
materials and on the surface of a material where we have convective and
radiative heat transfer to an ambient temperature.
The temperature field within the material is given as

T=T(x,y,z,t)
(5.1)
the heat flow rate at each point (x,y,z) is given as

q = - grad T
(5.2)
The heat production per unit volume is given as
= (x,y,z,t)
(W/m3)
(5.3)
The heat production can as an example be due to a chemical reaction as in

concrete being cured or due to absorbed solar radiation in a transparent
insulation material.
We now look at a small element in Cartesian coordinates. Assume for
simplicity that all variables are constant in the z direction that is
dT/dz = 0
(5.4)
= (x,y,t)
(5.5)
39

2005-10-24
-------------------------------------------------------------------------------------------------------------------
dx k-1
dy
dx k
T
k,j-1
j-1
2
Tk-1,j
dy j
dx k+1
3 T
k+1,j
Tk,j
4
dy
T
k,j+1
j+1
y
The material field is divided into rectangular elements k,j that have the
dimensions dxk and dyj and a unit length in the z direction. Each element
can have material properties given as thermal conductivity, , density, ,
and specific heat capacity, c. Each element is represented by its
temperature at a nodal point in the center of the element.
Since the temperature gradient in the z direction equals zero the heat
exchange between the element and the surrounding material goes through
surfaces 1 to 4.
During a time step dt the quantity of heat added to the volume is
dQ = dt (1 + 2 + 3 + 4 + .dx.dy)
(5.6)
Between each adjacent pair of nodal points we have a two layer construction
with dimensions given by dx and dy. The heat flow between two adjacent
nodal points can be calculated according to lecture 1.
40

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Tk-1
Tk
dx k-1
2
dy
dx k
2
If all the heat flows are defined as directed towards the element k,j we get
1 = k-1,k
dy
(5.7)
2 = j-1,j
3 = k+1,k
(Tk-1 - Tk,j)
dx k
(Tj-1 - Tk,j)
dy j 1 dy j
+
2 j 1 2 j
dy j
=
(Tk+1 - Tk,j)
dx k +1 dx k
+
2 k +1 2 k
(5.9)
4 = j+1,j
dx k 1 dx k
+
2 k 1 2 k
dx k
(Tj+1 - Tk,j)
dy j+1 dy j
+
2 j +1 2 j
(5.8)
(5.10)
We define the conductance s between adjacent elements as
k-1,k = sk-1,k(Tk-1 - Tk,j)

j-1,j = sj-1,j(Tj-1 - Tk,j)
k+1,k = sk+1,k (Tk+1 - Tk,j)
(5.11)
(5.12)
(5.13)
j+1,j
= sj+1,j (Tj+1 - Tk,j)
sk-1,k =
dy
(5.14)
(5.15)
dx k 1 dx k
+
2 k 1 2 k
41

2005-10-24
-------------------------------------------------------------------------------------------------------------------
dx k
dy j 1 dy j
+
2 j 1 2 j
sj-1,j =
(5.16)
sk+1,k =
sj+1,j =
dy
(5.17)
dx k +1 dx k
+
2 k +1 2 k
dx k
dy j+1 dy j
+
2 j +1 2 j
(5.18)
5.2
Expression of boundary conditions
The element can be next to a surface with a given boundary temperature.

The conductance is then calculated as
sk+1,k =
dy j
dx k
2 k
(5.19)
The element can be next to a surface with a given surface coefficient of heat
transfer hs and a known ambient temperature. The conductance is then
calculated as
sk+1,k =
dy j
1 dxk
+
hs 2 k
(5.20)
The element can be next to an adiabatic boundary due to symmetry or

extreme insulation. No heat flow can take place across an adiabatic
boundary. It follows that
sk+1,k = 0
5.3
(5.21)
Resulting heat balance
The quantity of heat Q added to an element k,j during a time step dt can
thus be expressed
42

2005-10-24
-------------------------------------------------------------------------------------------------------------------
dQ = dt { sk-1,k(Tk-1 - Tk,j) +sj-1,j(Tj-1 - Tk,j) + sk+1,k (Tk+1 - Tk,j)
+sj+1,j(Tj+1 - Tk,j)+ .dx.dy} = C.dT

(5.22)
C = cdxk,jdyk,j
(5.23)
The solution for the temperature in k,j at the end of a time step t+dt can
now be solved explicitly assuming that all temperatures at the time t are
known
Tk,j(t+dt) =Tk,j + dT= Tk,j+(dt/C){sk-1,k(Tk-1 - Tk,j) +sj-1,j(Tj-1 - Tk,j)
+ sk+1,k (Tk+1 - Tk,j)+sj+1,j(Tj+1 - Tk,j)+ .dx.dy}
(5.24)
When solving the variations in the temperature field with time we in many
cases do not have the starting values. If we know the starting values for all
temperatures the solution can be calculated correctly from the beginning. If
we do not know the initial temperatures we have to carry out the calculation
over a sufficiently long period with the prevailing boundary conditions, till
the calculated solution has converged to the correct solution.
5.4
Choice of time steps and element size - stability
The choice of the calculation time step dt is made to give a proper

representation of the boundary condition in time and to ensure the stability
of the solution with time. For the representation of normal daily variations
of temperature a satisfactory time step can be about one hour or 3600 s. The
maximum time step to avoid instability in equation (5.24) is for each cell
given by the thermal capacity of the cells divided by the sum of the
conductances to the adjacent cells
dtmax = Ck,j/( sk-1,k + sj-1,j + sk+1,k +sj+1,j )
(5.25)
With larger elements the stability time step is increased but since the
resulting solution is linearly varying between the nodes the inaccuracy is
increased. It is difficult to give a general recommendation for the choice of
element sizes but as a rule of thumb for the calculation of ordinary building
constructions the element size can be chosen 20 - 30 mm where larger
temperature gradients are expected and enlarged up to 5 to 10 times where
only small gradients are expected.
43

2005-10-24
-------------------------------------------------------------------------------------------------------------------
5.5
Simplifications for special cases
No generation of heat within the construction =0

Tk,j(t+dt)=Tk,j+(dt/C){sk-1,k(Tk-1 - Tk,j) +sj-1,j(Tj-1 - Tk,j) +
sk+1,k(Tk+1 - Tk,j)+sj+1,j(Tj+1 - Tk,j)}
(5.26)
One dimensional heat flow, T independent of j

Tk(t+dt)=Tk+(dt/C){sk-1,k(Tk-1 - Tk) + sk+1,k (Tk+1 - Tk))}
(5.27)
Steady state two dimensional heat flow, Temperature independent of the
time t.
0={sk-1,k(Tk-1 - Tk,j) +sj-1,j(Tj-1 - Tk,j) + sk+1,k (Tk+1 - Tk,j) +sj+1,j(Tj+1 - Tk,j)}
(5.28)
sk-1,kTk-1+sj-1,jTj-1+ sk+1,k Tk+1+sj+1,jTj+1 =Tk,j(sk-1,k +sj-1,j + sk+1,k +sj+1,j)
(5.29)
Tk, j =
T
T
+s
T
+s
T
+s
j + 1, j j + 1
k 1, k k 1 j 1, j j 1 k + 1,k k + 1
sk 1 + s j 1, j + s k + 1 + s j + 1, j
,k
,k
(5.30)
In this way each element has a balance equation expressing the

temperature in terms of the temperatures for the adjacent elements or
boundaries. This generates a system of N equations which can be solved
simultaneously where N is the number of cells. A common way of solving
two or three dimensional temperature fields is by iteration where a
calculation is started with the actual boundary condition but with arbitrary
initial temperatures in the cells. The temperatures are then calculated
repeatedly using equation (5.30) till the temperature field converges.
44

2005-10-24
-------------------------------------------------------------------------------------------------------------------
5.6
Finite element methods
Above formulations for the solution of multidimensional and non steady

state heat transfer problems are given. Finite difference solutions are today
widely used by experts and researchers due to the simplicity of the approach
and flexibility in combining different physical processes. An alternative
approach is the Finite element method (FEM) which also is a numerical
technique for solution of boundary-value problems. It was first developed in
the forties with the first applications within structural mechanics. The body
or system is divided into discrete regions, the so called finite elements. The
process within the elements and the constraints at element boundaries give
rise to a system of simultaneous equations which in turn can be solved with
known techniques from linear algebra or non linear solution techniques. In
this sense one could look at finite difference solutions as a limited subset of
the finite element technology.
The computer program FEMLAB that is introduced in the exercises on heat
transfer within the course is a general platform for the solution of various
kinds of differential equations and boundary value problems. It also
provides the possibility to combine different physical processes within the
same calculation. The program has an advanced pre-processor for
generation of geometries, material properties and boundaries and postprocessors for graphical display of the results. The generation of geometries
is similar to a normal CAD environment and geometries can also be
imported from CAD programs. The main practical difference compared to
common finite difference programs for two or three dimensional thermal
calculations is that the geometries are not limited to rectangular objects as
the elements can have triangular forms.
45

2005-10-24
-------------------------------------------------------------------------------------------------------------------
46

2005-10-24
-------------------------------------------------------------------------------------------------------------------
6 Convective Heat Transfer

For excursions in arctic or mountain climate, people take with them a table
that shows the equivalent cooling temperature based on the actual wind
speed. This is due to the fact that the cooling of the skin is not only taking
place by heat conduction from the skin to the ambient air but also enhanced
by air movements at the skin surface or, as we also could express it, by
convection.
Important applications of convective heat transfer in buildings are for
instance radiators heating the air, heating or cooling of air flowing in ducts,
air movements and their influence of human comfort, air leakage in the
building envelope leading to extra heat loss, air movements in cavities and
porous materials within insulated building constructions and heat transfer
between building surfaces and ambient air.
6.1
Basics of convective heat transfer
V1
Heat flow due to air flow between

two systems at temperatures T1 and
T2 can be expressed in terms of the
mass flow and temperature
difference. If a volume dV1 is
removed from system 1 and added to
system 2 and the same mass of air
from system 2 at temperature T2 is
added to system 1 the resulting
quantity of heat removed from
V2
system 1 is for constant c given by

dQ1= -dV1.1.c(T1-T2)
(6.1)
and the resulting quantity of heat added to system 2 is given by

dQ2= dV1.1.c(T1-T2) = dV2.2.c(T1-T2)
(6.2)
If we have a continuous flow of air between the systems

L=dV1/dt
(6.3)
the heat flow rate from system 1 to system 2 due to air flow is
47

2005-10-24
-------------------------------------------------------------------------------------------------------------------
12=L.1.c(T1-T2)
6.1.1
(6.4)
Convective surface heat transfer
An important application of convective heat transfer is the heat transfer

between the boundary surfaces and the air within a given volume. If the air
in the volume is standing still, heat is transferred with conduction in the air
only, but at higher temperature differences between the surface and the air
the density differences will generate air movements and the heat transfer
will be a combined process of conduction and convection. When referring to
convective surface heat transfer we usually mean the combined process
including conduction in the air as well.
6.1.2
Natural convection
Air movements are generated due to density

differences. Density differences usually occur as a
result of different temperatures as can be seen in
table 6.1. Warmer air will always tend to rise
upwards and cooled air will sink downwards tills an
T + equilibrium is gained. If heat is constantly supplied
Tor removed at different surfaces of a volume like
shown in the figure this will lead to continuous
rotation of the air. Examples of this can be seen in
rooms heated by a radiator with cooling at the inner
surfaces of the exterior walls and in vertical air gaps in multiple glazed
windows.
6.1.3
Forced convection
By forced convection we mean situations where air flow along the surface is
generated by an external force independent of the actual temperature
difference between the surface and the medium. Examples are wind on an
exterior surface, air gaps ventilated due to wind effects and air flow in a fan
coil unit.
6.1.4
Dimensionless numbers
The properties of the flowing medium and the mode of flow is characterized
by dimensionless numbers (ref. course in hydraulics).
48

2005-10-24
-------------------------------------------------------------------------------------------------------------------
hc l Nusselt : Actual heat transfer coefficient hc in relation to that of

still air over a given characteristic length l. When the air is
standing still the Nusselt number will be equal to 1. With
increasing air movements either due to temperature differences or to
external forces the Nusselt number will increase.
Nu =
Gr =
g l 3 T Grashof: Criterium for fluid movements due to natural

2
convection. As gravity g, volume expansion coefficient and
the kinematic viscosity for air are fairly constant in the

normal temperature range the Grashof number will depend strongly on the
geometry represented by the characteristic length l and the temperature
difference T.
u l Reynolds: Properties of the flow at forced convection. For a given

fluid and geometry Reynolds number will increase linearly with
the velocity of the fluid. With increasing velocity, when the
Reynolds number reaches a certain limit, we will have transition from
laminar to turbulent flow.
Re =
Pr =
c Prandtl: Properties of the flowing medium. For applications to

building air flow the Prandtl number can normally be set equal
to 0.7.
g = acceleration of gravity 9.81 m/s2
= volume expansion coefficient K-1

l = characteristic length, m, that can be the hydraulic diameter in forced
duct flow or the length of the surface in the direction of the flow for air flow
along exposed surfaces.
u = air velocity, m/s
= kinematic viscosity, m2/s

hc = the coefficient of surface heat transfer due to conduction and
convection, W/m2K.
When the Nusselt number has been calculated for a given characteristic
length l the convective surface heat transfer coefficient can be calculated as
hc = Nu./l
(6.5)
49

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Table 6.1 Properties of dry air at 1.013 bar (atmospheric pressure)
-20
+20
+40
1.395
1.293
1.205
1.127
kg/m3
0.00367
0.00343
0.00320
K-1
1006
1006
1007
1008
0.0243
0.0257
0.0271
J/K.kg
W/mK
13.30.10-6
16.11.10-6
16.97.10-6
m2/s
0.715
0.713
0.711
Pr
6.2
6.2.1
Expressions for different flow types
Forced convection
For forced convection we usually have a system with a given air flow or
given air velocity and the Nusselt number is a function of the Reynolds
number and the Prandtl number.
Nu = f(Re,Pr)
6.2.2
(6.6)
Natural Convection
For natural convection it is the temperature difference that generates the

air movements and thereby the conditions for surface heat transfer and the
Nusselt number will be a function of the Grashof number.
Nu = f(Pr,Gr)
(6.7)
Below are given useful expressions for different types of air flow. The
natural classification of the flow type is, if it is natural or forced flow, if it is
laminar or turbulent and the geometry of the boundary surfaces for the
flow.
6.3
Heat balance in a duct with air flow
In many applications air is heated or cooled on its way through a duct with
surfaces with temperature different from the air.
50

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Ts
u
T0
TL
Ts
x=0
x+dx
Assume that the duct has cross section area A, m2 and perimeter of the
surrounding surface P, m. For the volume element in the duct limited by x
and x+dx we have incoming volume flow u.A with a temperature T(x) and
the same outgoing volume flow at x+dx with the temperature T(x+dx). We
also have a heat flow from the surface to the flowing air depending on the
temperature difference, the convective surface heat transfer coefficient and
the surface area P.dx. Heat balance for the element will yield
u A c(T(x) T(x + dx)) + P dx hc (Ts T(x)) = 0
(6.8)
Using Taylor expansion for the temperature in x+dx
T (x + dx) = T( x) + dx
T
x x
(6.9)
we get the differential equation
T
P hc
P hc
+
T=
T
x u A c
u A c s
(6.10)
With the boundary condition T(0)=T0 we get the solution
T(x) = Ts + (T0 Ts )e
Phc
x
uA c
(6.11)
In many cases the surface temperature is not known a priori but rather
some ambient temperature on the
outside of the duct wall. The same
U
T(x)
Ta
equation then applies replacing hc with
the U-value of the duct wall and Ts with
the ambient temperature.
51

2005-10-24
------------------------------------------------------------------------------------------------------------------
T(x) = Ta + (T0 Ta )e
6.4
6.4.1
PU
x
uAc
(6.12)
Expressions for surface heat transfer coefficients
Forced laminar duct flow
The flow is generated by external forces and the criteria for laminar flow is
Pr > 0.6
Re < 2300
(6.13)
The characteristic length for calculation of the Reynolds number is the

hydraulic diameter of the duct which for non-circular geometries can be
calculated as four times the section area A divided by the length of the
perimeter P of the interior duct section.
dh =
4 A
P
(6.14)
It follows that the hydraulic diameter for a rectangular duct with sides a
and b will be
dh =
4 a b
2(a + b)
(6.15)
and if a>>b the hydraulic diameter will become 2 b. This condition is

typical for ventilated air gaps in the exterior part of insulated constructions.
The following expression gives the average Nusselt number along the
surface.
0.067(Re. Pr. d h / L)
Nu = 3.65 +
(6.16)
(1 + 0.045(Re. Pr. d h / L )2 / 3 )
L is the length of the duct in the direction of the flow, m.
52

2005-10-24
-------------------------------------------------------------------------------------------------------------------
6.4.2
Forced turbulent flow of air in a duct
d
l
0.7 < Pr < 100

l/d > 60
Re > 10000
The following expression is valid for a duct with smooth surfaces with
constant surface temperature. The characteristic length for calculation of
the Reynolds number is the hydraulic diameter of the duct which for noncircular geometries can be calculated as four times the section area divided
by the length of the perimeter of the interior duct section.
Nu = 0.023 . Re0.8 Prn = hc.dh/
(6.17)
n=0.4 if the surface is warmer than the air

n=0.3 if the surface is colder than the air
The expression can for air also be used approximately in the interval
2300 < Re < 10000 if n is set equal to 0.4.
6.4.3
Forced flow along flat surfaces

u
l
The convective surface heat transfer at the exterior surfaces of outer walls
and the roof is usually governed by the wind generated air flow along the
surfaces. The characteristic length here is the length of the surface in the
direction of the air flow at the surface i.e. the length or width of a roof or a
wall. A complication here is that the air velocity at different locations
around the house will differ substantially from the free wind speed observed
at a distance from the building. Calculations of this types are however
seldom carried out to get exact values for the heat transfer coefficient at the
53

2005-10-24
-------------------------------------------------------------------------------------------------------------------
exterior surfaces of a building but rather to study the influence of different

parameters.
6.4.4
Forced laminar flow along a flat surface
The conditions for laminar flow are as follows

0.6 < Pr < 2000
Re < 105
And the expression for the Nusselt number becomes:
Nu = 0.664 Re1/2Pr1/3=( hc.l)/
(6.18)
l, which is the length of the surface in the direction of the flow, m, is also
the characteristic length to be used in the calculation of the Reynolds
number.
6.4.5
Forced convection with turbulent flow along a flat surface
The criteria for turbulent flow along a flat surface are as follows:
0.6 < Pr
6.105 < Re < 107
Nu =
0.037 Re0.8 Pr
h c .l
=
.
0.1
2 /3
1+ 2.443 Re (Pr 1)
(6.19)
Expression (6.13) is mostly used to estimate the dependency of the

convective surface heat transfer coefficient on the convective surface heat
transfer coefficients on the exterior surfaces of buildings. Using those
equations one has to bear in mind that the air velocity around buildings
usually is different from the meteorological wind and there are many
elements around the building that can disturb the air flow.
If a building is 12 m wide Reynolds number will exceed 6.105 at 0.5 m/s and
107 at 10 m/s. The average wind velocity in Stockholm is about 3 m/s.
6.4.6
Natural convection on room surfaces
54

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The following equations apply for the

natural convection surface heat transfer
coefficients on room surfaces i.e. floors,
T
T+
walls and ceilings. The characteristic
length used in the calculation of the
Grashof and Nusselt numbers is usually
the height of the wall or the ratio
Area/Perimeter of the floor/ceiling,
which is being considered. All cases can be expressed with the same basic
equation (6.20) where A and B have different values for different situations.
Nu = A (Gr Pr)B =
hc l
(6.20)
For laminar flow B=1/4 and for turbulent flow B=1/3.
6.4.6.1 Vertical walls

For vertical walls we can use the same expression independent on whether
the wall is colder or warmer than the room air. The mode of flow is
determined from the Grashof number and the characteristic length to be
used in the calculation of the Grashof number is the wall height.
Laminar flow
Turbulent flow
(Gr.Pr<109)
(Gr.Pr>109)
A=0.59
A=0.13
B=1/4
B=1/3
(6.21)
(6.22)
6.4.6.2 Horizontal surfaces

For horizontal surfaces the heat transfer will depend not only on the
temperature difference but also on the thermal stability at the surface. For
a ceiling colder than the room air the density of the air at the surface will be
higher than below which will generate turbulent air movements at
relatively low Grashof numbers. Similar instability will appear at warm
floors where the density of the air in the vicinity of the surface is lower than
for the room air. The characteristic length will be the floor or ceiling area
divided by the perimeter of the floor or the ceiling. As an example consider a
room 3x6x2.4 m. The floor area is 18 m2 and the perimeter is 18 m which
will give the characteristic length 1 m.
For a relatively warm floor or a cold ceiling:
Laminar flow
Turbulent flow
(Gr.Pr< 2.107) A=0.54 B=1/4

(Gr.Pr>2.107) A=0.15 B=1/3
(6.23)
(6.24)
For a relatively cold floor or a warm ceiling we expect conditions to be stable

up to high Grashof numbers.
55

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Laminar flow
6.4.7
(Gr.Pr< 3.1010)
A=0.27 B=0.25
(6.25)
Natural air convection within an enclosure
It is not always easy to make a distinction between a room and an

enclosure. By an enclosure we mean a space where the distance between the
surfaces is so small that the convection flow generated at one surface affects
the other. The most common examples are thin non-ventilated air layers in
building constructions and the air gaps between the panes in a multi-glazed
window.
6.4.7.1 Horisontal gap with upwards heat flow

The thickness of the gap d, m, is the characteristic length to be used in the
calculation of the Grashof number
Td
T+
The following expressions are valid for air
Nu = 0.195 Gr 1 / 4
Nu = 0.068 Gr 1 / 3
3.7.104<Gr<3.7.105
3.7.105<Gr<3.7.107
(6.26)
(6.27)
And a more general expression for fluids is
Nu = 0.069 (Gr Pr )
1/ 3
Pr 0.074 1.6.105<Gr.Pr<109
(6.28)
The heat transfer coefficient hc is in this case given from surface to surface.
When Nu = 1 the heat transfer coefficient is given as
hc = /d
(6.29)
which means that the air in the gap is standing still and the heat transfer
in the air is taking place by conduction only.
56

2005-10-24
-------------------------------------------------------------------------------------------------------------------
6.4.7.2 Vertical gap with limited height H and horizontal heat flow
The thickness of the gap, d, m , is the characteristic length
to be used in the calculation of the Grashof number
T-
2.104 < Gr < 2.105 H>3d

Nu = 0.16.Gr1/4(H/d)-1/9
(6.30)
2.105 < Gr < 2.107 H>3d

Nu = 0.066.Gr1/3(H/d)-1/9
(6.31)
Expressions (6.25) and (6.26) are used for instance for the
calculation of the convective heat transfer between the
panes of a multi-glazed window.
The heat transfer coefficient is given as
Nu
(6.32)
d
and the average density of heat flow rate between the surfaces can be
calculated as
hc =
q = hc (T+ T )
(6.33)
At high Grashof numbers the air in the cavity will start rotating due to the
density differences and we can assume that we have downward flow of air
on the cold side and upward flow of air on the warm side. This means that
there will be a temperature and heat flow gradient along the surfaces.
6.5
Air flow through building components
The pressure difference as calculated above from various processes is the

governing potential which generates air flow through leakage paths in the
construction. The leakage paths are seldom desired or planned, which
implies a high degree of uncertainty concerning the geometry and other
properties of the paths. Uncontrolled air leakage through building
constructions is usually linked to unwanted consequences such as moisture
problems, distribution of odors, energy loss and draught. Important
exceptions are fresh air inlets and air gaps for ventilation of the exterior
parts of constructions. It is however of great value to provide means to
study the air leakage processes in order to quantify crucial construction and
material properties to ensure the required air tightness.
57

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The air leakage path through a construction is

usually of a complex nature. As an example we have
air entering from the inside through a crack in the
drywall, through an overlap in the vapor barrier,
through a porous insulation material to a crack in the
wind barrier into the external ventilation gap and out
through the ventilation openings in the exterior brick
cladding. The air tightness however is often provided
for in one layer in the construction such as the vapor
barrier, a concrete layer, a sheet metal deck or a wall
layer of aerated concrete. The rest of the construction
can then often be considered as non airtight and the
openings in the air tight layer can be studied as
connecting directly the inside and the outside.
6.5.1
Air flow through an opening
The equations used for the calculation of air flow through leakage paths are
the same equations as used for circular pipe and duct flow. The shape of a
crack in a building layer is of course rather far away from the cylindrical
form of a pipe. By using the hydraulic diameter for the actual configuration
we however hope to get useful results and the research conducted hitherto
in this field supports that.
Assume that we have a general

opening consisting of an inlet, a
narrow duct between the surfaces
and an outlet as shown in the
figure. The loss in static pressure
through the opening can be
considered in four steps as shown
in principle
1. As the air enters the inlet the

velocity is increased from 0 to a
mean velocity um given by the
flow and the dimensions of the
p2
opening. In the very opening of
1
the inlet a phenomenon occurs
3
4
called the vena contracta with
2
stagnation zones at the edges and
thereby a still higher velocity um+
before the flow develops. The pressure drop is given from Bernoullis
equation as
p1
58

2005-10-24
-------------------------------------------------------------------------------------------------------------------
dp1 > -(um+)2/2
(6.34)
2. When the flow develops a part of this is regained

dp2 < (um+)2/2-(um)2/2
(6.35)
3. Through the gap itself the friction between the wall and the air generates
a linear pressure drop. The friction loss is given by the equation
dp3 = (f.d/Dh) .(um)2/2
(6.36)
f is the friction coefficient which is a parameter depending on the Reynolds

number of the flow in the gap and the surface roughness.
Dh is the hydraulic diameter, m.
4. In the outlet of the gap the velocity goes down from to 0 and a part of the
pressure loss due to the increase of the velocity at the opening is regained.
dp4 < (um)2/2
(6.37)
The friction loss according to the processes 1, 2 and 4 are related and in
practical calculations given as
dp1+dp2+dp4 = .(um)2/2
(6.38)
For laminar flow in the gap i.e. when Reynolds number is below 2300, can
be set equal to 1.5 and the friction coefficient can be calculated as
f. = 64/Re for a circular geometry

f. = 96/Re for a rectangular air gap of infinite length
(6.39)
In the turbulent zone i.e. if the Reynolds number is above 4000, can be set
equal to 1.8 and the Colebrook-White formula is used for f
2, 51
k / b
= 2 log
+
f
Re f 3, 71
(6.40)
This formula does not have an explicit solution but has to be solved by
iteration.
k is the surface roughness of the surfaces along the air gap, m. For building
materials k can be of the order of magnitude 0.00001 m for smooth plastic
surfaces, 0.0001 m for smooth steel surfaces and 0.001 m for concrete.
59

2005-10-24
-------------------------------------------------------------------------------------------------------------------
In the range 2300< Re < 4000 or the so called transition zone, the type of
flow can vary. A practical approach is to use equation (6.17) or (6.18) for
calculation of f. depending on which choice gives results on the safe side for
the actual purpose of the calculation.
The total pressure loss over the gap can now be expressed as
p1-p2 = ( +f.d/Dh) .(um)2/2
6.5.2
(6.41)
Air flow through an orifice plate
In building applications air flow and pressure drops over holes or slots in
thin plastic films or sheet metal plates often has to be estimated. This can
be approximately treated with Dicks formula relating the air flow L through
the hole with area A to the pressure drop over the hole.
L = 0.65 A
6.5.3
2 p
m3/s
(6.42)
Air flow through porous materials
Since air flow through porous material is more often laminar, a linear
relation between the velocity and a pressure gradient can be established in
the material and for the flow of an incompressible fluid, Darcys law is valid.
u =
B0 p
(6.43)
= The specific permeability of the material, m

B0
= The dynamic viscosity of the fluid, Ns/m2

p/x = The pressure gradient, (m/s)
60

2005-10-24
-------------------------------------------------------------------------------------------------------------------
7 Radiative Heat Transfer

In chapter 3 we have stated that the surface heat transfer normally is a
combination of the two processes convection and radiation. In chapter 5 we
established formulae to calculate the convective heat transfer coefficient at
free surfaces, within enclosures and in duct flow. In this lecture we intend
to study the exchange of long wave radiation between surfaces and the
impact of absorption of short wave radiation on surfaces.
7.1
Exemplification of applications
Radiation exchange is important for several processes in our environment.

On cold nights the sky temperature can be at a temperature 10 to 20 K
below the air temperature. Horizontal surfaces are cooled by radiation
exchange towards the sky and reach a temperature below the dew point of
the air. Condensation of water vapor from the air will occur at the surface
and at temperatures below zero this can lead to frosting on car windows and
icy, slippery roads.
The temperature we sense on our skin is partly the air temperature and
partly the temperature of the surrounding surfaces due to the radiation
exchange. Therefore human comfort is depending not only on air
temperatures and velocities but also on the temperatures of the surrounding
surfaces.
In simplified building heat loss calculations we assume that heat transfer
takes place from indoor air to outdoor air through a series of resistive
layers. In more detailed studies of indoor heat exchange and comfort we
have to consider the heat transfer in a room as a complex system with
radiation exchange between the surfaces parallel to the convective heat
transfer between the surfaces and the air.
In porous insulation materials and in construction with air cavities and air
gaps, heat transfer is partly taking place by radiation between fibers and
cell walls.
Material surfaces emit long wave radiation that is a function of the surface
temperature. By measuring the infrared radiation from a surface,
information on the distribution of the surface temperature can be
monitored. Cameras for infrared thermography are widely used in industry,
in medical applications and of course in buildings where cold spots detected
on insulated surfaces indicate bad insulation, air leakage, moisture etc.
61

2005-10-24
-------------------------------------------------------------------------------------------------------------------
7.2
7.2.1
Basic theory
Thermal radiation
All bodies emit energy by

electromagnetic waves
Infrared
Ultraviolet
which we call thermal
radiation
radiation
radiation. The
characteristics of the
Thermal
radiation are depending
radiation
on the properties of the
Wavelength, m
surface material and on
the surface temperature.
Thermal radiation is
10-3 10-4 10-5 10-6 10-7 10-8 10-9 10-10
defined as radiation with
wavelength between 10-5
1
m
1 mm
1
and 10-8 m. This includes
visible light as well as a
part of the infrared and of the ultraviolet spectrum. At normal room
temperature, surfaces emit radiation far into the infrared spectrum which
we are not able to see. The visible light that we detect with our eyes, as
coming from a surface of a body, is not emitted radiation but reflected
radiation that originates from the sun or some artificial light source.
Visible
light
7.2.2
Black body radiation
When treating radiation from a real body surface we first find the radiation
from a black body at the same temperature. Black body is defined as a body
with a surface that absorbs all incident radiation for all wavelengths,
directions and polarizations. The radiation from the real body can then be
expressed as the black body radiation multiplied by the emissivity of the
real surface.
Black body total excitance is expressed by the Stefan Boltzman law
M= T4
(ISO 9288)
W/m2
(7.1)
Black body spectral excitance giving the distribution of the radiation over
the spectrum can be given as a function of the temperature, K.
o
M =
C 1 5
C2
W/m3
62
(7.2)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
is the wavelength in m.
C1 = 3.741.10-16 W.m2
C2 = 0.014388 m.K
7.2.3
Gray body properties
7.2.3.1 Emissivity
The emissivity of a surface is the ratio between the emitted radiation and
the radiation of a black body at the same temperature. The emissivity of
surfaces can vary with the wavelength. This is for instance utilized in
window glazing with so called LE or low emittance coating where the
surface is treated to have low emissivity for the infrared spectrum while
visible light is less affected.
Total hemispheral emissivity is the total excitance of the considered surface
M divided by the total hemispheral excitance of a black body M at the same
temperature. Since the spectral distribution of the black body excitance
varies with temperature the emissivity will also vary with temperature
=M/M
(7.3)
7.2.3.2 Total absorptance

Assume that we have an incident radiation with heat flow rate i towards
a gray surface. A part of this incoming radiation, a, is absorbed at the
surface. The absorptance of the surface, , is defined as
=a/i
(7.4)
7.2.3.3 Total reflectance

A part of the incident radiation, r, is reflected by the surface and the
reflectance, , of the surface is defined as
=r/i
(7.5)
63

2005-10-24
-------------------------------------------------------------------------------------------------------------------
7.2.3.4 Total transmittance

With transmitted radiation is meant incident radiation that is neither
reflected nor absorbed in the surface but passes further into or through the
material layer. Even for so called transparent materials the transmittance
will vary dramatically with the wavelength. For most opaque building
materials the transmittance is zero for the whole spectrum. For normal
window glass the transmittance for visible light will be about 90 % but for
long wave radiation as generated at normal temperatures the transmittance
is practically zero. The transmittance is defined as the radiant heat flow
rate transmitted through a surface, t, divided by the incident heat flow
rate, i
=t/i
(7.6)
From the above definitions it is clear that the incident radiation is either
reflected, absorbed or transmitted and it follows that
+ + = t/i +r/i +a/i = 1
(7.7)
For opaque bodies where is equal to zero it follows that
+ = 1
(7.8)
It can also be shown by regarding two identical surfaces facing each other
and at the same temperature that, since the net heat flow is zero, the
absorptance must be equal to the emittance. For longwave radiation
exchange this can be assumed to be valid with good accuracy.
7.3
7.3.1
(7.9)

Total radiosity
The total radiosity from an opaque surface is the heat flow rate for emitted
and reflected radiation from the surface, divided by the area of the surface.
J=/A
(7.10)
64

2005-10-24
-------------------------------------------------------------------------------------------------------------------
This is a useful quantity since it must be equal to the density of heat flow
rate supplied to the surface by other processes like absorbed radiation,
conduction and convection.
7.3.2
AjJj
AkJkF kj
j
The heat induced by long wave

radiation at a non transparent surface j
in an enclosure consisting of n diffuse
surfaces, each characterized by a
uniform temperature and emittance,
can be expressed as
n
j = A k J k Fkj A j J j (7.11)
k =1
By a diffuse surface is meant a surface that reflects or emits radiation

equally in all directions. An ordinary bathroom mirror is not diffuse since
the image of the incoming light is to a large part preserved in the reflected
radiation. After a hot shower however the moisture that condenses on the
surface makes it diffuse and we no longer see our image.
The density of heat flow rate leaving a diffuse surface as long wave
radiation is called the radiosity of the surface and is given the symbol J.
Fkj is the configuration factor or the view factor and may be defined as
the fraction of diffuse energy leaving area Ak that is directly incident on
area Aj.
The radiosity can be expressed as the sum of the emitted and reflected
radiation per area at the surface.
Jj = j Tj + (j/Aj)
4
k =1
JkFkj Ak
(7.12)
or since we further assume that the reciprocity theorem is valid

Fkj Ak = Fjk Aj
Jj = j Tj4 + j
(7.13)
n
k =1
JkFjk
(7.14)
is the reflectance and is the emittance of the surface. is the Stefan

Boltzmann constant 5.67 10-8 W/m2K4.
65

2005-10-24
-------------------------------------------------------------------------------------------------------------------
7.4
Calculation of view factors

j
The view factor between two diffuse

surfaces is a purely geometrical quantity.
From a plane surface k the radiosity is
equally distributed in all directions from
the surface. The amount of radiation that
reaches a remote surface j will be
determined by the proportion of the
hemisphere that is covered by the
k
projection of this very surface.
Mathematically this is somewhat complicated. In general the view factor
from surface j to surface k can be expressed in the form of a double integral
Fjk =
Aj
1
Aj
cos j cos k
dA jdA k
d2
A jA k
(7.15)
dAj
Normal
This general expression is in

practice difficult to calculate so
Ak
Normal
dAk
that formulae and diagrams are
given for special cases to cover
most applications that can arise in buildings. Starting with the simplest
case it can be stated that for a flat or convex surface no part of the radiation
leaving the surface will hit the surface itself i.e.
Fkk=0
(7.16)
For two infinite parallel planes all radiation leaving plane 1 will hit plane 2
and vice versa. This is the case when studying the radiation exchange
between window panes in a multi-glazed window.
F12= F21= 1
(7.17)
For a plane surface 1 surrounded by

a concave surface 2 it is evident
that all radiation leaving surface 1
will hit surface 2. However some of
the radiation leaving surface 2 will
hit itself. Therefore the view factor
F12 will be equal to unity
F12 = 1
66
(7.18)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
and using the reciprocity theorem gives the relation

F12 A1 = F21 A2
(7.19)
Which gives the view factor F21 as

F21 = F12 A1/ A2
(7.20)
This case is useful for instance when calculating the radiation exchange for
a radiator in a room where all the other surfaces can be assumed to have
the same temperature.
If the total surface of a volume consists of n different surfaces all radiation
leaving the surface has to hit one of the surfaces including the surface itself
if it is convex. The sum of all view factors from a surface therefore has to
equal unity
n
F
j =1
kj
=1
(7.21)
The practical use of this relation is that if the view factors to all surfaces
but one are known the last view factor can be found by simply subtracting
the known view factors from unity.
7.4.1
View factors for basic configurations
Especially when studying thermal comfort

it is of interest to study the radiation
exchange between a small surface and the
H
surrounding surfaces. For infinitesimal
2
surfaces there are explicit formulae for the
D
view factors to rectangular perpendicular
1
and parallel surfaces as given in the
L
following cases. For the radiation exchange
between an infinitesimal surface 1 and a
perpendicular surface 2 as shown in the figure the following equation is
valid.
1
1
L
D
+
F12 =
arctan
arctan
2
2
D
2
H
H
1
+
1+
D
D
67
(7.22)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
For the radiation exchange between an

infinitesimal surface 1 and a parallel surface 2
H
as shown in the figure the following equation is
valid. Observe that in both cases the location of
2
the infinitesimal element is in the normal
D
direction from the corner of the larger element.
1
If the infinitesimal element is not in the
L
normal direction of the element this is simply
achieved by enlarging the element and then
subtracting the view factor to the enlargement.
H
L
L
H
1
D
D
D
D
+
F12 =
arctan
arctan
2
2
2
2
2
H
H
L
L
1+
1+
1+
1+
D
D
D
D
(7.23)
The view factors between two arbitrary surfaces in a rectangular room can
be found by integration of the above formulae. This is mostly used in
computer codes that simulate the energy exchange within a room or a whole
building.
Solutions for the view factors for different configurations are given in forms
of tables or diagrams in various publications. The diagrams below give the
view factors for parallel and perpendicular rectangular surfaces.
68

2005-10-24
-------------------------------------------------------------------------------------------------------------------
69

2005-10-24
-------------------------------------------------------------------------------------------------------------------
7.4.2
Use of the addition rule
There are certain relations between the configuration factors which can be
utilized to reduce the calculation work needed for a given geometry. The
relation stated in equation (7.13), means that between two surfaces, the
configuration factor only has to be calculated in one direction.
From the definition of the configuration factor it is evident that the sum of
all fractions is equal to unity so that for a closed space with surfaces
numbered from 1 to n
n
kj
=1
(7.24)
j= 1
Furthermore if the surface j is divided in two partial surfaces j' and j" , the
configuration factor from the surface k to the surface j is given by
Fkj = Fkj' + Fkj"
(7.25)
j
j'
j''
And in a similar way

Fkj' = Fkj - Fkj"
7.5
(7.26)
The Radiosity matrix
Once the configuration factors are known the radiosities at different

surfaces can be calculated from equation (7.14) which can be put into matrix
form as
Jj - j
k =1
JkFkj = j Tj4
(7.27)
70

2005-10-24
-------------------------------------------------------------------------------------------------------------------
1 1 F11 1 F12
2 F21 1 2 F22
F
n
n1
1F1n J1
1T1 4
2 T2 4
J2
=
4
T
1 nFnn Jn
n n
J1
1 1F11 1 F12
J2
2 F21 1 2 F22
=

J
F
n
n n1
(7.28)
1 F1n 1 1 T1 4
2 T2 4

1 n Fnn nTn4
(7.29)
Where three or more surfaces are involved the inversion of the matrix
should be carried out numerically. The explicit solution for two surfaces
reads as follows.
[(1 2 F22 ) 1 T14 + 1 F12 2 T24 ]

J1 =
(1 1 F11 )(1 2 F22 ) 12 F12 F21
J2 =
(7.30)
[(1 1 F11 ) 2 T24 + 2 F21 1T14 ]
(7.31)
(1 1F11 )(1 2F22 ) 12 F12 F21
For long wave radiation the emittance and reflectance of a surface are
with good approximation related as
=1-
(7.32)
From the equations (7.11) (7.31) and (7.32) it can be deduced that the heat
generated by long wave radiation at the surface 1 for the simple
configurations with only two surfaces is given by the formula
1 = A 2 J2 F12 A 1 J1
A 2 [2 T24 + 2 F21 1 T14 ] A1 [(1 2 F22 ) 1 T14 + 1 2 T24 ]

=
(1 2 F22 ) 12F21
(7.33)
After some mathematical exercise this can be formulated as
1 = A1 12 (T24 - T14 )
(7.34)
12 is the so called resulting emittance between both surfaces, which is,

besides the surface emittances, depending on the geometry.
For the case with two infinite parallel surfaces the resulting emittance can
be expressed as
71

2005-10-24
-------------------------------------------------------------------------------------------------------------------
1/12 =1/1 +1/2 - 1
(7.35)
And if surface 1 is completely surrounded by surface 2

1/12 =1/1 +(A1/A2)(1/2 - 1)
(7.36)
If A2 >> A1
12 = 1
(7.37)
It is useful to define a surface heat transfer coefficient due to long wave

radiation at a surface in analogy with the convective heat transfer
coefficient. From the case with two different isothermal surfaces, the heat
transfer coefficient due to long wave radiation at the surface 1 hr can be
defined in such a way that
hr = 21/A1(T2-T1)
(7.38)
From equation (7.34)

hr = 12 (T24-T14)/(T2-T1)
(7.39)
Given that
(T24-T14) = (T22-T12)(T22+T12) = (T2-T1)(T2+T1)(T22+T12)
= (T2-T1)(2Tm)((T2+T1) 2-2 T2T1)
= (T2-T1)(2Tm)( 4Tm 2-2Tm 2+ (T2-T1) 2/2)
(7.40)
and as the temperature difference normally is small compared with the

total value, hr can be expressed as
hr = 4 12 Tm3
(7.41)
Tm = (T1+T2)/2
(7.42)
72

2005-10-24
-------------------------------------------------------------------------------------------------------------------
7.6
Combined radiation and convective heat transfer
Normally we assume that the air does not absorb radiation. This means
that radiation and convective heat transfer can be regarded as two separate
processes coupled through the surface temperature. For a thin air gap the
heat transfer coefficients can be calculated separately and the overall heat
transfer coefficient between the two surfaces, hcr, simply calculated as the
sum of these two.
q12 = (hc + hr ) ( T1 T2 ) = hcr (T1 T2 )
(7.43)
For the heat transfer at the surface in a room the situation is more
complicated. The convective heat transfer is taking place between the
surface and the room air while the radiative heat transfer is taking place
between the surface and other surfaces of the room. A common
simplification when calculating the surface heat transfer for insulated
building constructions is to assume that on the average the ambient air
temperatures and the surrounding radiating temperatures are the same.
Then a common surface heat transfer coefficient can be defined as
hs = ( hc + h r )
(7.44)
and the surface resistances as
R si =
1
hsi
(7.45)
R se =
1
hse
(7.46)
See chapter 3.
7.7
Solar radiation in building design
Solar radiation, and its impact on buildings and building surfaces, is of a

great concern in building design. Many examples of a bad design have their
origin in lack of knowledge and understanding of the process of solar
radiation.
With increased insulation and air tightness of buildings the solar radiation
through windows will play a relatively larger role in the power- and energy
balance for the building. The orientation and design of the windows will
73

2005-10-24
-------------------------------------------------------------------------------------------------------------------
influence the energy consumption for heating as well as the level of comfort
and the need for cooling in a building.
The performance of external solar collectors or photovoltaic solar cells is a
function of their orientation toward the sky. For real buildings it is often a
question of making the best use of available surfaces.
Solar radiation can heat up external exposed surfaces to temperatures
around 80 C and surfaces with partly transparent coating to more than
that. The temperature differences gives rise to mechanical stress,
dimensional changes and, together with moisture and chemical reactions,
cause a rapid decay of surface finishes. Solar radiation is therefore an
important factor in design for high durability, both in building constructions
and for infrastructural constructions like bridges and roads.
7.8
The elements of solar radiation
Solar radiation is available on meteorological weather files, as tabulated

values and from analytical and empirical formulae as given below. The solar
radiation on a weather file is usually given as the incident radiation on a
horizontal surface or as the cloudiness. The calculation procedure given
below can be used directly to calculate the incident solar radiation on a clear
day, but also as a tool to establish a relation between meteorological
weather data and the incident solar radiation on different surfaces.
74

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The total short wave radiation from the sun reaching an earthly surface, I
W/m2 can be split into three components
ID = direct solar radiation, W/m2
Rays directly from the sun to the surface
Id = diffuse solar radiation, W/m2
Secondary rays reradiated from the atmosphere
2
IR = reflected solar radiation, W/m
Secondary rays reflected from surfaces on the ground
7.9
Heat balance on exterior surfaces

The components of heat
transfer to be regarded at an
exterior surface are
Tsky
Te
Convective heat exchange

between the surface and
outdoor air
Solar radiation
Long wave radiation
exchange with the sky
Heat exchange between the
surface and the interior
climate
aI
T es
Ti
Other important components of the surface heat balance that are not
treated here are water transfer and phase change such as evaporation of
rain water and melting of snow.
A heat balance for the surface gives
aI+hc(Te-Tes)+hr(Tsky-Tes)+U(Ti - Tes) = 0
75
(7.47)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Absorptance for typical

building surfaces
Asphalt
Brick, red
Brick, yellow
White paint
0.93
0.75
0.55
0.30
a is the absorbtance of the surface for

shortwave solar radiation. U is the modified
U-value, i.e the heat transfer coefficient
between the outer surface and the ambient
indoor air.
1
(7.48)
U =
1
Rse
U
Rearranging the terms gives the equation for the temperature of the
external surface
Tes =
a I + hc Te + hr Tsky + U Ti
hc + hr + U
(7.49)
7.9.1 The equivalent outdoor temperature

Assume the following
(7.50)
Te = Tsky
Rse =
1
hr + hc
(7.51)
and that there is such a temperature Tequ that

(Tes-Tequ)/Rse = (Tes-Te)/Rse - aI
(7.52)
It follows that
Tes-Tequ = Tes-Te - Rse.aI
(7.53)
and explicitly for the equivalent temperature

Tequ = Te + Rse.aI
(7.54)
76

2005-10-24
-------------------------------------------------------------------------------------------------------------------
8 Heat Balance Problems

8.1
General heat balance problems
In chapters 3 to 9 we have come across models of various types to calculate

the separate heat transfer processes in building heat transfer. In reality,
most situations that we want to analyze, involve a combination of different
processes. We want to calculate the net heat flow to a surface or volume
under given conditions or to calculate the resulting temperature at a given
location and these quantities are influenced by several processes interacting
with each other.
To establish a solution for the heat balance for a given configuration we can
proceed in four steps:
1. Identification of the significant processes involved and choosing the
suitable algorithm to describe these. Heat flow through an insulated wall,
absorbed solar radiation and convective heat transfer are examples of
such processes.
2. Expressing the heat flows for the actual processes in terms of known
boundary values such as temperatures, heat flows, incident solar
radiation etc.
3. Making use of the fact that the sum of the heat flow to the surface or the
volume and the heat generated must be equal to zero.
4. Solving out the resulting temperatures directly or by iteration whereas
the heat transfer parameters may be temperature dependent.
8.2
Heat balance for a building
A simple heat balance model for a building volume presumes a constant

temperature in the whole building and that all energy is transferred to the
room air directly and that the indoor temperature is equivalent to the room
air temperature.
T+TB+S+V+H+P+C = 0
77
(8.1)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Transmission heat flow through homogeneous surfaces
T = (Te-Ti)A.U
(8.2)
2
A = area m
U = thermal transmission coefficient, W/m2.
Transmission heat flow through thermal bridges
TB=(Te-Ti)(.l +)
(8.3)
= Linear thermal transmittance

l
of a thermal bridge, W/mK

= length of a linear thermal bridge, m
= Point thermal transmittance of a thermal bridge, W/K

Heat generated from solar radiation through windows
S=A.F.I
(8.4)
2
A = window area, m
F = Total window transmittance
I = Total solar radiation at the exterior window surface, W/m2
Heat flow through ventilation
V=(Te-Ti).L..c
(8.5)
L = Flow of indoor air to the outside, m3/s

= density of indoor air, kg/m3
c = thermal capacitivity of indoor air, J/kgK
Indoor heat sources
H = Heat flow generated from the heating system, W

P= Heat flow generated from household appliances, persons etc, W
Thermal inertia
78

2005-10-24
-------------------------------------------------------------------------------------------------------------------
C = Heat flow to or from a capacitive heat storage, W.

For steady state calculations as for instance calculating the energy balance
for a building with seasonal or monthly averages this term can be set equal
to zero. For more exact non-steady calculations as for instance simulating
the temperature variations in a room on a sunny day a more complex
representation of the room heat balance is needed and such calculations are
usually carried out with the computer programs.
8.3
Window heat balance
The heat transmission through windows is complicated by the fact that the
transmission of solar radiation and the potential heat flow through the
window cannot be regarded as parallel, independent processes. The solar
radiation absorbed in the different panes causes a rise in temperature and
thereby disturbs the heat flow pattern.
The window can be treated as a series of nodal points representing the
temperatures of the panes Tj and with thermal resistances Rj on the left of
the pane representing long wave radiation and convection heat transfer.
The fraction of the total incident radiation that is absorbed in pane j is
denoted by aj. This is exemplified in the figure for a triple glazed window.
Te
T1
T2
T3
Ti
a1
a2
a3
R1
R2
R3
R4
A heat balance for a pane j then gives

(Tj-1-Tj)/Rj + (Tj+1-Tj)/Rj+1+aj.I = 0
(8.6)
79

2005-10-24
-------------------------------------------------------------------------------------------------------------------
In which aj is the fraction of the incoming solar radiation I W/m2 absorbed

by pane j. This implies that all aj's are dependent.
a1
Single
glazing
Double
glazing
Triple
glazing
Quadruple
glazing
a2
a3
a4
Examples of the relative

absorption of the different
window panes for different
numbers of panes of 4 mm
ordinary glazing are given in
the table.
0.086
0.091
0.073
0.095
0.077
0.061
0.098
0.081
0.066
0.052
Solving Tj
Tj (
T
T
1
1
+
) j1 j +1 = aj. I
R j R j+1
R j R j +1
(8.7)
for pane nr 1 T0 is known equal to Te.
T1(
1
1
T
T
+ ) 2 = a1 I + e
R1 R2
R2
R1
(8.8)
for pane nr n, Tn+1 is known equal to Ti.
1 Tn 1
T
1
+
= a n. I + i

R
R n+1 Rn
R n +1
n
(8.9)
This gives the matrix equation
(8.10)
Tn
1/R1+1/R2 - 1/R2
... 0
T1
.... 0
T2
... 0
T3
- 1/R2
1/R2+1/R3 -1/R3
- 1/R3
...............
....................
...
Tn-1
Tn
.............
...........
..............
- 1/Rn
.............
-1/Rn
1/Rn+1/Rn+1
a1.I + Te/R1
a2.I
a3.I
...............
an-1.I
an.I+ Ti/Rn+1
And the solution for the T vector is reached by inversion of the matrix and
multiplication on both sides
(8.11)
80

2005-10-24
------------------------------------------------------------------------------------------------------------------T1
1/R1+1/R2 - 1/R2
T2
- 1/R2
1/R2+1/R3 -1/R3
T3
- 1/R3
...
Tn-1
.............
Tn
... 0
-1
.... 0
... 0
a1.I + Te/R1
a2.I
a3.I
...............
....................
...............
...........
..............
- 1/Rn
.............
-1/Rn
1/Rn+1/Rn+1
an-1.I
an.I+ Ti/Rn+1
For a window with two panes the equation becomes

T1
T2
-1 a1.I + Te/R1
1/R2+1/R3
a2.I+ Ti/R3
1/R1+1/R2 - 1/R2
= - 1/R2
(8.12)
Inverting the matrix gives

T1
= 1/((1/R2+1/R3)(1/R1+1/R2)-1/R22) 1/R2+1/R3
T2
1/R2
1/R2
1/R1+1/R2
a1.I + Te/R1
a2.I+ Ti/R3
(8.13)
Solution for T1 becomes
T1=
(1/ R2 + 1/ R3 )(a . I + Te / R1 ) + (1/ R 2 )(a . I + Ti / R3 )

1
(1/ R 2 + 1/ R 3 )(1/ R1 + 1/ R2 ) 1/ R2 2
T1= R1R2R3
T2= R1R2R3
(8.14)
(1 / R 2 + 1 / R 3 )(a 1. I + Te / R 1 ) + (1/ R 2 ) (a 2 . I + Ti / R 3 )
(8.15)
R 1 + R2 + R3
(1 / R 1 + 1 / R 2 )(a 2 . I + Ti / R 3 ) + (1 / R 2 )(a 1 . I + Te / R 1 )
R1 + R2 + R3
(8.16)
The density of heat flow rate due to convection and long wave radiation
between the surface and the indoor environment can now be calculated
q=(T2-Ti)/R3
(8.17)
A solution of the system for two panes with area A, m2, gives the heat flow
from the inner surface to the room as
= (A/R3)(T2-Ti) = A[R1(a1I+a2I)+R2a2I + (Te - Ti)]/(R1+R2+R3)

(8.18)
or in more conventional terms
81

2005-10-24
-------------------------------------------------------------------------------------------------------------------
= AIU(Rse(a1+a2)+R2a2)+AU(Te-Ti)
(8.19)
The practical consequence of the above expression is that the fraction of the
incoming radiation that is absorbed in the windowpanes and then
transferred into the room by convection and radiation is with a good
approximation independent of the internal and external temperatures. This
also applies for windows with more than two panes.
The total transmittance F is defined as the fraction of the incident solar
radiation that reaches the interior of the room. The fraction F2 entering the
room by convection and long wave radiation is found from the first term in
equation (8.19)
F2 = U(Rse(a1+a2)+R2a2)
(8.20)
and the short wave radiation part is expressed as

q = F1I
(8.21)
in which F1 is the transmittance of the window for short wave radiation.

The total transmitted solar radiation is therefore given by the formula
q = I{F1+ U[Rse(a1 + a2) + R2 a2}= I(F1 + F2)
(8.22)
A mathematical procedure to calculate the absorptance of each window pane

and the transmittance for a multi-glazed window is given in Isflt (1974).
The transmittance
Transmittance %
for several types of
100
windows with
different kinds of
solar shading are
80
to be found in the
literature, e.g.
60
Double
ASHRAE
glazed
Handbook of
window
Fundamentals.
40
Some examples are
given below.
20
30
60
90
Angle of
incidence i
82
A complication is
that the
transmitted
radiation is
dependent upon

2005-10-24
-------------------------------------------------------------------------------------------------------------------
the angle of incidence of the solar radiation. The figure above shows the
total transmittance through a normal double glazed window as a function of
the angle of incidence. At angles up to 50 the total transmittance is
approximately constant while it decreases almost linearly to zero when the
angle exceeds 50 - 60. Consequently the total transmittance is different for
the three components of the incident radiation i.e. direct, diffuse and ground
reflected radiation.
Given that the curve-forms for different window types are similar, the
transmittances tabulated are given as a percentage of the transmittance
through a normal double-glazed window. Thus, knowing the transmittances
through a double-glazed window, the transmittances for any window given
in the tables are easily calculated. Tables and diagrams for the daily
variations of transmitted radiation through a certain type of window are to
be found in various publications. For Sweden see Brown and Isflt (1969).
The table below shows short-wave solar gain F1 and convective solar gain F2
for various types of glazing and shading. The values are derived from Bring
and Isflt (1979). They are given as a fraction of the total radiation through
a double glazed window and have to be used together with the above
diagram showing the transmittance for a double-glazed window. Observe
that F1 and F2 have different interpretation in the original text.
Glazing
Shading type
F1/Fdoublegl
%
F2/Fdoublegl
%
single glazed
double glazed
ext.pane heat absorbing
ext.pane heat reflecting
ext.venetian blinds
mid.venetian blinds
mid white weave.
mid.grey weave
mid.dark weave
triple glazed
ext. venetian blinds
ext. gap venetian blinds
int. gap venetian blinds
int. venetian blinds
109
93
40
8
11
14
17-54
10-43
4-33
80
6
9
11
13
3
7
20
6
28
51
14-8
30-21
42-30
11
5
21
37
53
83

2005-10-24
-------------------------------------------------------------------------------------------------------------------
8.4
The heat balance for a ventilated air gap
The solution for the temperature of an air stream flowing with constant
velocity along an air gap with isothermal surfaces is given in chapter 5. In
an air gap in an insulated construction with a significant temperature
difference over the air gap the radiation heat exchange between the surfaces
of the air gap is an extra complication and if that is taken into account the
solution will be as following.
R*
2
Ts2
T0
Ts1
equ
Tx
R1*
T
x =0
The heat balance in the air gap contains three processes

1. The convective heat transfer by the air flow along the air gap
2. The convective heat transfer between the surfaces of the air gap and the
passing airflow
3. The heat transfer by radiation between the surfaces
The expression for the variation of the air temperature along the air gap can
be expressed as
Tx = T - (T - To) e-kx
(8.23)
k = 1/(cbuR)
T is the temperature of the air in the gap at an infinitely large distance

from the inlet
T = Ti - (Ti-Tequ).R1/(R1+R2)
(8.24)
84

2005-10-24
-------------------------------------------------------------------------------------------------------------------
R =
r
1
+
2
1
1
h + 2h h
c
c r R +R
1
2
R1 = R1* +
(8.25)
(8.26)
h + 2h
c
r
*
1
R2 = R2 +
h + 2h
c
r
(8.27)
R1* is the thermal resistance between the indoor air and the inner gap
surface.
R2* is the thermal resistance between the equivalent outdoor temperature
and the outer gap surface. By using the equivalent outdoor temperature as
given in equation (7.22) the effect of the solar radiation can be accounted
for.
The solution for the surface temperatures along the gap can be given as
1
1
1
Tx +
Ti +
Tequ
R*1
R1
R2
Rx
Ts1 = (
Ti )
+ Ti
1
1
1
R1
+
+
R x R1 R2
1
1
1
Tx +
Ti +
T
Rx
R*
R1
R 2 equ
Ts2 = (
Tequ ) 2 + Tequ
1
1
1
R2
+
+
R x R1 R2
h
Rx = 2 r
hc + 2hc hr
85
(8.28)
(8.29)
(8.30)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
86

2005-10-24
-------------------------------------------------------------------------------------------------------------------
9 Examples - Heat Transfer Applications

Example 9.1
In a 200 mm thick concrete wall the one dimensional

temperature distribution at a certain time is given by
the equation
C
x
25
10. x 2
Where x=0 is at the center of the wall. The ambient
temperatures are at 20C. Looking up the material
parameters for concrete in in Appendix I we find
= 1.7 W/mK = 2300 kg/m3 c = 1000 J/kgK
24
m
22
50
The heat flow in the construction can be calculated

according to Fouriers law as
d
q x
.
x
q W/m2
dx
0
2
q 0.1 = 34 q 0.1 = 34 W/m
If the ambient temperature is 20 C we have a temperature
difference between the surface and the ambient room
50
temperature of 4 K. This would according to equation
0.1
0 m
0.1
3.25 give us a heat flow at the surface of
x
4
= 30.769 W/m2
0.13
using the standard value for Rsi. We also want to know how much energy is stored in 1
m2 of the wall if the reference temperature is set at 20 C. As the temperature in the wall
is not constant we have to find the total value by integration
0.1
J
Q
25
10. x 2 . . c d x Q = 1.135 107
0.1
Or in kilowatthours
Q = 3.243
1.135 . 107
3600. 1000
kWh
87

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Example 9.2
An outer wall is from the outside made of 120 mm brick, cellulose fiber
insulation and a 22 mm gypsum plasterboard. We want to know the
thickness of the insulation layer in order to reach a U-value of 0.2 W/m2K.
The U-value is calculated according to equ 3.32 and the thermal resistances
are given by equ 3.15 for the material layers and by equ 3.29 and 3.30 for
the surface resistances.
Input data
gypsum
0.25
brick
d gypsum
0.025
R si
0.13 R se
0.04
0.6
cellfib
0.042 d brick
0.12
Rearranging equation 3.32 we can calculate the insulation thickness

d cellfib
R se
U value
d cellfib = 0.19
R si
d brick
brick
d gypsum
.
cellfib
gypsum
We need 190 mm of insulation to reach the required U-value. We also want to find the
surface temperature on the inside when the ambient indoor and outdoor temperatures
are 20 C and -18 C respectively. Using equ 3.26 and equ 3.32 we get.
si
20
R si. 20
1
0.2
18
si = 19.012 C
Example 9.3
At a certain location we want to estimate the frost-free depth in the ground in order to
plan the foundation work for a new building. We assume that the temperature on the
surface over the year can written as a sinusoidal variation around the annual mean
temperature of 5 C with an amplitude of 10 C with minimum at the beginning of the
year. The soil is mainly clay and from appendix 1 we see that
clay
1.5 clay
1500 c clay
2000
= 1.991 10 7
clay
The temperature variation will have an
2.
clay. c clay
amplitude and angular frequency
Tamp
10
2.
365.25 . 24. 3600

88

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Period = 3.156 107

Period
365.25 . 24. 3600
= 0.446
As shown from the figure on p 29 we can

express the temperature variation with time
as
. .
T t
5 Im 0 10i . e t i
And the temperature variation at the
surface around its mean value can for the
given period and a corresponding angular
frequency be expressed simply as
T(0) = -10i
Equation 4.23 can now be used to calculate
the temperature variation with depth.
10
T(t)
0
time
T x
- .e 1
i . . x
We are now interested to find the lowest temperature at as a function of depth

and to visualize the amplitude reduction we also calculate the maximum values
15
10
Tmax x
Re T x 2
T x 2
Tmin x
Re T x 2
T x
Tmax x
Tmin x
0
0
Example 9.4
89
We see that according to this calculation

the frost-free depth should be at ca 1.6
m.

2005-10-24
-------------------------------------------------------------------------------------------------------------------
s
1
dconcrete
Lbalcony
Lfloor
dwall
The equation for a cooling fin is a useful tool to study thermal bridges. We now will
apply the equations to study heat flows and temperature distribution in an intermediate
floor slab that protrudes an insulated wall as a balcony slab. We assume that there is no
heat exchange between the wall and the slab and that in each section of the slab the
temperature is constant. The temperature on the inside is 20 C and on the outside -18
C. The wall is 300 mm thick and the concrete slab is 200 mm thick, the length of the
balcony is 1.2 m and the floor length to the center of the room is 3 m. The necessary
data for our calculation is
i
20
e
18
R si
0.13
R se
0.04
betong
1.7
L floor
3
d betong
0.20
L balcony
1.2
d wall
0.3
We now consider the slabs outside the wall as cooling fins therefore we can
calculate the heat flow between the border temperatures 1 and 2 using equ
4.53 since L is far larger than 2.0 in both cases
floor
balcony
2
R si. d betong. betong
floor. L floor = 20.18
R se. d betong. betong

balcony. L balcony = 14.552
We can express the conductances between the different temperatures as
following
betong. d betong
12
d wall
2i
betong. d betong. floor e1
betong. d betong. balcony
Since the conductances are coupled in series the temperature 2 can be calculated as
following:
90

2005-10-24
-------------------------------------------------------------------------------------------------------------------
e1
12
e1
12
2i
2 = 9.364 C
When the temperatures are known the heat flow through the construction can
be decide from any pair of temperatures and the corresponding conductance
We can also calculate the linear coefficient of thermal transmittance for the
thermal bridge
= 24.326 W/m
2i. i 2
= 0.64
W/mK
The U-value for a well-insulated outer wall is about 0.2 W/m2K which means
that the heat loss through the thermal bridge would be equal the heat loss
through 3 m2 of outer wall.
20
15
s x
10
0.5
x
When we now know the

temperature at the wall floor
junction we now use equ.
4.49 to study the
temperature distribution on
the floor surface, s.
.x
s x
i
2 i . e floor
Such low temperatures will
cause problems with bad
comfort and high levels of
relative humidity.
It is now obvious that this

construction is not acceptable both regarding surface temperatures and the
heat loss. As a further exercise try to propose improvements of this construction
and carry out a new analysis.
Example 9.5
91

2005-10-24
-------------------------------------------------------------------------------------------------------------------
A 50 mm air gap in the outer part of a construction is ventilated with air with
velocity 0.5 m/s. The average temperature in the gap is about 0 C. We want to
estimate the convective surface heat transfer coefficient hc.
Basic input data see table 5.1
air
0.0243
a
0.05
0.5
Pr
13.3 . 10 6
0.715
Assuming that the width of the air gap, b, is much larger than the thickness, a
dh
2. a
We find the Reynolds number using the hydraulic diameter as characteristic
length
u. d h
Re
Re = 3.759 103
Equ 5.17 is valid for this Reynold number if we use n=0.4
Nu
0.023 . Re0.8 . Pr0.4
Nu = 14.573
And from the definition of the Nusselt number comes

Nu . air
2
h c = 3.541 W/m K
hc
dh
Example 9.6
An inlet air duct is placed above the insulation in a cold attic space before
entering the house. The duct is 10 m long and 200 mm in diameter. The air
velocity in the duct is 2 m/s. We now want to see what the necessary U-value
for the duct wall should be if we do not want the temperature to fall from 22 C
at the inlet of the duct to 21 C where the duct leaves the attic.
Basic data
l
10
T0
22
1.205
d
0.2 u
2
Tl
21
We can now calculate the perimeter P
P
. d
and the area of the duct
d2
A
4
Rearranging equ. 5.12 gives us the U-value
Tl
u. A. . c
. ln
P. l
T0
T att
T att
1007
W/m2K
92

2005-10-24
-------------------------------------------------------------------------------------------------------------------
U = 0.564
This U-value can be reached with about 20 mm of mineral wool insulation. Note
however that this would mean that the energy content of the inlet air is reduced
by 5% with reference to the outdoor air at 0C and 50 % with reference to the
indoor air at 20 C. The duct insulation should therefore be improved
considerably.
Example 9.7
In a double glazed window the distance between the glazings, d, is 25 mm and

the height of the window, H, is 1 m. The temperature of the outer pane, T1, is
2C and on the inner pane, T2, 17C. We want to estimate the convective heat
transfer coefficient between the panes, in the air gap. Using the average
temperature in the gap we can find basic data for the air in table 5.1
air
0.025
14. 10 6
0.0035
9.8
Since this is a vertical air gap with natural convection we calculate the Grashof
number with d as the characteristic length
Gr = 4.102 104
g. . d3. T 2
Gr
T1
Equ(5.25) is valid for the Nusselt number

1
Nu
0.18 . Gr 4 .
Nu = 1.7
And the convective heat transfer coefficient can be calculated as
hc
Nu. air
d
h c = 1.7
W/m2K
Example 9.8
The temperature on the tip of the nose, Tnose is measured to be 27C and the
emittance of the nose nose is estimated to be 0.9.
93

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The total exitance at the surface, M, can be calculated by combining equ(6.1)

and equ(6.3)
M
nose. . T nose4
2
M = 414.17 W/m
This is a heat flow corresponding to little less than half a kilowatt per square
meter and the nose tip would be cooled down considerably. If we are facing
surfaces in a room at 20C the total heat flow at the nose tip due to long wave
radiation will consist of radiation from the nose tip to the walls, and the radiation
from the walls to the nose tip. Using equ. (6.34) for the radiation exchange
between two surfaces, and assuming that the nose tip is completely
surrounded by room surfaces, and the area of the walls is far larger than the
nose tip, and equ. (6.32) for the resulting emittance is valid. The net rate of heat
flow due to long wave radiation at the nose tip can therefore be calculated as.
q nose
12. . T nose4
T walls4
q nose = 37.306
W/m2K
Which is far less than the original excitance at the nose tip.
Example 9.9
At the inner surface of an outer wall of height, Hsurf, 2.5 m and with emittance, ,
0.9 we have a temperature, Tsurf, 17C. Other surfaces in the room have the
temperature, Twalls, 20C. We want to decide the combined cofficient of surface
heat transfer, hs. We start with finding the convective coefficient of surface heat
transfer. First we decide the Grashof number
Gr
g. . H surf3. T walls
T surf
Gr = 8.203 109
And then Nusselt number can be found according to equ(5.20) noting that
Pr.Gr >109.
1
Nu
0.13 . Gr . Pr 3
and from the Nusselt number the convective surface heat transfer coefficient hc
can be calculated
hc
Nu. air
H surf
h c = 2.344
W/m2K
Using equ(6.39) for the radiative surface heat transfer coefficient hr and
assuming that the condition according to equ(6.37) is valid we get
94

2005-10-24
-------------------------------------------------------------------------------------------------------------------
hr
12. . T walls4
T walls
T surf4
T surf
h r = 5.064
W/m2K
The combined surface heat transfer coefficient can now be calculated using
equ(6.44)
2
hs
hr hc
h s = 7.408 W/m K
The surface resistance is then found by
1
2
= 0.135 m K/W
hs
Which can be compared with the standard value generally used for interior
surfaces
R si
0.13
m2K/W
It is interesting to note that at the corner between two outer walls the radiative
surface heat transfer will be reduced to approximately 2.5 resulting in an Rsi
value of 0.20 m2K/W
.
95

2005-10-24
-------------------------------------------------------------------------------------------------------------------
96

2005-10-24
-------------------------------------------------------------------------------------------------------------------
10 Moisture in Buildings - Introduction

10.1 What do we mean by moisture?
Within the field of building physics moisture signifies water, (H2O) in its
different phases, vapor, liquid and ice, appearing in indoor and outdoor air,
within the materials in building constructions and on their surfaces.
10.1.1 Moisture problems
Moisture problems involve situations where the moisture content in air, on

surfaces or in materials, or the gradient in moisture content, exceeds the
limits for normal biological and mechanical decay or affects the performance
of a construction or the building in some other significant negative way.
10.1.2 Functional demands - performance

In the design process, the moisture related functional demands on a
building construction are, that it protects the interior environment from
moisture from external sources such as rain, snow and ground water and
also that the construction protects itself against moisture problems and
damages.
10.2
Exterior timber framed wall
To exemplify the importance of moisture for the design of normal building

constructions, lets look at a normal timber framed exterior wall.
97

2005-10-24
-------------------------------------------------------------------------------------------------------------------
interior cladding
vapour barrier
insulation
wind barrier
ventilated layer
exterior cladding
studs, timber frame
The construction consists from he inside of :

Interior cladding that can consist of a chip board or a gypsum board. Gives
integrity to the inner surface and serves as a basis for exterior finish and
mounting of pictures etc.
Vapor barrier that usually in such construction consists of a 0.2 mm
polyethylene foil. Is usually laid with overlaps and carefully attached to
window frames etc to provide the high degree of air tightness required in
modern buildings. The vapor barrier also has a high vapor resistance and
prevents diffusion of water vapor from the interior into the construction to
keep it as dry as possible.
98

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Timber frame, studs and mineral wool. The timber frame, sometimes
together with the adjacent layers, stands for the load bearing capacity of the
wall. The mineral wool is a highly porous material with a low thermal
conductivity and thereby contributes with most of the thermal resistance of
the wall. Since timber has a higher thermal conductivity the thermal bridge
effect of the timber can reduce the thermal resistance of the layer by 5-15 %.
In well insulated constructions the frame can split to avoid continuous
thermal bridges through the insulated layer.
Wind barrier. Pressure differences along the construction surfaces could
easily create air movements in the highly porous insulation layer and
cavities and imperfections in the insulated layer which could jeopardize the
insulation function. The function of the wind barrier is therefore to prevent
this. However the wind barrier has to have high permeability to effectively
allow diffusion of moisture from the construction to the outdoor air to keep
the construction as dry as possible. The wind barrier can also be exposed to
rain or snow penetrating through the exterior cladding. Typically wind
barriers are made of tarred low density fiberboards, gypsum of a special
type, but also from polystyrene insulation and some special foils with
properties similar to modern outdoor garment fabric such as Goretex.
Ventilated layer. The ventilated layer usually has openings at the bottom
and the top of the wall so that temperature differences and differences in
wind pressure along the surface can generate air flow in the air gap. The air
gap has the function to even out the pressure difference over the exterior
cladding to reduce the risk for rainwater penetration, to dry out the
excessive moisture in the outermost layers of the construction and to drain
out water that penetrates the exterior cladding.
Exterior cladding. The exterior cladding is, besides being the exterior finish
of the building, the outermost protection against the elements. For as an
example a wooden panel the important features for the moisture protection
is that the paint and the details are such that the driving rain water runs of
effectively and that the ventilation through the air gap behind reduces
pressure differences and moisture gradients.
10.3 Boundary conditions

In ventilation technology the water vapor content of air is expressed as
humidity by mass, x, kg of water per kg of dry air. In building physics a
common convention is to express the vapor content as humidity by volume,
v, kilograms of water per m3 of air.
v = pv/(461.4.T)
(10.1)
99

2005-10-24
-------------------------------------------------------------------------------------------------------------------
pv is the partial water vapor pressure, Pa

T is the thermodynamic temperature, K
The reason for this is that transfer of moisture in materials can be
expressed with vapor content by volume as a potential, independent of
temperatures.
Obviously, the relative humidity can be expressed as
=pv/psat = v/vsat
(10.2)
The humidity of air by volume at saturation can be given with adequate

accuracy with the formulae
v sat( > 0) =
v sat( 0) =
v sat( > 0) =
288.68 (1.098 +
)8.02
100
( + 273.15) 461.51
4.689 (1.486 +
(10.3)
12.3
)
100
( + 273.15) 461.51
288.68 (1.098 +
)8.02
100
( + 273.15) 461.51
4.689 (1.486 +
12.3
100
( + 273.15) 461.51
which are tabulated in Appendix II.
v sat( 0) =
For a typical winter period, the indoor conditions can be

T = 20 C, vsat = 17.28 g/m3,
= 40 % , v = 6.91 g/m3
(10.4)
and the outdoor conditions

T = 0 C, vsat = 4.84 g/m3,
= 90 % , v = 4.36 g/m3
(10.5)
The saturation temperature for the indoor air is 5.3 C and for the outdoor
air -1.3 C. By placing the vapor barrier on the inside of the insulation we
make sure that in the whole construction the temperature is well above the
saturation point. If we place the vapor barrier on the exterior side of the
insulation the colder parts of the insulation will reach equilibrium with the
indoor air, resulting in condensation of vapor in these areas.
100

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The above example shows that moisture protection is a very significant

element in the design of exterior constructions.
The following lectures will provide the theoretical basis for moisture
transfer in constructions and how calculation rules can be expressed in the
design process.
10.3.1 Outdoor temperatures and humidity

Outdoor air temperatures and relative humidity are available, for a number
of locations in Sweden, on weather files with meteorological data given hour
by hour. For moisture calculations we often use monthly means to and carry
out our calculations with successive monthly balances or even longer
periods like winter and summer. Monthly means for temperatures and
relative humidity for a few locations are given in Table A.5.
10.3.2 Indoor humidity

Over a small time period dt the moisture stored in the indoor air equals the
difference in moisture generated,G, kg/s, in the building minus the moisture
loss due to ventilation
(10.6)
V.dvi = dt(G - nV(vi-ve))
vi is the interior vapor content and ve is the exterior vapor content. V is the
building volume, m3, and n, h-1, is the number of air exchanges per hour.
vi+(1/n) dvi/dt = G/(nV) + ve
(10.7)
If G is constant,
vi = Ae-nt + G/(nV) + ve
(10.8)
If the moisture production G starts at time zero and vi(0) =ve

A = -G/(nV)
(10.9)
vi = G/(nV)(1- e-nt ) + ve
(10.10)
For constant moisture generation and ventilation rate over a long time, e-nt
tends towards zero
vi = G/(nV) + ve
(10.11)
101

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The term G/(nV) gives the moisture production per m3 ventilation air.
Typical values for dwellings in Sweden with normal ventilation rates are
between 2 and 4 g/m3 and for wet rooms such as shower rooms a value of 6
g/m3 can be expected.
10.4 Moisture in building materials

There are several ways to express the moisture content in materials. By
evaporable water is meant water that is not bound in chemical or crystalline
form.
w = moisture content, mass by volume, kg/m3.
Mass of evaporable water divided by volume of material
= moisture content, volume by volume, m3/m3.

Volume of evaporable water divided by volume of material
u = moisture content mass by mass, kg/kg.
Mass of evaporable material divided by dry mass of material
102

2005-10-24
-------------------------------------------------------------------------------------------------------------------
10.4.1 Building moisture

Many building materials have a higher moisture content in the initial state
at the end of the construction phase than when in its normal long term
environment indoor or within the building components. Wet concrete,
plastering and timber are examples of materials that normally, in the initial
state, have a moisture content that has to be dried out during the
construction phase and during the first months or years that the building is
in operation.
10.4.2 The sorption curve

For hygroscopic materials, the relation between relative humidity of
surrounding air and the moisture content of the material at equilibrium is
fairly independent of the surrounding temperatures. The sorption curves
which are commonly called sorption isotherms are, for a number of
materials, given in Table A6.
10.5 Rainwater
The total precipitation and intensity of rain serves as a basis for the
dimensioning of drainage systems etc. Usually we do not assume that
vertical rain affects the thermal or hygric properties of the building
components, since the roof drainage systems are usually made of water
tight materials. However, during a rainfall in combination with wind, the
direction of the raindrops deviates from the vertical direction resulting in a
horizontal velocity component. The amount of rain on a vertical surface,
often referred to as the driving rain, can then be expressed
S= N
103
u
uv
(10.12)
S is the quantity of
driving rain, kg/m2
N is the precipitation,
kg/m2
u is wind velocity, m/s
uv is the vertical
component of the velocity
of raindrops, m/s

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The maximum velocity of the raindrops in the vertical direction is ca 10 m/s.

Due to the large variations in air velocities, the amount of driving rain
hitting the facades can vary over the surfaces. During a rainstorm, one can
clearly see how a pattern forms over the surface resulting in wet areas
around corners and on the upper parts of the facades. The driving rain on
those parts can be estimated from equ. (10.12) using the free wind velocity.
10.6 Ground water
The parts of the building below ground level can naturally be exposed to
moisture in different forms. Water is added to the ground under the house
from potential flow from higher levels and from precipitation at the location
of the house. The groundwater level around the house is kept low with
special drainage pipes, but the water level under the house can rise above
the level of the pipes due to capillary action.
Precipitation
Drain
Drain
Potential
flow
Capillary flow
Normally the relative humidity in the ground is assumed equal to 100 %
and the lowest part of the fundaments of the building should be supposed to
be exposed to water.
104

2005-10-24
-------------------------------------------------------------------------------------------------------------------
11 Moisture transfer in constructions - diffusion

11.1 Moisture transfer - introduction
In chapter 10 we have treated boundary conditions and the relation between

the moisture content of a material at equilibrium and the surrounding
climate. Insulated building constructions normally have their surfaces
exposed to two or more different climates which results in a distribution of
moisture in a similar way to that of temperatures. Our next task is
therefore to develop relations between the boundary values, moisture flow,
moisture content, and material properties to predict the state of moisture
and temperatures in different parts of the construction.
11.2 Examples of applications
Predicting the preconditions for microbiological growth in constructions.

Drying of wood
Decay of moisture content in concrete floors before carpeting
Estimating risks for condensation and biological growth on surfaces
Estimation of moisture related movements in materials exposed to varying
climatic conditions
Drying out of rain wet structures - energy loss and risk for freezing
damages.
Design of ventilated constructions
Effect of moisture content on fire resistance
11.3 The physics of moisture flow in porous materials
Moisture is transferred in building materials and building constructions

mainly by four processes, diffusion, air convection, capillary suction in
fine pores, and flow of water due to external forces such as wind
pressure and gravity. As long as we only have vapor at low relative
105

2005-10-24
-------------------------------------------------------------------------------------------------------------------
humidity, moisture transfer takes place by diffusion and convection. At the

state of saturation within materials and on surfaces or just below, capillary
and gravity flows occur.
11.4 Diffusion
Assume that we have a mixture of n gases in a volume, each having a

concentration cj which can be expressed by number of molecules per volume
or in kg per volume. At equilibrium, the concentration cj will be evenly
distributed over the volume. If the concentration of cj at a certain time point

is not evenly distributed, a flow of gas molecules will take place in order to
reach equilibrium. The gas flow rate g will be proportional to the gradient
in concentration in the same way as the heat flow rate is proportional to the
gradient in temperature.
g = -Dj.grad cj
(11.1)
This is known as Ficks law

g = gas flow rate, kg/m2s.
Dj = coefficient of diffusion for the gas number j, m2/s
cj = concentration of gas nr j, kg/m3
We note that the symbol v for humidity of air is in reality the concentration
of water vapor molecules given in kg/m3.
The water vapor flow rate can then be expressed as
(11.2)
g= -D grad v
2
D = coefficient of diffusion water vapor in air, m /s

or in one dimension
(11.3)
g = - D dv/dx
Empirically, D varies with temperature as
D = (22.2+0.14).10-6
(m2/s)
(11.4)
106

2005-10-24
-------------------------------------------------------------------------------------------------------------------
11.5 Diffusion in porous materials
Ideally we can assume that the section of air that can transfer vapor by
diffusion in a porous material is reduced to the sum of the cross sections of
the pores. As we will explain later, this is not that simple, since liquid water
will form in the pores below saturation point and in fibrous materials and
the size of the pores can depend on the relative humidity. By a good
approximation, the diffusion coefficient for vapor in a material can be
related to the coefficient of diffusion in air by a constant describing the
resistance of the material to water vapor flow such that the vapor
conductivity v for a certain material is given as
v=D/
(11.5)
Examples of are 2-10 for mineral wool, 20-100 for concrete and 103 - 106
for plastic foil or glass.
The density of vapor flow rate through a porous material layer of thickness
dx and the difference in vapor content dv between the surfaces can thus be
expressed as
g = - v dv/dx
(11.6)
The water vapor conductivities for different building materials are given in
table A:
The analogy with Fouriers law for heat conduction is now evident
g = - v dv/dx
q = - dT/dx
(11.7)
This means that the theory we have developed for heat conduction can be
applied to vapor flow and distribution of humidity in constructions.
v
1
Using this analogy we can define the vapor

resistance Z for a given layer with thickness d and
v(x)
vapor conductivity v
Z= d/v
(s/m)
(11.8)
and for a layer with vapor conductivity v, at

steady state we express the vapor flow rate as
d
107

2005-10-24
-------------------------------------------------------------------------------------------------------------------
g =v(v1-v2)/d = (v1-v2)/Z
(11.9)
For some layer types of a complex structure of different material the vapor
resistance Z is given directly as in table F12:4.
And for the humidity distribution in a multi-layer construction
vn
vn-1
v2
v1
gn
g1
Z1
Z2
Zn-1
k1
v k = v1
(v 1 v n ) Z j
j=1
(11.10)
Ztot
11.6 Moisture surface resistances
The exchange of vapor between the surfaces of the construction and the
ambient air is governed by the same equations as the ones for convective
heat transfer. Lewis law gives the coefficient of surface moisture transfer
and the surface moisture resistance Zs, in relation to the convective heat
transfer coefficient hc as
= hc /c
g = (vs-va) = (vs-va)/Zs
(11.11)
(11.12)
vs = humidity at the surface, kg/m3
va = humidity of ambient air, kg /m3

Normally, the surface vapor resistance is very small compared to the vapor
resistance of material layers in a construction.
108

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Values for Zs that are normally used are Zsi = 360 s/m on the inside of a
construction and Zse = 60 s/m on the outside. These are rather small
compared to the vapor resistance of material layers and generally only of
interest when estimating the rate of surface condensation and evaporation.
11.7 Distribution of relative humidity
The distribution of humidity gives us the moisture content of the air in the
pores of a material. For the following reasons we are interested to know also
the relative humidity.
The risk for damages in a building material in the hygroscopic range is
usually related to the relative humidity. Fungi growth starts at 70-80 % RH
at temperatures over 5 C, steel corrosion starts at 50 - 60 % RH, carpet
glue decays at relative humidity over 90 % etc.
Our diffusion theory is only valid at a relative humidity up to ca 95 %. When
the air in the pores is close to saturation capillary water is formed in the
pores and the moisture transfer is governed by new processes.
The moisture capacity of porous materials in the hygroscopic range is
strongly related to the relative humidity.
A calculation of the moisture transfer in a construction in the hygroscopic
range can be done in the following steps:
1. From the ambient boundary temperatures and the thermal properties of
the layers calculate the temperature distribution T(x).
2. From the ambient boundary humidity and the vapor conductivities and
the vapor resistances of the material layers given in Appendix III, calculate
the humidity distribution v(x).
3. From the temperature distribution the distribution of the humidity of
saturation of the air in the pores vsat can be found, either from table
A: or from the approximate expression
a (c +
)b
100
v sat ( ) =
( + 273.15) 461.4
> 0 a = 288.68 b = 8.02 c = 1.098
0 a = 4.689 b = 12.3 c = 1.486
109
(11.13)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
4. Now the relative humidity distribution can be found as
(x)=
v (x )
.
v sat ( x )
(11.14)
5. If the relative humidity is larger than 1.0, we are outside the possible
limits. Condensation occurs and the pores will be partly filled with water.
The pure diffusion theory is no longer valid. If the diffusion is taking place
on the border between two material layers or at the inner surface we can
put v equal to vsat at the border and make a new calculation for each part
of the construction where v at the location of the saturation now serves as
boundary condition for both sides.
6. When the distribution of humidity is known in the construction, the
density of vapor flow rate g at the different surfaces can be calculated from
equ (11.12). If a condensation occurs within the construction, the rate of
condensation gcond can be found as the difference between the rate of
vapor flows to and from of the construction
g kond = si (v i v si ) se (v se v e )
(11.15)
Or if the surface layers are not included in the calculation the flows are
calculated over the layers closest to the surfaces
g kond =
v 1 v 2 v n1 v n
Z1
Z n 1
(11.16)
110

2005-10-24
-------------------------------------------------------------------------------------------------------------------
12 Moisture transfer in constructions - capillary

transfer
12.1 Water in porous materials
We have mainly three ways of storing water within porous materials.

At a low relative humidity the air in the pores contains some water vapor.
Due to forces between the water molecules and the material in the pore
walls, a layer of water molecules is adhered to the wall. The bond to the
next layer will be weaker with increasing distance so that the amount of
moisture stored in this way is limited. When the partial pressure of the
water vapor in the surrounding air increases, the adhesion forces at concave
surfaces dominate the pressure difference between the saturated surface
and the air in the pores. This leads to condensation at the concave surfaces
even though the air in the pores is not saturated.
Capillarity
2r
Adhesion forces
Circular pipe
in water
Water molecule
The molecule next to a concave surface will have a shorter mean distance to
the surface and therefore the adhesion forces will be stronger than for a
molecule at the same distance from a plane surface. We consider a single
pore in a material idealized as a vertical pipe with a representative
diameter. The lower end of the pipe is in contact with free water and a
water gauge is drawn up into the pipe with capillary forces. The angle is
dependent upon the properties of the liquid in the pipe and the material of
the inner pipe surface. For water and most building materials can be
assumed equal to 0. The water molecules closest to the wall have a strong
111

2005-10-24
-------------------------------------------------------------------------------------------------------------------
bond to the surface. The force that holds up the water gauge is then
dependent on the surface tension along the perimeter of the pipe. This
upward force can be expressed as
Fup = 2 r cos
(12.1)
is the surface tension N/m
The pressure drop over the top surface of the water gauge is given as
p=
F 2 r cos 2 cos
=
=
A
r2
r
(12.2)
The downward force will be due to gravity action on the water gauge
Fdown = g h r2
(12.3)
At equilibrium, balance of the forces will give
g hequ r2 = 2 r cos
(12.4)
2 cos
g r
(12.5)
H = hequ =
hequ is the height of the water gauge at equilibrium also denoted as

maximum height H.
As cos can be set equal to 1 and given that for water at 10 C the surface
tension is 0.074 N/m and that is equal to 1000 kg/m3 the maximum
height can be found to be approximately
H = hequ =
1.5 10 5
r
(12.6)
12.2 Capillary flow
For our purposes we are not only interested in the fact that the pores
eventually will be filled up. This is a transient process and since surfaces
can be exposed to free water for a limited period of time the water velocity
in the pores is of interest.
112

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The friction between water with velocity u and the inner surface of a pipe
with radius r can be expressed as
= 4u/r
(N/m2)
(12.7)
is the dynamic viscosity (Ns/m2)

Assume that we have a water gauge in a capillary pipe with the height z
and it has had contact with water since t = 0. The force balance will now be
Upwards
Downwards
Surface tension force =
Weight
gzr2
2rcos =
+
+
Friction
2rz.4u/r
(12.8)
Solving for u gives

u=
r cos 1
r g
(
)
4
z 2 cos
(12.9)
Substitution by equ (12.5) for maximum height

u=
r cos 1 1
( )
4
z H
(12.10)
For a horizontal pipe the effect of gravity equals zero and the equation for
the velocity becomes
u=
r cos
3 z
(12.11)
The relation between the velocity and length is given as

u=
dz r cos
=
4 z
dt
(12.12)
The differential equation (12.12) is fulfilled with z as a function of time, t

z=
r cos
t
2
(12.13)
This can be written as
113

2005-10-24
-------------------------------------------------------------------------------------------------------------------
z = (t/m)
(12.14)
t = mz2
(12.15)
m is the capillary resistance, s/m2

Examples: Brick m = 0.5.106 , aerated concrete m = 10.106 s/m2
12.3 Water absorption
The resulting water content per unit area of the material surface in contact
with water can be calculated from the porosity of the material pk that is the
fraction of pores in the total material volume
Water molecules adsorbed

to pore surfaces
Capillary condensation to a
concave surface
114

2005-10-24
-------------------------------------------------------------------------------------------------------------------
G= zpk = . pk(t/m) = B (t)
(12.16)
B is the coefficient of capillarity, kg/(m2.s)

Due to capillary suction the relative humidity in the pores seldom reaches
higher than 98 %, the hygroscopic limit. The terms humidity and relative
humidity now lose their meaning and the state of moisture is now
characterized with the moisture content w, kg/m3 which includes the vapor
and the water held by capillary and van der Waals forces related to the
volume of the materials, not only the pores.
Over the hygroscopic limit, added moisture can only exist in the form of
water. Gradually the pores will fill up and water will be distributed along
the pores by capillary forces. When the pores fill up to form a continuous
network of filled pores, moisture transfer in liquid phase can start. We say
that we have passed the critical moisture content, wcrit.
12.4 Capillary moisture transfer within materials
Our analysis of capillary water transfer above assumed one end of the pore
in contact with free water. In reality this does not have to be the case.
Imagine a porous material as a system of pores of different diameters
forming a complex network of pores through the material. For each pair of
pores that are in contact, water will be transferred to the pore with the
smaller diameter since the pulling pressure is proportional to 1/r.
p1
p2
p= 2 cos
r
(12.17)
As a result of this, water will flow from the pore with the larger diameter to
the pore with the smaller diameter until the latter is filled up. In this way
water will be distributed in the material to fill the pores with the smallest
diameter.
115

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The precondition for such a transfer to take place is that the water content
of the pore system is sufficiently high so that the pores become connected
through the material. This is called the critical moisture content Wcrit. A
common convention is to express the moisture transfer in a material as a
sum of two processes
g = - dv - kw dw
dx
dx
(12.18)
Under the critical moisture content, kw 0 and the moisture transfer

takes place by diffusion only. Over the critical moisture content kw is high
and the diffusion term becomes insignificant.
A major drawback for moisture calculations is the lack of reliable data for
-12
2
kw. A typical value for concrete is kw 200 . 10 m /s between 90 and 100
% RH.
116

2005-10-24
-------------------------------------------------------------------------------------------------------------------
13 Moisture transfer in constructions convection - ventilated constructions
13.1 Water flow through porous materials
When a material is exposed to external water pressure the pore system can
be completely filled and capillary forces no longer exist. The transient
process can be analyzed in the same way as for filling up pores under the
influence of gravity. In the steady state the flow of water through a material
layer can be expressed as
g=
B0 p
d
kg/m2.s
(13.1)
B0 is the permeability of the material, kg/m

is the kinematic viscosity of water (/), m2/s
p is pressure difference, Pa
d is thickness of material layer, m
For permeability of different materials, see table A:1 in Appendix I.
13.2 Water flow through a crack.
Water can be driven through a crack in the building surface by water or

wind pressure. The flow of water through a crack can be expressed as
p2
d
b
G=
l b3 p
12 d
p1
kg/s
(13.2)
l is the length of the crack, m

b is the width of the crack, m
d is the depth of the crack, m
117

2005-10-24
-------------------------------------------------------------------------------------------------------------------
is the density of water kg/m3

When the width of the crack is so small that the capillary height of the
crack becomes significant compared to the depth of the crack this will affect
the transient flow in the crack but as soon as the crack is filled to the end
the capillary force drops. In vertical gaps with upward flow the external
pressure and the capillary force may add together to fill the crack and cause
intermittent water flow over the upper edge.
13.3 Moisture flow by air convection
We have two air volumes V1 and V2 with humidities v1 and v2 respectively.

If a small volume dV1 from V1 is moved over to V2 and a small volume from
V2 is moved back to V1 the resulting moisture flow between the volumes 1
and 2 will be
G=
dV1
(v v 2 ) = L12 (v 1 v 2 )
dt 1
(13.3)
L12 is the air flow, m3/s.

If a quantity dV1 is moved to a location 2 and cooled to a temperature T2
which is lower than T1sat condensation will occur. The rate of condensation
will be
G=
dV1
(v v 2 ) = L12 (v 1 v 2 )
dt 1
(13.4)
The air flow through a material layer of thickness d, m, between the

environments 1 and 2, and with permeability Bo can be calculated as
L = u A = A
B0 dp
B (p p2 )
= A 0 1
d
dx
(13.5)
u is the apparent velocity of air in the material, m/s.

For a narrow crack of width b and length l through a material layer of
thickness d and for flow with low Reynolds number
L = u b l =
(p1 p2 ) b3 l
12 d
(13.6)
is the dynamic viscosity of air Ns/m2
118

2005-10-24
-------------------------------------------------------------------------------------------------------------------
13.4 Heat and moisture balance in a ventilated air gap.
When the air passes an air gap in a construction the moisture exchange
with the surfaces will change the state of the air. The resulting moisture
flow from the construction to the passing air can be expressed as
G = u . b . l (vin - vout) = L (vin - vout)
(13.7)
v out
u
v in
v
u
vx
x+dx
x+dx
x
x=0
Moisture balance for the element of the air gap with length dx gives
u b v x + 2 (v s v x ) dx = u b v x +dx = (v x + dx
119
v
) u b
x
(13.8)

2005-10-24
-------------------------------------------------------------------------------------------------------------------
2 (v s v x ) dx = u b
vx +
v
dx
x
(13.9)
u b v
= vs
2 x
(13.10)
x 0 v v0
x v vs
v = v o + (v s v o )(1 e
2
ub x
(13.11)
Equation (12.11) is valid if the distribution of vs is even along both surfaces.

For an air gap where the vs for one surface is considerably higher than the
other this can be approximately dealt with by a reduction of .
can be estimated from Lewis law
hc
c
(13.12)
and the equations for calculation of hc in an air gap has been given in
previous lectures.
In most cases the construction is ventilated with out-door air with low
temperature and low initial humidity, To, vo. The surfaces of the air gap
being in an insulated construction are usually warmer than the inlet air and
the inlet air is heated on its way at the same time as the humidity increases
along the air gap.
The solution for the air temperature and the surface temperatures
of the air along the gap has been given in chapter 10 equ(10.2310.27).
120

2005-10-24
-------------------------------------------------------------------------------------------------------------------
14 Applied Moisture Transfer - Damages

In building engineering practice the problems that we come across,
concerning building physics, are more often of a complex nature. We have to
involve several different processes in our study and to form balances
including the transfer of heat, air and moisture. A difficult task is usually to
decide upon the level of modeling used in the study. Can we consider a
volume as limited by homogenous surfaces with one-dimensional heat and
moisture flow or do we have to consider the two-dimensional heat flow at
the corners at the edges or even the three-dimensional heat flow at the
corners? Such questions have of course no general answer. We have to be
able to decide if more detailed calculations will significantly improve our
results and also if it is meaningful considering the uncertainties in input
data, such as boundary conditions, material properties and dimensions.
14.1 Attic ventilation

An approximate procedure can be as follows
Find monthly mean values for the boundary conditions such as
temperatures, solar radiation and wind velocity.
Identify the wind form factors for the roof openings.
Calculate the airflows and resulting pressure in the attic neglecting
the airflow from the indoor environment.
Estimate the pressure difference over the ceiling to the attic and
calculate the airflow through the ceiling based on the estimated air
tightness.
Calculate the resulting temperature and moisture content of the attic
air assuming no moisture flow through the outer roof.
Calculate the inner surface temperatures of the outer roof.
121

2005-10-24
-------------------------------------------------------------------------------------------------------------------
T sky
Tequ1
T equ2
L2
T s1
u0 p0
L1
T s2
Ta V a p a
L4
Te v e
Ti v i
L3
Calculate the moisture flow between the surface and the attic air
assuming 100 % RH at the surface and estimate the amount of water
stored in the inner surface material of the outer roof at the end of
each month.
From the sorption curve the relative humidity can be estimated and
from the relative humidity and the surface temperature the risk for
mould growth and rot can be evaluated.
122

2005-10-24
-------------------------------------------------------------------------------------------------------------------
14.2 Solar condensation

A phenomenon that is called solar condensation has appeared in timberframed walls with exterior brick
cladding. During long periods of
driving rain the pores of the
Chipboard
bricklayers will eventually be
PE foil
Minwool
water filled by capillary suction.
Gypsum
Air gap
When the sun starts shining
Brick
again the brick can be warmed up
to 30 - 40 C. When the outmost
brick surface has dried out the
drying out of the brick layer will
be governed by diffusion. The
ventilated air gap does not have
enough capacity to dry out the
inner brick surface and the air
will remain at 100 % RH for a
long time. It can be shown by
calculations that the diffusion from the air gap towards the PE foil will be
sufficiently large so that drops can form at the foil surface, which eventually
can appear as free water at the floor corner, or cause damages within the
building. The problem can be studied in the following steps.
Calculate the time for the brick layer to be water filled with a
continuous water film on the surface
Estimate the solar radiation and resulting exterior surface
temperature
Calculate a temperature distribution in the wall with solar radiation
Calculate the resulting moisture flow towards the foil
Estimate the resulting amount of water that can be collected at the
foil.
14.3 Smell of fungi, the search for possible reasons

A common first indication that moisture damages are present in building
constructions is a growing smell of fungi that contaminates the indoor air
and eventually also the fabric of the interior and clothes etc. Usually the
source is fungi growth on surfaces in construction parts where relatively
high temperature and relative humidity is present at the same time. Since
123

2005-10-24
-------------------------------------------------------------------------------------------------------------------
we have under pressure in the building at most times, especially at the

ground floor, contaminated air can pass through the construction into the
building where and if we have failures in air tightness. In case of more
severe damages, emission gases can also pass through constructions and
vapor barriers by diffusion and cause smell on the inside. Rot in the load
bearing members of the structure can of course also lead to structural
failure such as when parts of the roof give in under heavy snow loads. The
problems usually appear in the following places:
14.4 The wooden foot rail

The situation occurs
where a timber-frame
outer wall rests on a
masonry floor slab or
foundation beam. There
Foot rail
are several different
processes that can cause
high relative humidity in
the foot rail. If we have a
semi permeable weather
shield there will be some
water transport on the
inside on the outer
cladding or on the wind
protection. If this leakage
water is not drained out
properly at the bottom of
the wall the foot rail will be in contact with free water and we will have
capillary suction into the wood.
The solar condensation, which is treated in 16.1.2 can in this case be a
contributing factor. With condensation against the interior vapor barrier
free water can run down on the foot rail. Also when the moist exterior
cladding is heated with solar radiation the foot rail standing on the concrete
foundation will remain relatively cool, and we can have diffusion directly
from the bottom part of the exterior cladding to the foot rail.
If the fundament is in contact with ground water or drainage water on the
outside, the foundation under the foot rail can be moist due to capillary
moisture transfer. If there is not a sufficient protection with a capillary
breaking material between the masonry and the foot rail there will be a
further capillary moisture transfer into the foot rail with the growth of
mildew and rot fungi as a result.
124

2005-10-24
-------------------------------------------------------------------------------------------------------------------
14.5 Wooden floors on concrete slabs

A popular construction
in the seventies was the
so called slab on grade,
where an in situ cast
concrete floor slab was
laid out on the ground
and then a suspended
floor construction with insulation between wooden beams would be built on
top. The concrete slab is thereby in thermal and hygric balance with the
ground below and has to be assumed to have almost 100 % relative
humidity. The bottom part of the wooden floor construction will therefore
have high relative humidity since the wood construction would not be
protected against vapor diffusion from the inside. In cases where there is a
capillary contact between the concrete slab and the woodwork the decay of
the construction can happen very fast.
14.6 The crawl space paradox
radiation
convection
A traditional founding of buildings in Scandinavia has been to use crawl

space foundations. The crawl space is normally an un-insulated, unheated
space, less than 1 m high, ventilated with outdoor air. Above the heated
space there is an insulated, suspended floor usually made of wooden beams.
More often there is a vapor barrier in the bottom floor of the crawl space to
prevent moisture transfer from the ground into the crawl space. In
wintertime the crawl space is well ventilated with air with a rather small
vapor content since the temperature in the crawl space is higher than for
outdoor air the relative humidity is kept rather low. In summertime
however the earth mass under the house is still cold after the winter. The
exchange of long wave radiation between the floor of the crawl space and
the bottom of the suspended floor above keeps the surface temperature of
125

2005-10-24
-------------------------------------------------------------------------------------------------------------------
the floor construction towards the surface low. The outdoor air summertime
has high vapor content and we get high relative humidity and even
condensation at the surface. Since the temperatures in summer are
relatively high this will lead to fast growth of mildew fungi on the wooden
surfaces. Since we normally have under pressure at the bottom floor in the
house, contaminated air will be sucked into the building through leakages
in the floor construction.
14.7 High relative humidity indoors

High relative humidity indoors can be either due to excessive moisture
protection or insufficient ventilation. During the first year of operation
moisture from building materials can also be a contributing factor.
With high relative humidity and high vapor content in the indoor air there
will be a risk for still higher relative humidity or even condensation at
surfaces with lower temperatures such as at thermal bridges in the
insulated envelope, behind upholstered furniture, around cold water pipes
etc. With increased indoor humidity the risk for condensation and moisture
damages due to air leakage through the envelope will also be increased.
14.8 Water leakage

External water leakages are most common in roof details and around
windows and doors. Internal leakages are most common where water in
liquid form is present such as around the kitchen basin with dishwasher,
around bath and shower places and around floor gullies. In public buildings
water can penetrate the flooring when the floors are being washed.
126

2005-10-24
-------------------------------------------------------------------------------------------------------------------
15 Appendix I. Data for Calculations

Table A1 Thermophysical properties of materials.
The values given are typical values to be used for calculations within the
course. Observe that normally the materials can have significant variations
in properties based on different density, moisture content etc.
Material
Thermal
Density
conductivity
Specific heat
Perme-
capacity
ability
B0
W/mK
kg/m3
J/kgK
m2
Glass
1.0
2500
750
Ice at -10oC
2.3
920
2000
Ice at 0oC
2.2
900
2000
Fresh snow
0.05
100
2000
Compact snow
0.6
500
2000
Water at 10oC
0.6
1000
4190
Air
0.025
1.23
1008
Carbon dioxide
0.014
1.95
820
Argon
0.017
1.7
519
Krypton
0.009
3.56
245
Xenon
0.0054
5.68
160
Aluminum alloys
160
2800
880
Copper
380
8900
380
Cast iron
50
7500
450
Steel
50
7800
450
Stainless steel
17
7900
460
Rubber
0.17
1200
1400
Plastic
0.25
1700
1400
Linoleum
0.17
1200
1400
Polycarbonate
0.2
1200
1200
PVC
0.17
1390
900
Polypropylene
0.22
910
1800
Foam rubber
0.06
70
1500
Gypsum plasterboard
0.25
900
800
127

2005-10-24
------------------------------------------------------------------------------------------------------------------Material
Thermal
Density
Specific heat
Perme-
capacity
ability
B0
W/mK
kg/m3
J/kgK
m2
Plaster, gypsum/sand
0.7
1600
1000
10.10-15
Plaster, cement /sand
1.0
1700
1000
2.6.10-15
Soil, clay or silt
1.5
1500
2000
Soil, sand and gravel
2.0
2000
1000
Basalt
3.5
2900
1000
conductivity
Granite
3.5
2600
1000
Natural pumice
0.12
400
1000
Construction timber
0.13
500
1600
0.2.10-15
Cement particle board
0.23
1200
1500
5.0.10-15
Chipboard
0.14
600
1700
5.0.10-15
Oriented strandboard
0.14
600
1700
5.0.10-15
Plywood
0.13
500
1600
4.0.10-9
Mineral wool
0.036
30
1000
4.0.10-9
Expanded polystyrene
0.036
25
1000
5.0.10-9
Polyurethane
0.022
25
1000
Cellulose fiber insulation
0.042
50
1000
Concrete
1.7
2300
1000
1.0.10-15
Brick
0.6
1720
1000
70.10-15
Aerated concrete
0.10
400
1000
1.8.10-12
Exp clay concrete
0.10
400
1000
1.0.10-9
128

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Table A2. Hygric properties of materials.

The values given are typical values to be used for calculations within the
course. Observe that normally the materials can have significant variations
in properties based on different density, moisture content etc.
Material
Density
Vapor
Capillary
Coefficient of
conductivity
resistance
capillarity
kg/m3
106 m2/s
s/m2
kg/m2s-0.5
Glass
2500
Ice at -10 C
920
Ice at 0 C
900
Fresh snow
100
Compact snow
500
Water at 10 C
1000
Air
1.23
25
Carbon dioxide
1.95
25
Argon
1.7
25
Krypton
3.56
25
Xenon
5.68
25
Aluminium alloys
2800
Copper
8900
Cast iron
7500
Steel
7800
Stainless steel
7900
Rubber
1200
Plastic
1700
Linoleum
1200
Polycarbonate
1200
PVC
1390
Polypropylene
910
Foam rubber
70
Gypsum plasterboard
900
3.6
Plaster, gypsum/sand
1600
0.25
Plaster, cement /sand
1700
0.2
0.03
Soil, clay or silt
1500
Soil, sand and gravel
2000
Basalt
2900
Granite
2600
0.05
129

2005-10-24
------------------------------------------------------------------------------------------------------------------Material
Density
Vapor
Capillary
Coefficient of
conductivity
resistance
capillarity
kg/m3
106 m2/s
s/m2
kg/m2s-0.5
Natural pumice
400
Construction timber
500
0.8
0.004
Perp. to fibers
0.016
Cement particle board
1200
1.0
Chipboard
600
0.3
Oriented strandboard
600
Plywood
500
0.9
Mineral wool
30
20
Expanded polystyrene
25
1.3
Polyurethane
25
0.7
Cellulose fiber insulation
50
17
Concrete
2300
0.25
0.02
Brick
1720
0.37
0.08
Aerated concrete
400
Exp clay concrete
400
* To avoid misunderstanding it is made clear that the entries for vapor

conductivity are to be multiplied by the 10-6 to get the actual value. The
vapor conductivity for concrete is then 0.25.10-6 m2/s.
Table A:3 Miscellaneous thermophysical properties

Melting heat: Ice - water
Evaporation heat: Water - vapor
334.103
J/kg
2500.103 J/kg
130

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Table A:4 Water vapor content at saturation, vsat in g/m3. The table
is calculated by the approximate formula (13:13).
TC
-20
-19
-18
-17
-16
-15
-14
-13
-12
-11
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
0
0.89
0.97
1.06
1.16
1.27
1.39
1.52
1.65
1.80
1.97
2.14
2.33
2.53
2.75
2.99
3.25
3.52
3.82
4.14
4.49
4.86
4.85
5.19
5.56
5.95
6.37
6.80
7.27
7.76
8.28
8.83
9.41
10.02
10.67
11.35
12.07
12.83
13.63
14.48
15.37
16.30
17.29
18.32
19.41
20.55
21.75
23.02
24.34
25.73
27.19
28.71
30.31
31.99
33.75
0.1
0.88
0.96
1.05
1.15
1.26
1.38
1.50
1.64
1.79
1.95
2.12
2.31
2.51
2.73
2.97
3.22
3.50
3.79
4.11
4.45
4.82
4.88
5.23
5.60
5.99
6.41
6.85
7.32
7.81
8.33
8.89
9.47
10.09
10.74
11.42
12.15
12.91
13.72
14.57
15.46
16.40
17.39
18.43
19.52
20.67
21.88
23.14
24.48
25.87
27.33
28.87
30.48
32.16
33.93
0.2
0.87
0.95
1.04
1.14
1.25
1.36
1.49
1.63
1.77
1.93
2.10
2.29
2.49
2.71
2.94
3.20
3.47
3.76
4.08
4.42
4.78
4.92
5.27
5.64
6.03
6.45
6.90
7.36
7.86
8.39
8.94
9.53
10.15
10.80
11.49
12.22
12.99
13.80
14.65
15.55
16.50
17.49
18.53
19.63
20.79
22.00
23.28
24.61
26.01
27.49
29.03
30.64
32.34
34.11
0.3
0.86
0.94
1.03
1.13
1.24
1.35
1.48
1.61
1.76
1.92
2.09
2.27
2.47
2.69
2.92
3.17
3.44
3.73
4.04
4.38
4.74
4.95
5.30
5.68
6.07
6.50
6.94
7.41
7.91
8.44
9.00
9.59
10.21
10.87
11.57
12.30
13.07
13.88
14.74
15.64
16.59
17.59
18.64
19.75
20.91
22.13
23.41
24.75
26.16
27.64
29.19
30.81
32.51
34.29
0.4
0.85
0.94
1.02
1.12
1.23
1.34
1.46
1.60
1.74
1.90
2.07
2.25
2.45
2.66
2.89
3.14
3.41
3.70
4.01
4.35
4.71
4.98
5.34
5.72
6.12
6.54
6.99
7.46
7.96
8.50
9.06
9.65
10.28
10.94
11.64
12.37
13.15
13.97
14.83
15.74
16.69
17.69
18.75
19.86
21.03
22.25
23.54
24.89
26.30
27.79
29.34
30.98
32.69
34.48
0.5
0.85
0.93
1.02
1.11
1.22
1.33
1.45
1.58
1.73
1.88
2.05
2.23
2.43
2.64
2.87
3.12
3.38
3.67
3.98
4.31
4.67
5.02
5.38
5.75
6.16
6.58
7.03
7.51
8.02
8.55
9.11
9.71
10.34
11.01
11.71
12.45
13.23
14.05
14.92
15.83
16.79
17.80
18.86
19.97
21.15
22.38
23.67
25.03
26.45
27.94
29.50
31.14
32.86
34.66
131
0.6
0.84
0.92
1.01
1.10
1.20
1.32
1.44
1.57
1.71
1.87
2.03
2.21
2.41
2.62
2.85
3.09
3.36
3.64
3.95
4.28
4.63
5.05
5.41
5.79
6.20
6.63
7.08
7.56
8.07
8.61
9.17
9.77
10.41
11.08
11.78
12.53
13.31
14.14
15.01
15.92
16.89
17.90
18.97
20.09
21.27
22.50
23.80
25.16
26.59
28.09
29.66
31.31
33.04
34.84
0.7
0.83
0.91
1.00
1.09
1.19
1.31
1.43
1.56
1.70
1.85
2.02
2.20
2.39
2.60
2.82
3.07
3.33
3.61
3.92
4.24
4.60
5.09
5.45
5.83
6.24
6.67
7.13
7.61
8.12
8.66
9.23
9.83
10.47
11.14
11.85
12.60
13.39
14.22
15.10
16.02
16.99
18.01
19.08
20.20
21.39
22.63
23.94
25.30
26.74
28.25
29.83
31.48
33.21
35.03
0.8
0.82
0.90
0.99
1.08
1.18
1.29
1.41
1.54
1.68
1.84
2.00
2.18
2.37
2.58
2.80
3.04
3.30
3.58
3.88
4.21
4.56
5.12
5.49
5.87
6.28
6.71
7.17
7.66
8.17
8.72
9.29
9.90
10.54
11.21
11.93
12.68
13.47
14.31
15.19
16.11
17.09
18.11
19.19
20.32
21.51
22.76
24.07
25.45
26.89
28.40
29.99
31.65
33.39
35.21
0.9
0.82
0.89
0.98
1.07
1.17
1.28
1.40
1.53
1.67
1.82
1.98
2.16
2.35
2.55
2.78
3.02
3.27
3.55
3.85
4.18
4.52
5.16
5.52
5.91
6.32
6.76
7.22
7.71
8.23
8.77
9.35
9.96
10.60
11.28
12.00
12.76
13.55
14.39
15.28
16.21
17.19
18.22
19.30
20.44
21.63
22.89
24.20
25.59
27.04
28.56
30.15
31.82
33.57
35.40

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Table A:5 Climate examples. Monthly averages for two locations.

Stock-
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Temp C
-4.1
-4.1
-1.4
4.1
10.1
14.6
17.2
15.8
11.3
6.3
1.9
-1.0
RH %
84
82
74
66
62
65
69
74
81
83
86
86
I hor
10
30
77
135
188
210
200
160
99
47
15
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
-3.1
-0.7
-1.4
3.8
6.8
9.6
11.2
10.3
8.2
4.1
0.8
0.6
RH %
81
86
87
79
81
75
83
77
84
85
89
86
I hor
25
65
130
190
210
200
150
80
35
holm
W/m2
Reykjavik
Temp
C
W/m2
Table A:6 Sorption curves for some materials
Tegel,densitet 1680 kg/m
w kg/m3
20
10
0
0
50
RF %
132
100

2005-10-24
-------------------------------------------------------------------------------------------------------------------
w kg/m3
Lttbetong,densitet 500 kg/m3
200
100
50
100
RF %
w kg/m3
Betong K25
100
50
0
0
50
RF %
133
100

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Gran,densitet
3
420 kg/m 3
w kg/m
200
100
0
0
50
RF %
134
100

2005-10-24
-------------------------------------------------------------------------------------------------------------------
16 Examples for Moisture Transfer

16.1 Air humidity
Expression 12.3 gives us a relation between the temperature of the air and
the vapour content at saturation.
288.68 .
v sat ( )
1.098
8.02
4.689 .
.( > 0 )
100
273.15 ) .461.51
1.486
12.3
100
273.15 ) .461.51
.(
0)
0.03
0.025
v sat( )
0.02
0.015
0.01
0.005
0
30
20
10
10
20
30
We can study how the vapor content of the indoor air at saturation in kg/m3
varies with the air temperature in oC.
We see that air at 30 C contains less than 1 g per m3 at saturation while

air at 30 C can contain 30 g/m3.
16.2 Taking a shower in Trondheim

In a hotel room in Trondheim, where visitors to NTNU are often placed, the
ventilation in the bathroom is rather poor. During a hot shower the
temperature in the bathroom quickly rises to approx. 27 C and the air
becomes saturated. The temperature in the hotel room itself, both in the air
and on the surfaces is at 18 C. The initial relative humidity in the room is
40 % and the room is ventilated with air with the same values as the room .
The volume of the room itself is 50 m3. The room itself was ventilated with
0.5 air exchanges per hour and the air exchange between the room and the
bathroom through a door that stood ajar was corresponding to 1 air
135

2005-10-24
-------------------------------------------------------------------------------------------------------------------
exchange per hour. The moisture production in the hotel room itself can be
neglected. Now what happened?
If we look at equation 12.6 and rewrite it for ventilation towards two
environments we see that
Vroom dvi = dt (G 0.5 Vroom (vi ve ) 1.0 Vroom (vi vbath ) )

Rewriting the equation and taking into account that G=0
vi +
2 dvi 1
= (ve + 2 vbath )
3 dt 3
From Table A:4 we see that the saturation point at 18 C is 15.37 g/m3 and
40 % of that is 6.148 g/m3. This is our initial vapor content in the room and
also the vapor content of the ventilation air. The vapor content of the air in
the bathroom is constant equal to 25.73 g/m3.
This equation obviously has the solution
vi( t) :=
1
2
( 6.48 + 2 25.73 ) + ( 6.48 25.73 ) e
3
3
3
t
2
20
vi( t)
15.37
15
10
What happens is that the relative humidity of the room will relatively
quickly close in on the saturation point of the original room temperature.
Even if the temperature of the room air will rise some, the surfaces will not
follow so fast including the sensor for the fire alarm so now you can guess
the happy ending.
136

2005-10-24
-------------------------------------------------------------------------------------------------------------------
16.3 Moisture distribution in a multilayer construction

The purpose of this example
is to demonstrate how the
moisture distribution and
condensation can be
calculated for an arbitrary
multi-layer construction. The
shown calculation is shown
for 10 layers. The shown case
is from a paper mill with 40 C
and 80 % RH on the inside
and 0 C and 90 % RH on the
inside. The construction is a
sandwich of 80 mm concrete
100 mm mineral wool and 80
mm concrete. The concrete
layers are divided into 3
sublayers and the mineral
wool layer into 4 sub layers.
0.03
0.028
0.026
0.024
0.022
Vapor content kg/m3
0.02
0.018
0.016
0.014
0.012
0.01
0.008
0.006
0.004
0.002
0
20
10
0
10
20
Air temperature deg C
30
40
Climate data
Ti := 40
RHi := 80
Te := 0
RHe := 90
Boundary conditions
Rse := 0.04
Rsi := 0.13
Material data
j := 1 .. 3
dj :=
j := 1.7
0.08
3
j := 4 .. 7
0.10
4
0.08
dj :=
3
dj :=
j := 8 .. 10
j := 1.2 10
j := 0.036
j := 20 10
j := 1.7
j := 1.2 10
d0 := 0
d11 := 0
Boundary conditions
T12 := Ti
R11 := Rsi
T0 := Te
Z0 := 60
R0 := Rse
Z11 := 360
Approximate function for m

Indoor and outdoor moisture content
v12 := vsat ( Ti)
RHi
100
v0 := vsat ( Te)
RHe
100
Accumulated values vectors for thermal resistance, vapor resistance

and distance, starting values
Racc 0 := R0
Zacc 0 := Z0
dacc 0 := 0
137

2005-10-24
-------------------------------------------------------------------------------------------------------------------
Thermal and vapor resistances for the material layers

s := 1 .. 10
Rs :=
ds
s
Zs :=
ds
s
Calculation of temperature and vapor content distribution

k := 1 .. 11
Racc k := Racc k 1 + R k
Accumulated thermal resistance

dacc k := dacc k1 + dk1
dacc 12 := dacc 11
Accumulated distance
Zacck := Zacc k1 + Zk
Accumulated vapor resistance

Temperature at point k
Tk := T0 +
(T12 T0) Racc k1

Racc 11
vsatv 0 := vsat ( T0)
vsatv 12 := vsat ( T12)

Saturation vapor content in
point k
vsatv k := vsat ( Tk)
Vapor content in point k, This solution is only correct if no saturation occurs

(v12 v0) Zacck1
v k := v 0 +
Zacc 11
vk := min( ( vk vsatv k ) )
In case we have saturation we have to calculate the resulting vapor

distribution by iteration
v1 k , 0 := vk
Using the calculation above as starting values

r := 1 .. 200
200 iterations
w := 1 .. 11
For all element boundaries

v1r1 , 0 := v0
v1 r1 , 12 := v12
Boundary conditions are constant. For each element boundary we calculate

a new balance vapor content
assuming that the adjacent boundary contents are known
v1 r1 , w1 Zw + v1 r1 , w+1 Zw1
vsatv w
Zw + Z w 1
v1 r , w := min
We now calculate the accumulated moisture flow at each border

expressed in kg per m2 and month
v1 199 , w+1 v1 199 , w v1 199 , w v1 199 , w1
30 ( 24 3600 )
Zw
Zw1
g w :=
g0 := 0
g12 := 0
138

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The results can now be given in a diagram form

m := 0 .. 12
We can now calculate the

moisture accumulation in
kilogram per month at each
element boundary. At the
boundary between mineral
wool and concrete on the
outside we get condensation
equal to 1.245 kilograms per
m2 and month. The quantities
indicated at other boundaries
are extremely small and are
merely a result of the
inaccuracy of the iteration
process.
55
50
45
40
35
Tm
v1199 , m 1000
30
vsatvm 1000
25
gm 10
20
15
10
5
0
5
0.05
0.1
0.15
0.2
0.25
0.3
daccm
16.4 Capillary action
16.4.1 Brick in contact with water
d := 0.12
10
mdi
3600
m := 0.5 10
Time for filling up

2
m d
3600
t = 2 hours
t :=
0.1
0.2
di
139

2005-10-24
-------------------------------------------------------------------------------------------------------------------
16.4.2 Aerated concrete in contact with water
d := 0.12 m := 10 10
Time for filling up
m d
t :=
3600
t = 40 hours
200
150
2
mdi 100
3600
50
0.05
0.1
0.15
0.2
0.25
di
16.4.3 Concrete in contact with water
m := 30 10
d := 2.1
B := 0.02
Time for filling up
mdi
3600 24 365
m d
t :=
3600
t = 3.675 10
Years
3
2
1
0
0.5
1
di
140
1.5

2005-10-24
-------------------------------------------------------------------------------------------------------------------
16.4.4 Equivalent radius of materials
:= 0 := 1000 := 0.074
:= 1.31 10
g := 9.8
For brick
m := 0.5 10
requ :=
requ = 7.081 10
H :=
m cos ( )
requ :=
requ = 3.541 10
H :=
m cos ( )
H = 4.265 10
2 ( cos ( ) )
requ :=
requ = 1.18 10
H :=
m cos ( )
H = 1.28 10
requ g
For concrete
m := 30 10
requ g
H = 213.273
For aerated concrete
m := 10 10
2 ( cos ( ) )
2 ( cos ( ) )
requ g
16.5 Ventilated air gaps
v
u
x+dx
x
x=0
x+dx
In this example we want to analyze the ventilated air gap. The

preconditions given are that the outside temperature is 0oC, the solar
radiation on the roof surface is 300 W/m2, the absorptance of the surface is
141

2005-10-24
-------------------------------------------------------------------------------------------------------------------
0.8 , the thermal resistance on the outside of the air gap is 0.2 m2K/W and
on the inside is 5 m2K/W. The width of the air gap is 50 mm and the length
of the roof is 30 m. We assume values for convective and radiative heat
transfer in the gap.
Te := 0 a := 0.8 Rse := 0.04 R2x := 0.2 u := 1.0 R1x := 5 L := 30

Ti := 20 I := 300 := 1.2 c := 1000 T0 := Te hc := 3 hr := 5
b := 0.05
The equivalent outdoor temperature including the effect of the solar
radiation can be calculated as
Tequ := Te + Rse a I
From equations for the ventilated air gap we no calculate the necessary
intermediate quantities to be able to calculate the temperature distribution.
1
hc + 2 hr
hr
1
hc + 2 hr
hr
1
Rx :=
R :=
+
2
2
1
1
hc + 2 hc hr
hc + 2 hc hr
+
R1 R2
R1
1
k :=
Tinf := Ti ( Ti Tequ )
R1 + R2
c b u R
R1 := R1x +
R2 := R2x +
The temperatures for the air and the surfaces along the air gap can now be
calculated
i
k xi
Txi := Tinf ( Tinf T0 ) e
100
1 Tx + 1 Ti + 1 Tequ
Rx i R1
R1x
R2
Ts1 i :=
Ti
+ Ti
1
R1
1
1
+
+
Rx R1 R2
1 Tx + 1 Ti + 1 Tequ
Rx i R1 R2
R2x
Ts2 i :=
Tequ
+ Tequ
R2
1
1
1
+
+
Rx R1 R2
i := 0 .. 100
xi := L
The temperatures are plotted in the following diagram. The equivalent outdoor
temperature is almost 10oC even if the outdoor air temperature, which also is
the initial temperature in the air gap, is no more than 0oC. The surface
temperatures are higher than the air temperature in the gap.
142

2005-10-24
------------------------------------------------------------------------------------------------------------------10
Txi
Ts1i
Ts2i
Tequ
10
15
20
25
30
xi
We now want to
study the moisture balance in the air gap assuming that both surfaces are wet
assuming that we are in the beginning of a drying out phase. We start with
defining the saturation curve for air
8.02
vsat ( T ) :=
1.098 + T
100
288.68
( T + 273.15) 461.51
( T > 0) ...
12.3
1.486 + T 4.689
100
( T + 273.15) 461.51
( T 0)
The coefficient of surface vapor transfer is calculated from Lewis law and
the initial state of the inlet air and the surfaces is decided
:=
hc
1200
v0 := vsat ( Te ) 0.9
vsi :=
vsat ( Ts1 i) + vsat ( Ts2 i)

2
Since the surface vapor content is not constant along the surface we make a
stepwise calculation assuming constant properties for each segment.
j := 1 .. 100
( ( xjxj1) )
ub
v j := v j1 + ( vs j1 v j1) 1 e
vxsatj := vsat ( Tx j)
143

2005-10-24
-------------------------------------------------------------------------------------------------------------------
The air becomes

quickly saturated
but because the
air is heated
along the air gap
the vapor content
is increased
along the air gap.
0.009
0.008
vj
vsj
0.007
vxsatj
Now we calculate
the vapor flow
0.006
from the surfaces
to the air along
the air gap and
0.005
by multiplying by
3600 we get the
0
5
10
15
20
25
30
density of vapor
xj
flow in g/m2h.
The drying out is 4 times faster at the beginning of the air gap compared to
the end
gj := 2 ( vs j v j) 3600
g1 = 0.038
0.04
0.03
gj
0.02
0.01
0
10
15
xj
144
20
25
30

2005-10-24
-------------------------------------------------------------------------------------------------------------------
145

Lectures On Building Physics

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Lectures On Building Physics

Enviado por

Direitos autorais:

Formatos disponíveis

Prof Gudni Jhannesson

Div. of Building Technology

Lectures on Building Physics

Hotel Hamar. Borgarnes Iceland.

Heat and Moisture Transfer

Prof. Gudni Jhannesson - Lectures on Building Physics

Prof. Gudni Jhannesson - Lectures on Building Physics

Prof. Gudni Jhannesson - Lectures on Building Physics

Prof. Gudni Jhannesson - Lectures on Building Physics

Prof. Gudni Jhannesson - Lectures on Building Physics

BUILDING PHYSICS IN CIVIL ENGINEERING

What building physics is mainly about

Examples of heat, air and moisture transfer

New mathematical tools make life easier

Building physics and the environment

INTRODUCTION TO BUILDING HEAT TRANSFER

The basic elements of building heat transfer

Steady state calculations

Two-dimensional heat conduction

HEAT CONDUCTION EQUATION ANALYTICAL SOLUTIONS

Solution for a one dimensional slab with harmonic boundary temperatures

Steady state heat flow - simplifications

The control element - two dimensional heat flow

Expression of boundary conditions

Resulting heat balance

Choice of time steps and element size - stability

Prof. Gudni Jhannesson - Lectures on Building Physics

Simplifications for special cases

Finite element methods

CONVECTIVE HEAT TRANSFER

Basics of convective heat transfer

Expressions for different flow types

Heat balance in a duct with air flow

Expressions for surface heat transfer coefficients

Air flow through building components

RADIATIVE HEAT TRANSFER

Long wave radiation exchange

Calculation of view factors

The Radiosity matrix

Combined radiation and convective heat transfer

Solar radiation in building design

The elements of solar radiation

Heat balance on exterior surfaces

HEAT BALANCE PROBLEMS

General heat balance problems

Heat balance for a building

Window heat balance

The heat balance for a ventilated air gap

EXAMPLES - HEAT TRANSFER APPLICATIONS

MOISTURE IN BUILDINGS - INTRODUCTION

Prof. Gudni Jhannesson - Lectures on Building Physics

What do we mean by moisture?

Exterior timber framed wall

Moisture in building materials

MOISTURE TRANSFER IN CONSTRUCTIONS - DIFFUSION

Moisture transfer - introduction

The physics of moisture flow in porous materials

Diffusion in porous materials

Moisture surface resistances

Distribution of relative humidity

Water in porous materials

Capillary moisture transfer within materials

Water flow through porous materials