!""#"
!$%&'%"&("
  )*+**))*,-"."(#



  

  

   
 
 
 
   ! "#$$%
&'"( 

 (
 ()*+ 
 , 
 - *


 .*#$$%*

 ,
/ + */
(+  /
0

 /0 
 + /
*1
  

 *

  
   
 

23"*$**,6%1%456%6""364"&63*

7   

 

 (    

(( 
  

  
*8 



 
 

     , 
  


  
 ,
  9
 
 

   
 (
 


 ( 
 *
+   , (   ,  
     
 

  


  
   
   (   
    ,  
 
    

:     
*;


  (( 


 
 

*+

   (6 <
  (( 

(

 *
8   
   
  ( 
   
     ( 
    
   
  9       (      

 
 
 
   *;
 
 
    

*+ 
 

  , 

, 9,  =

 *+ 

,
6:
 (  6(  
 (  
  
 
*.

  
(
  
 


   >


,
   
6
,
(
 9



  , ,

 
( 
 ,  

 
 
*+
(  

  (
 
 

  
 

   
*
? 

,

(( 
  
 , 

  
 
 
  

 
 
, 
   9


 *1,  
  


 ,

 
 

 
*+
 



 

 
 *

  

    
 
 
 9

 

   

 

   

,
  (6

 

! 
 "#  $#%&'()#
#
$*'+,+
#

@.  #$$%

8//A2"62#3
8/;A%456%6""364"&63
 '  '''6$&#"B

'CC *9*C D E '  '''6$&#"F

 Till Pr...

... och resten av min familj

This thesis is based on work conducted within the
interdisciplinary graduate school Energy Systems. The
national Energy Systems Programme aims at creating
competence in solving complex energy problems by
combining technical and social sciences. The research
programme analyses processes for the conversion,
transmission and utilisation of energy, combined together in
order to fulfil specific needs.

The research groups that participate in the Energy Systems Programme are the Department of
Engineering Sciences at Uppsala University, the Division of Energy Systems at Linkping Institute
of Technology, the Department of Technology and Social Change at Linkping University, the
Division of Heat and Power Technology at Chalmers University of Technology in Gteborg as well
as the Division of Energy Processes at the Royal Institute of Technology in Stockholm.

www.liu.se/energi
List of Papers

This thesis is based on the following papers, which are referred to in the text
by their Roman numerals.

I Lundh, M., Blom, T., Wckelgrd, E. (2008) Antireection treat-

ment of Thickness Sensitive Spectrally Selective (TSSS) paints
for thermal solar absorbers. Submitted to Solar Energy.
II Lundh, M., Wckelgrd, E. (2007) Evaluation of the solar heating
model in a building simulation tool. In manuscript.
III Widn, J., Lundh, M., Vassileva, I., Dahlquist, E., Ellegrd, K.,
Wckelgrd, E. (2009) Constructing load proles for household
electricity and hot water from time-use dataModelling approach
and validation. Energy and Buildings. In press.
IV Lundh, M., Wckelgrd, E. (2009) Inuence of different load
proles on the performance of a solar hot water system. In
manuscript.
V Lundh, M., Henning, A. (2009) The Formation of a
Socio-Technical System: the example of Swedish solar and
wood-pellet heating. Submitted to Energy Policy.
VI Lundh, M., Zass, K., Wilhelms, C., Vajen, K., Jordan, U. (2009)
Inuence of store dimensions and auxiliary volume conguration
on the performance of medium-sized solar combisystems. Sub-
mitted to Solar Energy.
VII Lundh, M., Dalenbck, J.-O. (2008) Swedish solar heated
residential area with seasonal storage in rock: Initial evaluation.
Renewable Energy 33:703-711.
VIII Lundh, M., Wckelgrd, E. (2007) Theoretical study on solar and
pellet heating systems to reduce electricity demand for residen-
tial heating. Proceedings of CIB World Building Congress, Cape
Town, South Africa, 14-17 May 2007.

Reprints were made with permission from the publishers.

Other Publications

IX Jonsson, A., Lundh, M., Lfstrm, E. (2005) Hushll med

solvrme ett svenskt pilotprojekt i Anneberg. Arbetsnotat
Nr 30, ISSN 1403-8307, Program Energisystem. In Swedish.
X Lundh, M., Blom, T., Wckelgrd, E. (2006). Anti-reection
treatment of TSSS paints for thermal solar absorbers.
Proceedings of Eurosun 2006, 27-30 June 2006, Glasgow,
Scotland.
XI Lundh, M. (2007) Ssongslagrad storskalig solvrme. In: Wck-
elgrd, E. and Ellegrd, K. (Eds.), Energianvndning i bebyggelse
boendes och aktrers val av teknik, ISBN: 978-91-506-1957-7,
Uppsala, Sweden. In Swedish.
XII Lundh, M., Lfstrm, E. (2007) Large scale pilot project with
solar heating for a residential area success and disaster. Pro-
ceedings of ECEEE Summer Study, 4-9 June 2007, Cte dAzur,
France.
XIII Lundh, M., Wckelgrd, E., Ellegrd, K. (2007) Design of hot
water user proles for Swedish conditions. Proceedings of ISES
Solar World Congress 2007, pp. 2074-2078, 18-21 September
2007, Beijing, China.
XIV Lundh, M., Wckelgrd, E., Ellegrd, K. (2008) Design of hot
water user proles for Swedish households based on time diaries.
Proceedings of the Scientic Conference on Energy saving and
Green energy at Alvsjo fair, Stockholm, March 12-13 2008 in con-
nection with the Energitinget 2008.
XV Lundh, M., Wckelgrd, E., Henning, A. (2008) Qualitative
investigation of the development of combined solar and pellet
heating systems in Sweden. Proceedings of Eurosun 2008,
7-10 October 2008, Lisbon, Portugal.
XVI Lundh, M., Vassileva, I., Dahlquist, E., Wckelgrd, E. (2008)
Comparison between hot water measurements and modelled
proles for Swedish households. Proceedings of Eurosun 2008,
7-10 October 2008, Lisbon, Portugal.
XVII Lundh, M., Wckelgrd, E. (2008) Solar heating more than
a technical system. Proceedings of World Sustainable Building
Conference 2008, September 2008, Melbourne, Australia.
Contents

1 Market actors and technology an introduction to solar heating . . . 11

1.1 Why study solar heating? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2 Overall objective and research questions . . . . . . . . . . . . . . . . . 13
1.3 Solar heating systems in a social context . . . . . . . . . . . . . . . . . 14
1.3.1 System boundaries and system levels . . . . . . . . . . . . . . . . 14
1.4 Research odyssey and paper overview . . . . . . . . . . . . . . . . . . . 15
1.5 Thesis summary outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.1 Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Systems approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 Socio-technical perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.4 Case study methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Modelling and dynamic system simulations . . . . . . . . . . . . . . . 24
2.6 Time-geographic methods and time-use data . . . . . . . . . . . . . . 25
2.7 Interviews . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3 Solar energy and how to make use of it . . . . . . . . . . . . . . . . . . . . . 29
3.1 Solar radiation and technical applications . . . . . . . . . . . . . . . . 29
3.2 Electromagnetic radiation and its interaction with materials . . . 31
3.3 Solar collectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3.1 Different collector types . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3.2 Flat plate collectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4 Solar absorbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.1 Spectrally selective materials . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.1.1 Intrinsic selective materials . . . . . . . . . . . . . . . . . . . . . . . 38
4.1.2 Tandem absorbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.1.3 Textured surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
4.1.4 Antireection treatment . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.2 Thickness Sensitive Spectrally Selective paints . . . . . . . . . . . . 39
4.2.1 Project objective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.2.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.2.3 TSSS paints in a larger perspective . . . . . . . . . . . . . . . . . 42
5 Solar heating systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5.1 System solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5.2 The solar heating system market . . . . . . . . . . . . . . . . . . . . . . . 45
5.3 Planning new solar heating plants . . . . . . . . . . . . . . . . . . . . . . 46
 5.3.1 Reliable planning tools . . . . . . . . . . . . . . . . . . . . . . . . . . 47
5.3.2 Hot water load model . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
5.3.3 User inuence on solar heating systems . . . . . . . . . . . . . . 54
5.4 Small-scale system solution: Solar and wood pellet . . . . . . . . . 56
5.4.1 Qualitative interviews . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
5.4.2 The market as experienced by the informants . . . . . . . . . . 58
5.5 Medium-sized mid-term system solution . . . . . . . . . . . . . . . . . 59
5.5.1 Inuence of store dimensions . . . . . . . . . . . . . . . . . . . . . . 60
5.6 Large-scale long-term system solution: borehole storage in
Anneberg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.6.1 Socio-technical evaluation of Anneberg . . . . . . . . . . . . . . 65
5.7 Heating system assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.7.1 Electricity production . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.7.2 Natural assets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.7.3 Assessment from a marginal electricity perspective . . . . . 70
6 Discussion and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.1 Project result overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
6.2 Overall conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
7 Future work and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
8 Final remarks and self-evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
APPENDIX A: Interview guide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
APPENDIX B: Word list and abbreviations . . . . . . . . . . . . . . . . . . . . . 87
9 Summary in Swedish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
1. Market actors and technology an
introduction to solar heating

When I ask family and friends how they perceive solar heating, most of them
respond by saying something about boxes on the roof, or that it better suits
a country with higher insolation, or maybe by questioning me; is the tech-
nology really mature? Most importantly, they all focus on technology. We
tend to focus on the technical components when discussing a heating system,
or any technological system, but is it that simple? In this thesis the perspective
is widened to include more aspects and values than the purely technical. For
example, a heating system will always appear in a social context; it is used by
people and inuenced by the surrounding society and current rules, legislation
and trends, but it also affects the people around it. There is a close interaction
between technology and people. This thesis presents a way of understanding
that interaction by investigating some of the aspects of solar heating technol-
ogy in a social context. Through that understanding, some of the obstacles pre-
venting wider implementation of solar heating may be identied and thereby
contribute to nding a solution to the slow implementation of this technology,
because what is more reasonable than using free energy that ows to us in
excess, every day?

1.1 Why study solar heating?

At present, more than one third of the total nal energy is used within the
residential and service sector in Sweden, and from this about 60 percent is
for space heating and hot water in residential and commercial buildings [1].
There are more than 1.7 million detached houses in Sweden [2] and almost
half of the energy for heating is electricity, used both in electrical heating sys-
tems and to operate heat pumps [1]. Moreover, about 80 percent of all single-
family houses built today use electrical water based central heating systems,
and another eight percent utilize electric resistance heating [3]. This means
that electricity is still the most common way to heat detached houses in Swe-
den. After the oil crisis in the 1970s the Swedish government even encour-
aged households to convert from oil boilers to electrical heating, for example

11
by offering conversion subsidies. In the present political discussion, the large
number of electrically heated single-family houses is rather seen as an obsta-
cle in the ambition towards a sustainable society [3], which has resulted in
new conversion subsidies [4] [5].

When considering actions to increase sustainability in the built environment

the energy demand can in the rst place be minimized to reduce the resource
use and the negative environmental effects. This may be achieved for example
by increasing the insulation or air-tightness of a building. Low-energy or pas-
sive houses, which only require a small amount of purchased energy to be
operated, are built more frequently [6]. Even old buildings are retrotted into
passive house standard. There will, however, always remain a need of energy
supply for appliances, space heating and hot water. Energy supply that origi-
nate from renewable sources may be used to meet the requirement of sustain-
ability. Solar energy is a vast renewable resource that is only used to a small
extent for heating and electricity production at present.

A frequent comment on solar heating is the so called lack of solar radiation

in the Nordic countries [7]. Correctly, a solar heating system can not cover
all heat demand on an annual basis, but that is more due to the mismatch be-
tween periods of high insolation and high heating demand, than actual lack
of solar radiation. A substantial part of the hot water demand can be supplied
directly by solar heating during summer, but solar energy can also contribute
with low-temperature space heating or pre-heating hot water during spring
and autumn. Heat storage issues are essential and an auxiliary heating system
is always required. Therefore, appropriate and well-functioning system solu-
tions, with heat sources adjusted to combinations with solar collectors, have to
be developed and commercialised. Similarly, heat stores, enabling storage of
energy on different time-scales, from hours to seasons, have to be developed
to overcome the mismatch in time.

Thermal solar collectors for hot water production have been markedly devel-
oped since the 1970s and the technique to utilize solar radiation for heating is
today well-developed and has reached cost-effectiveness in many applications.
But if the systems are both technically well-developed and cost-effective, why
is solar heating not more widely used? Considering the simple and safe tech-
nology, utilizing an energy source free of charge, solar heating should already
be a success. As is shown in this thesis, the reasons are many and of differ-
ent nature and in order to grasp the complexity of the issue, a socio-technical
framework has been applied.

12
1.2 Overall objective and research questions

The main objective of this thesis work is to widen the system boundaries and
method basis in solar heating research; from purely technological studies of
materials and components to studies of solar heating technology and technical
systems in a social context. The implementation rate of solar heating in Swe-
den is modest. My hypothesis is that both possibilities and obstacles appear
in the interaction between technology and people, and not only in technology
itself. That means that obstacles toward implementation can often be related
to people and their knowledge; the users expectations on the technology and
the planners expectations on the user. Those do not always match. To under-
stand why they do not match the connections between technology and market
actors have to be investigated.

There are several question marks to be straightened out in order to convince

a customer to invest in solar heating, for example the auxiliary heat source
and heat store issues mentioned in the previous section. In this work some
of the obstacles that can appear in the planning, installation and utilization
phases are identied using Sweden as a case study. To solve the problems, both
engineering methods and methods traditionally used within other disciplines
are applied, all under proper supervision. The questions upon which the thesis
work is based are the following:

How can technical studies and studies of market actors be brought together
to get a wider perspective on solar heating and improve the technical studies?
(Model of hot water demand in Swedish households in Paper III; Simulations
with hot water proles in Paper IV; Simultaneous technical and social evalu-
ation in Section 5.6; The trades experiences of the market in Paper V.)

How can solar heating utilization be optimized and in what way could solar
heating be part of a larger energy system?
(Improvement of absorber materials in Paper I; Evaluation of a solar heating
model in Paper II; Interviews with solar and pellet trades in Paper V; Medium-
sized mid-term storages in Paper VI; Evaluation of a large-scale pilot plant in
Paper VII; Assessment of different heating systems in Paper VIII)

The focus of the thesis is twofold; one part focuses on heat storage (Paper II,
III, IV, VI, VII) and one part on system solutions (Paper V, VIII). A smaller
part of the thesis is about absorber materials (Paper I).

13
1.3 Solar heating systems in a social context

Technically, a solar heating system consists of solar collectors, a storage tank,

a circulation pump, an auxiliary heat source, heat exchangers, valves, heat
distribution systems and so on. The solar collectors on the roof and the ac-
cumulator tank in the basement are also what the user mainly experiences.
But what does it mean to actually own and use a solar heating system? Above
all, the end-user demands that the system should supply heat and hot water,
be safe and easily operated. But to reach that state, the solar collector sys-
tem has to be installed properly, after being planned and dimensioned by the
installer, construction company or architect. Appropriate solar collector mod-
ules and storage tanks have to be chosen and ordered. They are delivered from
the manufacturer, via a retailer, to the installer or to the home address of the
customer. Only considering the solar collector, components like cover glass,
absorber, insulation and frame, have to be produced and combined in different
factories. To facilitate the decision and investment for the customer, the gov-
ernment may decide to support the customer by subsidies, while companies
offering heating systems try to persuade the customer by different types of
marketing. This means that there is a long chain or network of actors involved
before the solar heated hot water reaches the bathtub or the low temperature
oor heating system reaches the desired temperature.

1.3.1 System boundaries and system levels

In this study a systems approach is applied to understand the interaction be-

tween market actors and solar heating technology. A system is in this case
dened as components connected to form an integrated whole (see further in
Section 2.2). By this approach the technology can not be considered as indi-
vidual artefacts1 , but rather as elements in a complex system [8]. Furthermore,
some parts of this study take social factors 2 into account. Therefore a socio-
technical systems approach is in most cases applied, but technical studies are
also presented. The main advantage with a socio-technical approach is the pos-
sibility of using methods that are traditionally not used in technical research.
The focus in this thesis is still on technical systems, but by using time-use data
the user is introduced in the study of technical performance, while technology
is seen in a social context by carrying out interviews.

1 An artefact is a man-made object, in contrast to naturally existing objects.

2 Social factors may be market actors, such as user, manufacturer, installer, as well as political,
cultural or economical conditions.

14
The papers are dealing with different system levels and with varying system
boundaries to get a wide perspective on solar heating. A schematic gure of
those boundaries and levels are found in Figure 1.1. All papers, except Paper I,
take off at higher system levels than the detailed component technology itself,
from heating system issues in buildings up to market and society level. This
means that the perspective of this thesis is raised from studies in technology
to studies of technology in a social context, i.e. a more holistic picture of the
technology in focus.

Society and market

Paper VIII

Residential area
Paper VII
Paper II
Building Paper III
Paper IV
Paper V
Component:
Paper I Paper VI
Solar collector

Figure 1.1: The system boundaries and levels used within this thesis work. The posi-
tion of the appended papers are also found in the gure.

1.4 Research odyssey and paper overview

It all started with a study on cheap absorber materials, in form of spectrally

selective paints. The aim was to investigate the possibility of improving the
optical properties of the paints. In Paper I it is shown that an improvement
is possible, but it requires an additional manufacturing step and has a rather
small impact on the total amount of useful heat from a solar collector. Fur-
thermore, there are already highly efcient and fairly cheap absorbers on the
market. I therefore asked myself if the price and quality of absorber materi-
als are the main reasons for the modest use of solar heating. As part of the
interdisciplinary graduate school, The Energy Systems Programme, I had the
opportunity to move on and continue my journey in a slightly different di-
rection. I therefore decided to focus on system issues, such as how to store
solar heat on different time-scales, but also the auxiliary heat source required
in temperate climates.

The rst step toward installation of solar heating is planning. Simulation tools
are often used to show possibilities with different technical solutions. Up to
now detailed solar heating simulations have mainly been performed within re-

15
search groups. Due to a growing interest in and regulations about energy use
in buildings there is, however, an increasing number of commercial simulation
tools for planners available on the market. Correct estimations of possible so-
lar energy supply are prerequisites to make planners choose solar collectors.
One commercial and widely used Swedish building simulation program, en-
abling detailed simulations of both buildings and supply systems, was there-
fore evaluated. This is reported in Paper II.

Furthermore, to properly estimate the system requirements and to take the

dynamics in a solar heating system into account the heat load must be known.
To improve the description of a household, from the very simplied hot water
proles normally used in simulations, a model was developed to translate
time-use data into hot water use. This study shows one possibility of bring-
ing the user in to simulations of technological heating systems according to
research question one. The model is described and compared to measured
data in Paper III and applied in simulations of a solar hot water system in
low-energy buildings in Paper IV. This study also shows possibilities with
small-scale short-term storage of heat and how the user behaviour and habits
of individual households may be advantageous or disadvantageous in solar
heating utilization.

Another small-scale solution that is getting more frequently used, and which is
promoted as a renewable system, is combined solar and wood-pellet heating.
Although several solar and pellet companies make efforts to offer those sys-
tems, the dissemination is slow. The technology is well-developed and rather
cost-effective, and different market actors promote as well as prevent the im-
plementation process. In Paper V people in the trade give their view on ob-
stacles and possible solutions to reach increased implementation. This study
relates to both research questions by constituting a way to include the actors
in the study of solar technology as well as investigating how solar heating may
be part of the larger energy system.

Storage of heat is also essential for solar heating systems and a great deal
of research has been done in the area. It is, however, not trivial to store heat,
especially over long periods of time and for small systems. Different solutions
have to be available on all system levels, from small-scale to large-scale, and
from short-term to long-term storage. Additionally, they have to be adopted
to the different users appearing at the different system levels, from individual
households to residential areas or cities.

The auxiliary heater is the basic heating system in most existing solar heating
systems, while solar energy only contributes with a smaller fraction of the
demanded energy. Increased fractional energy savings are therefore important
to let solar heating constitute the foundation of the system. One possibility

16
is to increase the size of conventional water tanks, which would constitute
a medium-sized mid-term storage solution. The space available and physical
restrictions of detached houses, however, limit the size of the store. To answer
the question about optimization of solar heating for individual households the
inuence of those physical constraints are investigated in Paper VI.

By use of seasonal storage the fractional energy savings can be increased even
more, but long-term stores are only economically reasonable for larger de-
mands. One example is the pilot plant in the residential area Anneberg, out-
side Stockholm, utilizing a borehole storage. Well established technologies
are used, while the system solution is state of the art. Both research questions
in Section 1.2 form the basis of the investigation. First, there was a paral-
lel social and technical evaluation performed in an interdisciplinary research
project, discussed in Section 5.6. Secondly, the area comprise a new system
solution for solar heating, utilizing a seasonal borehole storage. The technical
evaluation is presented in detail in Paper VII.

To nd a proper system solution is not trivial. Several different aspects have

to be considered. Due to the political discussion on the need of reducing the
electricity demand for heating, the last system level to be examined within
this thesis was the larger energy system. Simulations of different heating sys-
tems were performed in a model of a contemporary single-family house in
Paper VIII. The results were used to discuss possible effects of converting the
heating systems of individual households, both on the national energy system
and international carbon dioxide emissions.

Not a ten, but ve year long expedition into the world of solar heating is
close to an end and I can almost see the contour of my Ithaca. It has been an
adventure out of the ordinary. I have visited many islands along the journey,
and all of them offered me new, exciting, but sometimes also scaring, chal-
lenges. To actually nish this stage of the journey it is time to stand up and
defend my Penelope, which you have in your hand. I will now tell you my
account of the journey. Welcome onboard.

1.5 Thesis summary outline

This part of the thesis both summarizes and complements the appended
papers. Several different methods have been applied within this thesis
work, all presented in Chapter 2. Solar energy and the properties that make
it possible to utilize solar radiation in thermal solar collectors are then
described in Chapter 3. In Chapter 4 the idea of solar absorbers is introduced

17
and a project on improvement of absorber paints presented. Most projects
performed within this thesis are described in Chapter 5, which focuses on
solar heating systems. Available system solutions and the present market is
briey described and after that the focus is on projects on planning solar
heating, system solutions of different size and heat stores on different
time-scales. The difculties and some suggestions for assessment of heating
systems are also discussed. Finally, the overall outcome of this thesis work
is summarized in Chapter 6. Important work not covered by this thesis is
proposed in Chapter 7 and at the end I evaluate what I have achieved during
those years.

18
2. Methodology

The present research in the eld of solar heating systems has a major focus
on complex dynamic simulations of different system solutions. The simula-
tions, however, most often only focus on the technical components and highly
simplies the social context in which the solar collector system is operated. To
introduce the social context in the simulations, but also to studies of solar heat-
ing technology in general, methods that traditionally are not used within en-
gineering have been applied within this thesis. To ensure the scientic quality
the methods have been used in collaboration and under supervision of a social
anthropologist and a cultural geographer respectively. The different methods
are presented in the following sections, jumping off in traditional engineer-
ing experimental methodology, on to the general method of system studies,
which pervades all research projects performed within this thesis work. There-
after case study, modelling and simulations, time-geographic studies as well
as interview methodology are introduced. The aim of this chapter is to explain
how the different methods complement and interact with each other in this
thesis.

2.1 Experiments

According to Jnsson and Reistad (1987) the experimental method is one

of the most important characteristics of natural science in the West. They
continue saying that to perform experiments from a natural science point of
view means to interfere with nature and register how nature respond to the
disturbance. [9] In contrast to natural processes, which can not easily be con-
trolled, the narrow system boundaries and study objects in experiments allow
systematic observations of a phenomenon. The scientist thereby rules over
the conditions and controls the process. Well-dened system boundaries min-
imize unwanted and uncontrolled disturbances during the experiment. Ideally,
only one parameter is varied at a time to isolate the effect of the particular
disturbance.

Already in the 16th century Galileo Galilei stressed that true science con-
nect mathematical theories and empirical observations, but also that physics

19
should describe the world as it is, i.e. objectively. And still, the empirical sci-
ences focus on exploring, describing, explaining and predicting phenomena in
nature. The present technological research and development to a large extent
take place in small-scale laboratory setups, where experiments are typically
used to connect a phenomenon to a theory or to map causalities.

One of the fundamental requirements is that the experiments should be re-

producible; the results should be independent on the particular occasion, re-
searcher or equipment. Repeated experiments, calculations or different equip-
ments may also be used to verify the results. Reproducibility, however, claims
both systematics, well-dened conditions and standard international proce-
dures to enable comparisons.

The benets of experiments are, however, not uncontroversial. By some it is

seen as an objective method to collect observations, while others stress the
ambiguity, due to the conclusions being dependent on the theory on which the
interpretation is based. In addition, there are always intentions with a partic-
ular experiment; it is based on certain questions, which foremost may affect
the analysis of the results. The scientist is often considered objective, not
inuencing the experiment, but may in reality both affect the results by the in-
tentions, interpretations and by the carefulness and systematics applied during
the experiment.

Narrow system boundaries are necessary to allow precise experimental results,

but they also constitute a limitation in the understanding of the large-scale
performance of the object. The small technical detail will eventually end up in
a larger technical system, interacting with all those factors that are considered
belonging to the surroundings during the experiment.

2.2 Systems approach

The concept of systems and system research have been discussed and dened
by many theorists over the last decades, among which several are presented,
along with their theories, in Ingelstam [10]. Thus, a system can be dened in
several different ways. In this thesis a system is considered comprising a set of
components interacting with each other through connections, as illustrated in
Figure 2.1, but also coordinated to reach a mutual goal, according to the theory
presented in Churchman [11]. A minimum of aspects have to be considered
and dened when performing a system analysis:

- the general objective and performance of the system as a whole

20
- the environment of the system; the xed restrictions
- the available resources
- the components constituting the system and their performance
- the management of the system.

The focus in system studies is the performance of the system as a whole,

which is most often different and more than the sum of the performance of the
components. Moreover, the methods or tools to be used to explore a system
are not given, but should be carefully selected with regard to the problem to be
solved, which thereby enables interdisciplinary use of methods and theories.

Environment

Connection

Component

Figure 2.1: A schematic gure of a system model, after Ingelstam [10].

The system boundaries do not need to be the physical boundaries or limits,

such as the walls of a building. What is not inuenced by the system is instead
considered belonging to the environment, while the environment still inu-
ences the performance of the system [11]. Furthermore, resources are found
within the system and are used by the components to perform the tasks. The
difference between resources and the environment is that the resources can
be controlled, while the environment is given. Finally, the management of the
system formulates the aim of the components, distributes the resources and
controls the system performance.

Churchman stresses the importance of studying the system as objectively as

possible, but also highlights the subjective inuence of the system analyst,
who also in some way constitutes a part of the system being observed. Further,
he emphasizes the weight of looking at the system in a wider perspective and
to include values that are difcult to understand and quantify. Consequently,
by this system view the social context, which is traditionally regarded as be-
longing to the environment and not being inuenced by the technological sys-
tem, may now be considered part of the system and intimately interacting with
the technology. The system approach is therefore highly relevant in the case of
studying solar heating from a socio-technical perspective. Relating to the re-

21
search questions in this thesis, different actors may be introduced in the wider
solar heating system by adding them as system components. Additionally, the
systems approach allows solar heating to be seen as part of the overall energy
system, which has to be considered when evaluating the use of solar heating.

Just like Churchman introduces his book The Systems Approach, questioning
why most of the fundamental problems in the world have still not been solved,
I introduced my thesis by asking why solar heating is not more widely used.
Churchman continues arguing that the main reason is that the problems are
very complex and can not really be separated. In the same way I argue that
solar heating technology can not be isolated to solve the problem of imple-
mentation, but a systems approach has to be applied.

2.3 Socio-technical perspective

The contextual description of a solar heating system in Section 1.3, includ-

ing both artefacts and market actors, is a typical socio-technical system in the
sense used in this thesis. This means that the socio-technical perspective is
used to relate to the problems within system studies; a way to widen the per-
spective from the purely technical to also include the social context in the
study of technology. The socio-technical system perspective, however, origi-
nates from a movement towards multi-disciplinary research within the social
sciences in the 1980s and 1990s [8], but there is no unambiguous denition of
socio-technical systems.

It more or less started with the historian of technology Thomas P. Hughes,

who strongly inuenced the development of the theory on Large Technical
Systems (LTS) which, in the 1980s, was illustrated by the development of the
electrical system. Hughes focus was the mutual inuence between society
and technology, while similar previous studies exclusively focused on tech-
nologys inuence on the society [10, 12]. The theory was developed in con-
trast to the traditional technological determinism. Technological development
was no longer seen as a straight, predetermined path, but more as a winding
track of trial and error.

Another theory, partly following Hughes, is the Social Construction of Tech-

nology (SCOT). According to social construction, there is no unambiguous
best solution, but the result depends on the social groups that formulate the
problem and use the artefact. There are always negotiations and compromises
between actors within a social group before the artefact reaches the nal de-
sign. [13]

22
Yet another theory connected to the socio-technical perspective is that of Sys-
tems of Innovation [14]. The focus is the inuence of strategies and actions
taken by the engineers and contractors in the technological development. En-
gineers build networks to reach their goals; they negotiate and persuade other
actors and the required artefacts to form them into the network. Conicts and
power struggles are characterizing the technological development.

In a socio-technical system, artefacts and technology are not isolated, mean-

ingful objects, but rather intimately embedded in a network or system of both
social and technical factors. There is no clear boundary between technical and
social systems, but as Hughes expresses it: they together form a seamless web.
Everywhere along the path of development there are market actors strongly in-
uencing the technology by what they do, but also by what they refuse to do.
Politics, culture and the physical surrounding also inuence the artefact. But,
as was mentioned in the approach to this thesis, people tend only to see the
concrete and isolated artefact. However, by a socio-technical perspective, the
artefact is only valuable when seen in a social context; when it interacts with
people and is used by people. We can not only ask what, but also have to ask
who, to understand how a technology develops and disseminates. [15]

Although socio-technical systems surround us in everyday life the focus of

research is in most cases either on social or technical aspects. In this the-
sis, applying a socio-technical perspective means that both social and tech-
nical considerations are taken into account and brought together. The socio-
technical perspective is thereby seen and used as an interdisciplinary frame-
work to widen the system view of an artefact. In reality this means to introduce
market actors in the usually strictly technical world of solar heating systems,
but also to reect over peoples choices, political decisions and organisational
structures, and how it inuences the use or lack of use of solar heating.

2.4 Case study methodology

One method to systematically examine a phenomenon is case study. Just like

the systems approach, but in contrary to many other research methods, the
case study does not claim any specic methods to collect or analyse informa-
tion. The uniqueness of case studies is not the method used, but the questions
that constitute the foundation of the study. In Yin (2003), the case study is
described as suitable for how and when questions [16]. Furthermore, Merriam
(1994) writes that a limited system has to be identied to apply case study to
a research object. A case study is thereby examination of a specic occasion,
where the system is chosen due to its importance, interest or since it forms a

23
hypothesis. A case study does not try to predict the future or generalize, but
understand the present situation or context [17]. A holistic view on the object
should be applied.

Case studies are particularly suitable when the variables concerning the study
object can not be separated from the context or the surrounding situation; the
investigator has no or little control over the object and can not manipulate the
relevant variables. The main advantage of the methodology is its ability to
manage different types of empirical material. Furthermore, the case study can
be used to show the complexity of a situation. It may explain why a problem
has occurred and give the background to a certain situation, explain why a
change works or does not work, discuss and judge the chosen alternatives, as
well as evaluate and summarize a situation. [16] The result could give indi-
cations on what should and should not be done in similar situations. It only
applies to the particular object or occasion, but should show a general prob-
lem.

Close to all larger solar heating projects are evaluated, but very few evalua-
tions are categorized as case studies. One of the few is a study of the Sr solar
heating plant [18]. Another solar heating case study is found in Chapter 5.6. In
the latter, not only a technical evaluation was performed, but interviews were
also carried out to give an idea of how the technology was received by the
residents. Case study was thereby used to position technological studies in a
larger context.

2.5 Modelling and dynamic system simulations

As mentioned in the approach to this chapter system simulations, using for

example the dynamic simulation tool TRNSYS [19], constitute an important
method in the present solar heating research. The technical system under in-
vestigation is built up by physical relations between the components consid-
ered inuencing the performance of the system. Modelling and simulations
are thereby intimately correlated to system studies. One limitation is, how-
ever, the transparency of many commercial simulation programs, where the
user often has no or restricted insight in to the actual numerical algorithms1 .
This limits the understanding of the results, but may also affect the outcome,
since errors in the model are not easily discovered.

1 TRNSYS is one exception, since the program user can read, modify and share the model de-
scriptions by open source codes.

24
In this thesis dynamic system simulations are used to investigate different sys-
tem solutions, which applies to the research question on how solar heating
utilization may be optimized. But simulations are also used to compare solar
heating with alternative heating systems, which introduce solar energy in a
larger energy system.

Simulations are used in many different research areas as a cost-effective

means to examine the performance of a technical system that is not yet in
use. The mathematical model comprise a set of equations describing the
physical phenomenon by thermodynamic relations. Ingelstam expresses the
benet of models in its ability to answer questions about the system without
performing experiments [10]. Simulations can, thus, be considered numerical
experiments where the object is a mathematical model. A model should,
however, never be considered to fully represent the reality or deliver the
truth. The basic conditions of the model are always limited and governed
by the programmer. Only a limited number of properties can be taken into
account and components and relation descriptions are often simplied and
sometimes ignored. Furthermore, Churchman discusses the difculty of
including other values than the technical or economical, i.e. values that are
not easily quantied, such as feelings, behaviour or experiences [10] [11].

Different methods have previously been used to include social factors in math-
ematical models, such as the Predicted Percentage Dissatised (PPD) index.
In this thesis a time-geographic study (see further in the next section) is used
to model residential energy demand for hot water. This is one method to in-
troduce the user behaviour in mathematical modelling and achieve improved
descriptions of the dynamics between user and technology.

2.6 Time-geographic methods and time-use data

In a case study, system simulations can be seen in a context by use of several

different methods, so called triangulation [17]. Still, the social context, in form
of the people using the technology, has not been included in the simulations.
One way to actually introduce the household, or rather the behaviour of the
household members, is to use the outcome of studies on peoples arrangement
of their time, which indirectly tells how they use the technical systems that are
to be simulated.

Time-geography originates from the cultural geographer Torsten Hgerstrand

and his research group, who applied an interdisciplinary approach to geog-
raphy. By time-geography they connected the spatial and timely perspectives

25
in geography and history respectively and focused on the time-space. The ev-
eryday life of a person is the interconnected and related activities she or he
performs as time ows; every person has a lifeline in the time-space. This in
turn means that society is built on those everyday activities of its individual cit-
izens, resulting in a web of connected biographies [20]. In time-geography
uxes of matter, energy and information is therefore not only seen as move-
ments in space, but also as movements in time.

Ellegrd describes a method to empirically collect knowledge about the timely

distribution of everyday activities by so called time diaries, resulting in time-
use data [21]. The respondents write down all the activities they perform dur-
ing the day; the type of activity, when and where it takes place, accompanying
persons, means of transportation and so forth. To minimize the inuence from
the researcher, such as communicating a notion of important activities, the
diaries are preferably open, meaning that the informants dene the activities
and time-steps themselves. To visualize the time-use data a computer program
(Visual-TimePAcTS) has been developed at Linkping University [22, 23, 24].
After being imported to the program, the data can be used to plot time use for
different activities, both for individuals and groups of individuals.

Time diaries often focus on the effects of changes and can be used for a set of
different purposes (within therapy, research on everyday life [21] as well as on
changes caused by new technological systems [25]) and have previously been
used to visualize energy use in households, for example by Lfstrm [26].
A fairly new approach is to use time-use data to improve the description of
human behaviour in technical simulations. By translating the activities to en-
ergy use realistic individual load proles may be generated. And since they
are based on time-diaries they do not only supply an average energy use, but
are connected to an arrangement of everyday activities that makes sense to the
person performing them. By letting those load proles describe the user of the
simulated system a more realistic interaction between technology and user is
achieved.

2.7 Interviews

The theoretical performance, as well as advantages and drawbacks of techni-

cal systems, may be studied by simulations, but not even an extremely detailed
model will tell why a particular system solution, such as combined solar and
wood-pellet systems, has not yet reached wide implementation. And there is
no unambiguous answer. But there are people working for increased dissem-
ination of those systems and they have experiences, perceptions and ideas

26
about the market situation. Their experiences could increase the understand-
ing of the social context and the market arena, on which combined solar and
pellet systems have to ght against other heating solutions. By qualitative in-
terviews the informants get the possibility of expressing their thoughts about
the situation, from inside the market. By interviews, it is possible to position
technology in a larger social context.

The interview is a professional conversation aiming at the informants experi-

ences and opinions. It is particularly well suited to investigate such issues that
can not be directly observed, such as the past [17]. It is a means to understand
how the interviewed person perceives a specic topic or phenomenon of com-
mon interest to the interviewer and interviewee [27]. There are similarities
with everyday discussions, but also differences. The everyday conversation
mainly takes place between two equal parts and focuses on the subject, while
the research interview has specied objectives and structures, but it is also
a one-sided questioning; the interviewer does not argue for her or his opin-
ion [27]. According to Kvale knowledge is brought forth in the interaction
between the interviewers and the interviewees opinions. This means that
the end-product depends both on the interviewer and the informant and their
interpersonal interplay, but the interviewer is the one dening the situation. In
absorber material studies different optical measurement equipment constitute
the instruments, while in interview studies the interviewer herself is the tool.

The principal advantage of qualitative semi-structured interviews is their

openness. There are no strict standard techniques or rules for how to perform
it [27]. Instead it is exible and the approach may be modied during
the interview, depending on the situation and the informant in question.
Follow-ups can be made on new interesting topics or issues, but the interview
guide, dening possible questions or themes, may also be strictly followed.

All the above methods may be used individually, but in this thesis they were
rather used complementary. As was described above, one method may ll the
gap where another can not supply a full explanation of a phenomenon. The
usefulness and advantages of triangulation, i.e. when several methods are used
together or in parallel to reach further, is stressed. How the different methods
were used in practise is described in the respective project description.

27
3. Solar energy and how to make use
of it

Solar energy is a natural part of everyday life and may be used both actively
and passively and both for lighting, electricity generation and heating. It pas-
sively heats buildings when passing through windows or being absorbed by
building walls. Solar radiation may also be used actively to heat water in solar
collectors. In this chapter properties of the solar radiation, as well as materi-
als for solar energy utilization, are described and thermal solar collectors are
briey introduced.

3.1 Solar radiation and technical applications

The sun is the basic condition for life on Earth, both for plants, animals and
human beings. It has an estimated age of 4 600 million years and has always
been an intensively used source of heat and light for people, both actively
and passively. The solar radiation reaching Earth contains about 10 000 times
more energy than the present total energy use. One difculty so far has been
to convert it to useful energy, but also to store it over time.

The nuclear reactions in the core of the sun generates a lot of heat, and thereby
a surface temperature of about 5 500 C. All objects with a temperature above
the absolute zero (0 K = -273 C) emit so called blackbody radiation in a cer-
tain wavelength interval depending on the temperature. The radiation inten-
sity, I , depends on the temperature, T , of the object as well as the wavelength,
, of the radiation and is described by Plancks law

2hc2
I( , T ) =  hc  (3.1)
e kB T 1 5

where h is Plancks constant, c is the speed of light and kB is Boltzmanns

constant. The wavelength, at which the maximum intensity occurs, moves to-

29
wards shorter wavelengths as the temperature increases, according to Wiens
displacement law

b
max = (3.2)
T

where b is a constant. The solid line in Figure 3.1 shows the spectral distri-
bution of the solar radiation reaching the surface of the Earth, while the black-
body radiation from objects at 100, 200 and 300 C are found as dashed lines.
The high temperature of the sun has moved the peak intensity towards short,
visible wavelengths. On its way to the Earth, the solar radiation is ltered,
above all when passing the atmosphere. Some of the radiation is reected,
some is absorbed in molecules, and another portion is scattered, resulting in
the radiation distribution in the gure. The solar radiation reaching Earth is
mainly in the wavelength range 0.3 to 3.0 m. Within this interval, wave-
lengths shorter than 0.4 m correspond to ultraviolet radiation, about 0.4 to
0.8 m is visible light and wavelengths longer than 0.8 m are infrared radi-
ation or heat. Furthermore, a temperature of 100 C corresponds well to the
working temperature of an ordinary solar collector. Interesting to note is that
most of the solar radiation has shorter wavelengths than 3 m, while most of
the blackbody radiation at 100 C exceeds 3 m. The overlap at about 3 m is
close to zero.

1.0

0.9

0.8 300 C
0.7
Relative intensity

Reflectance

0.6
3 m
0.5

0.4

0.3 200 C
0.2

0.1
100 C
0
0.3 0.5 0.8 1 1.6 2 2.5 5 10 20
Wavelength (m)

Figure 3.1: Intensity of the solar radiation reaching the surface of the Earth and the
blackbody radiation at 100, 200 and 300 C normalized to maximum radiation. The
overlap at 100 C is negligible. The gray line shows the reectance of an ideal spec-
trally selective solar absorber for low temperature applications (less than 100 C).

30
3.2 Electromagnetic radiation and its interaction with
materials

An electromagnetic wave, i.e. light, that hits a surface will either be reected
by the surface, absorbed in the material or, if the material is transparent, trans-
mitted through it. The fraction between reected, absorbed and transmitted
radiation depends on the wavelength of the light, as well as the properties of
the material, but it will always add up to one according to

R( ) + A( ) + T ( ) = 1, (3.3)

where R is the reectance, A is the absorptance, T is the transmittance and is

the wavelength. For opaque materials, such as solar absorbers made of metal-
lic substrates, and thereby non-transparent to light, T = 0 and Equation 3.3
becomes

R( ) + A( ) = 1. (3.4)

A further relation, Kirchhoffs law, says that the radiation emitted by a sur-
face (emittance E ) at a certain wavelength is equal to the absorbed radiation
(absorptance A) at the same wavelength at thermal equilibrium, according to

E( ) = A( ). (3.5)

Combined with Equation 3.4 this yields

E( ) = A( ) = 1 R( ). (3.6)

An ideal solar absorber would distinguish between different wavelength

ranges according the gray line in Figure 3.1. The emittance should be as low
as possible, and accordingly the reectance high, in the long wavelength
range to minimize the thermal losses. In the solar wavelength range, on
the other hand, the reectance should be as low as possible to admit high
absorptance. A material with considerable different optical properties in
different wavelength intervals is called spectrally selective.

A solar absorber is characterized by the optical properties solar absorptance,

sol , and thermal emittance, therm , at 100 C, both dened at near normal

31
incidence. Absorptance is, however, difcult to measure, while reectance is
easily measured. Absorptance and emittance are therefore calculated by nu-
merical integration, using the relation in Equation 3.6. The solar absorptance
is dened as the average absorptance weighted against the solar spectral in-
tensity [28], according to

 4.1m
Isol ( )(1 R( ))d
sol = 0.3
 4.1m (3.7)
0.3 Isol ( )d

where Isol ( ) is the intensity of the solar radiation and R( ) is the reectance
at wavelength . Similarly, the normal thermal emittance is given by the av-
erage emittance over the infrared wavelength range weighted with the black-
body intensity according to Plancks blackbody radiation distribution, Ib , at
100 C, according to

 50m
Ib ( )(1 R( ))d
therm = 2.5
 50m . (3.8)
2.5 Ib ( )d

Ideally, the solar absorptance should be one and the thermal emittance zero.
Commercial absorbers rather have absorptance values of 0.90-0.95 and emit-
tance values of 0.05-0.15.

3.3 Solar collectors

A thermal solar collector is a device that absorbs a large portion of the solar
radiation and efciently converts it to useful heat. The energy gain is the dif-
ference between the absorbed solar energy and the losses from the collector,
which is thoroughly described in [29]. The losses are of three types, conduc-
tion, convection and radiation. To minimize the convection a cover glass is
placed on top of the absorber. The conduction losses are reduced by insulat-
ing the collector box and nally, the radiation losses can be suppressed by
modifying the absorber material. The collector output per unit collector area
is given by

Qu = Qcoll QL1 QL2 Qinert (3.9)

32
where Qcoll is the absorbed solar energy and QL1 mainly describes the losses
through conduction and convection. Losses through radiation are foremost
represented by QL2 , while Qinert is the losses due to the thermal inertia of the
collector, which causes a delay in heat production as the ambient temperature
changes. In total, this yields the equation

dT f
Qu = 0 (Kb ( )Gb + Kd Gd ) F UL1 T F UL2 (T )2 (mCp ) , (3.10)
dt

where 0 is the zero loss efciency and F  is the mean temperature collector
efciency factor. The angle modiers for beam and diffuse radiation, Kb and
Kd , take the non-normal incident radiation into account, Gb and Gd are the
irradiance for beam and diffuse radiation in W/m2 , while UL1 and UL2 are the
rst and second order heat loss coefcients. Further, T is the temperature
difference between the heat carrier and the ambient air, m is the mass ow
rate of the uid, Cp is the heat capacity of the uid, T f is the uid temperature
and t is time. [29, 30]

This means that the most important parameters for solar collector output are
the irradiance on to the collector and the temperature difference between the
heat carrier and the ambient air. The three parameters 0 , UL1 and UL2 are
often found on data sheets for commercial solar collectors and are used as
input data in solar heating simulations. The efciency of the solar collector,
, is the ratio between collector output and the global solar irradiation, G,
according to

Qu
= . (3.11)
G

3.3.1 Different collector types

There are four main types of solar collectors, the at plate solar collector,
the vacuum tube collector, the concentrating collector, and the unglazed col-
lector, see Figure 3.2. The simplest and cheapest kind of solar collectors is
the unglazed, which only consists of the absorber and no glazing or box. The
losses are high and it is therefore preferably operated at low temperatures.
Swimming pool applications with unglazed collectors is today the single most
economically protable solar collector application.

33
The main part of the concentrating collector is a parabolic reector that con-
centrates incoming solar radiation onto the absorber. The advantage is the cost
per unit area, since reector material is less expensive than the absorber. It fur-
ther enables higher utilization of the low standing winter sun, as well as avoid-
ing overheating in summer. However, only direct and not diffuse radiation can
be focused on the absorber and to maximize the efciency an expensive solar
tracking system is required.

(a) Flat plate collector on a single-family house (b) Vacuum tube collectors on a multi-family
in Vnns, north of Sweden. house in Xian, China.

(c) One type of concentrating collector on lab- (d) Unglazed collector on outdoor swimming
oratory in Lund, south of Sweden facility in Malm, south of Sweden

Figure 3.2: The four main types of solar collectors; the at plate collector, the vacuum
tube collector, concentrating collector and the unglazed or swimming pool collector.
Photos: M. Lundh.

So called vacuum tube collectors are frequently used in China, but are also
gaining increasing market shares in Sweden and other European countries.
The absorber is placed in an evacuated tube to utilize the insulating effect of
vacuum, and the collector comprises several such tubes in a frame. The collec-
tors can not be integrated in the roof construction, but the absorber can easily
be tilted inside the tube and does not need to follow the tilt of the roof. Due
to the vacuum technique high temperature heat can be generated and the ef-
ciency is high even at high temperature differences to the surroundings and
at low solar irradiation. The average annual system efciency is 45-50 per-

34
cent [31], which is higher than for the usual at plate collector, but vacuum
tube collectors are also more expensive.

3.3.2 Flat plate collectors

The at plate collector dominates the world market for hot water heating by
85-90 percent [32]. It was mainly developed during the 1980s in terms of the
absorber, insulation, convection barriers and larger modules, but also manu-
facturing costs. The main advantages are the simple and stable construction,
the well developed technique and the low manufacturing costs, but also that
it allows roof integration. A schematic gure of a at plate collector is found
in Figure 3.3. The main component, the absorber, is placed in a metal box
which is insulated with glass bre or mineral wool both on the back and edges
to minimize the losses. The box is covered by a glass, preferably a low-iron
glass with an antireection coating to allow high transmittance of solar radia-
tion, but reduce the convection losses.

Figure 3.3: Schematic gure of a at plate solar collector with the most important
parts highlighted. The absorber is the heart of the collector, while the cover glass
and insulation reduce the losses.

An ordinary modern at plate solar collector annually generates on average

350-450 kWh/m2 at a global solar irradiance of 1 000 kWh/m2 and year. This
yields an average annual system efciency of 35-40 percent [31]. The working
temperature of the at plate collectors varies between 20 and 90 C [32]. A
lower working temperature, and thereby lower temperature difference to the
surrounding, increases the efciency and thereby the energy yield.

The radiation is converted to heat in the absorber, which usually consists of

a metal sheet made of copper, aluminium or steel, the so called n, covered
by a thin, visually dark, absorbing layer, see further in Chapter 4. Copper
pipes are attached to the back or inside of the absorber, in which a liquid
or gaseous medium is circulated to remove the heat from the ns. In the next

35
chapter the absorber and favourable optical properties of the absorbing coating
applied to the n are introduced in detail. The focus is on absorber paints,
which are easily applied to metal sheets, and a study made to improve the
optical properties of two paints.

36
4. Solar absorbers

In this chapter different solar absorber materials are introduced. The focus
is on so called Thickness Sensitive Spectrally Selective (TSSS) paints. The
possibility of improving the optical properties of two such paints has been
investigated by antireection treatment. The procedure and results, as well as
a discussion on painted absorbers in a larger perspective, are presented.

As mentioned above, the absorber is the main component in a thermal so-

lar collector. It should efciently transform incident solar radiation to useful
thermal energy (heat) by a photo-thermal conversion process. The thermal en-
ergy should then be transferred to a heat carrier medium with as small losses
as possible. To achieve a high efciency the absorber should absorb all solar
radiation and, ideally, emit no thermal radiation, i.e. be spectrally selective.
But what properties should a material possess to differentiate in response to
wavelength?

4.1 Spectrally selective materials

According to Figure 3.1 an ideal spectrally selective absorber should absorb

all incident solar radiation, without re-emitting the energy as heat. According
to Equation 3.6 and 3.7 this means that a good spectrally selective absorber
should have low reectance, and thereby high absorptance, for wavelengths
shorter than 3 m and high reectance at wavelengths longer than 3 m. The
transition should be moved towards shorter wavelengths for higher operation
temperatures, according to Equation 3.2. Fortunately, reectance is an easily
measured quantity by use of a spectrophotometer1 .

Spectrally selective materials can be designed in several different ways. The

main types of spectrally selective absorber materials, the intrinsic absorber, the
tandem absorber and textured surfaces, are presented in the following sections.

1 Optical measurement equipment.

37
4.1.1 Intrinsic selective materials

The simplest selective solar absorber would be materials that possess spec-
tral selectivity intrinsically. There are, however, no natural materials where
the transition from low to high reectance appears at the appropriate wave-
lengths. Instead, the materials have to be modied to suit the application as
solar absorber. For example, there are alloys with intrinsic spectral selectivity,
as well as pure metals. In metals, however, the transition edge often occurs at
too short wavelengths. Semiconductors may also possess intrinsic spectrally
selective properties, but the transition in reectance often occurs at too long
wavelengths, in the infrared, which causes high thermal losses. The edge can,
to some extent, be moved towards shorter wavelengths by heavy doping. [33]

4.1.2 Tandem absorbers

Most modern solar absorbers are so called tandem absorbers, which consist
of two materials with different optical properties. Usually a highly absorptive
material is applied on to a highly reecting substrate, such as a metal sheet.
The absorptive layer allows high absorption of solar radiation, but when made
sufciently thin it is transparent to the infrared radiation which is reected by
the underlying metal sheet. Both semiconductor-metal tandems with appropri-
ate band-gaps and metal-oxide particles in a dielectric matrix are common tan-
dem materials. Another is the Thickness Sensitive Spectrally Selective (TSSS)
paints, which consist of a highly absorptive inorganic pigment dispersed in a
resin. This material is further described in Section 4.2.

4.1.3 Textured surfaces

By using micro structuring techniques the surface of a material can be made a

light trap by multiple reections. The structure should have the size of the so-
lar wavelengths to trap the solar radiation and thereby increase the absorption
in this wavelength range. The surface will, however, appear smooth to the long
wavelength thermal radiation, which will then be reected. An advantage of
the textured absorbers is the stability at high temperatures, although too high
temperatures could cause oxidization and thereby change the microstructure
of the material.

38
4.1.4 Antireection treatment

The amount of radiation reected by a surface depends on the refractive index

of the material compared to that of air. For normal incidence the reectance,
R, is given by

   
 N1 N2 2  n1 ik1 n2 + ik2 2
R =   = 
 n1 ik1 + n2 ik2  , (4.1)
N1 + N2 

where N1 and N2 are the complex refractive indices of the incident medium
and reecting medium respectively, and N = n ik, where n is the real refrac-
tive index and k is the extinction coefcient. The refractive index of air is 12 .
The solar absorptance of an absorber can often be improved by reducing the
front reectance. This can be done by adding an antireection (AR) layer on
top of the absorber surface. The AR coating should have a refractive index
between that of the surface material and air. The thickness of an antireection
coating should be made thin enough not to increase the thermal emittance,
usually 50-100 nm.

With non-scattering materials thin lm interference is utilized to obtain a good

antireection coating. By optimizing the refractive index of the AR coating
destructive interference can be utilized to minimize the surface reectance.
The optimum AR coating has a refractive index that is the square root of the
refractive index of the absorber material, which cancels the reection com-
pletely at a particular wavelength.

On rough surfaces, on the other hand, destructive interference can not be uti-
lized, since the radiation will be reected in the front surface, but also highly
scattered by the particles in the surface lm. The reectance of the front sur-
face can still be reduced by introducing an AR coating, creating a broad spec-
tral reduction in reectance by reducing the difference between the refractive
index of air and that of the surface material.

4.2 Thickness Sensitive Spectrally Selective paints

One cheap type of absorber material is the commercially available Thickness

Sensitive Spectrally Selective (TSSS) paints. They are easy to use, since they
could be applied with an ordinary hand spray gun, but are also suitable for

2n = 1, k = 0

39
large area preparation. The selectivity of the paints lies, as indicated by the
name, in the thickness of the paint layers. The paints consist of active pig-
ments that absorb solar radiation, and up to a certain thickness they will be
transparent to longer, thermal, wavelengths.

The TSSS paints consist of a resin binder, proprietary pigments, curing agents
and adhesion promoters admixed in a solvent [34]. The paint layer thickness
should be in the order 2-5 m. Compared to ordinary black paints TSSS paints
have considerably better optical properties, due to the selectivity and thereby
relatively small thermal losses. Their main drawback compared to other ab-
sorber coating materials, such as metal-ceramic composites, is their relatively
high emittance values.

Two different TSSS paints have been investigated in this study. One is Solkote
HI/SORB-II, which was developed in the USA and has been in use since
1982 [35]. According the product information, application on to aluminium,
copper or stainless steel will cause absorptance values between 0.88 and 0.94,
while the emittance ranges from 0.28 to 0.49, depending on the substrate and
the paint layer thickness. The paint consists of xylene as a solvent, proprietary
oxide pigments and a silicon polymer as a binder. The last constituents are not
specied by the manufacturer. [35]

The second paint is the Solarect-Z, developed in Slovenia. The paint consists
of a silicon resin, an organic solvent, and an active inorganic pigment speci-
ed as FeMnCuOx . The pigment has a primary particle size of 400 nm. Opti-
mum absorptance values of 0.90 to 0.92 with corresponding emittance values
ranging from 0.20 to 0.25, depending on the paint thickness, are given by the
producer [34].

4.2.1 Project objective

Both the absorptance and emittance of the TSSS paints increase with increas-
ing paint layer thickness up to a certain maximum absorptance value. The limit
in absorptance for the Solkote paint was slightly less than 0.95 and somewhat
higher for Solarect. A further increase in paint thickness only yields an in-
crease in emittance for both paints. To optically compete with other absorber
materials further enhancement in absorptance, without increasing the emit-
tance, would be desirable. The aim of this study was therefore to investigate
the inuence of an antireection coating on the optical properties of the paint
layers, since a reduced front reectance could result in enhanced absorptance.
As the paint surfaces are rough and the pigments will cause internal scattering
in the material, the effect of thin lm interference can not be utilized. Instead,

40
 the ambition was to get a broad spectral decrease in reectance in the solar
spectrum. Only one kind of antireective material was examined and an opti-
mization of the AR coating was not performed. The study is reported in detail
in Paper I as well as in [36].

4.2.2 Results

The cross sections of typical painted samples with an AR coating are found in
Figure 4.1, while optical properties of all studied samples are shown in Fig-
ure 4.2. It can be seen that the maximum absorptance may be exceeded by anti-
reection treatment with the studied AR solution. The absorptance increased
after the treatment for all samples. A difference between the two paints was,
however, observed.

300 nm 1 m

(a) Solkote (b) Solarect

Figure 4.1: SEM micrographs of thin Solkote and Solarect samples with AR coat-
ings. The aluminium substrate, paint layer and AR coating are all seen [36].

0.6 0.6
Solkote paint Solarect paint
AR treated paint AR treated paint
0.5 0.5

0.4 0.4
Emittance

Emittance

0.3 0.3

0.2 0.2

0.1 0.1

0 0
0.88 0.9 0.92 0.94 0.96 0.98 0.88 0.9 0.92 0.94 0.96 0.98
Absorptance Absorptance

(a) Solkote (b) Solarect

Figure 4.2: Emittance versus absorptance for samples painted with Solkote and So-
larect respectively, before and after AR treatment.

41
The largest improvement is achieved for thick paint layers. The absorptance of
the paint did not for any thickness exceed 0.95, but after AR treatment reached
almost 0.97 for Solarect and almost 0.96 for Solkote, while the emittance was
still slightly decreased. For thin Solkote samples the same optical properties
are achieved by increasing the paint layer thickness, thereby avoiding the ad-
ditional time and effort of AR treatment. On Solarect samples, on the other
hand, an AR coating can not in any way be compensated by a thicker paint
layer.

4.2.3 TSSS paints in a larger perspective

To discuss TSSS paints in a larger perspective the outcome in Adsten et

al. [37] was used. The inuence of the optical and thermal properties of
the solar absorber on the annual energy output from a solar collector was
investigated by calculations in the simulation program Minsun. Swedish
climate data and xed average operating temperatures were used [37]. The
reference solar collector had a normal absorptance, sol , of 0.95 and a
thermal emittance, therm , of 0.10. Increasing the absorptance to 0.97, keeping
the emittance constant, approximately corresponds to antireection treatment
of thick TSSS paint layers. The emittance in Adsten et al. is somewhat
lower than that of the paints. This, however, only affects the absolute energy
output, while the relative increase is assumed to be representative even for
higher emittance values. It is shown that increased absorptance only yield a
moderate improvement; less than ve percent for all working temperatures
between 30 and 70 C. In the case of painted absorbers the improvement is
too small to economically and energetically compensate for the additional
process of antireection treatment. Even though increased absorptance is
shown to be more important than a corresponding decrease in emittance,
the combination is shown to inuence the annual energy output most
signicantly [37]. Optimized AR coatings may therefore make the process
more economically viable.

Conclusively, cheap solar absorber materials are essential to lower the cost
of the collector, but in the case of TSSS paints the achieved improvement
does not justify the additional manufacturing step. There are also, already, ab-
sorbers with excellent optical properties on the market, although not as cheap
as the paints. Another way to increase the cost-effectiveness, and thereby the
implementation rate of solar thermal, is to increase the overall supply of solar
energy by system optimization. In the next chapter some system solutions and
other system aspects on solar heating are presented.

42
5. Solar heating systems

Improved materials and components are important for the development of so-
lar heating, but different system aspects are also essential to actually reach
implementation and dissemination of the technology. In this chapter different
possible system solutions are presented, from small-scale solutions for single-
family houses to large-scale long-term solutions for residential areas. Apart
from proper system designs, both energy savings and inuence on the over-
all energy system must be taken into consideration. Solar heating is therefore
discussed at different system levels. First, a general introduction is given to
different system congurations, then the past and present solar heating market
is presented. The different system projects performed within this thesis work
are then described; planning of solar heating, small-scale solutions with solar
and wood pellet, medium-sized storages for single-family houses, a seasonal
storage for a residential area and nally assessment of heating systems.

5.1 System solutions

Small-scale solar heating systems are mainly designed in two different ways,
either as hot water systems or as combi-systems, which supply both hot water
and space heating, see Figure 5.1 and 5.2 respectively. Hot water systems are
still more common in the world as a whole [31], while in Sweden, combi-
systems are more frequently installed [38]. The distribution between different
system solutions in Sweden is 25 percent hot water systems for single-family
houses, 65 percent combi-systems for multi-family houses and district heating
and 10 percent combi-systems for single-family houses [39].

In temperate climates, like in Sweden, a small-scale solar hot water system is

often dimensioned with 0.6-1.0 m2 solar collector area and 40-60 litres storage
volume per person. This yields annual fractional energy savings1 for hot water
of about 50-60 percent [31]. The optimum size of a combi-system can not be
dened as easily. The systems are often dimensioned for annual fractional

1 Fractional
energy savings is the ratio between auxiliary energy needed in the solar heating
system and with conventional heating.

43
energy savings of 20-25 percent [40] and according to [41] a system with
15 m2 collector area with 30 degrees tilt in Stockholm (latitude 59.3 N) could
yield energy savings of almost 30 percent, while measurements on a 12 m2
solar combi-system in Bors (latitude 57.7 N) yielded 11 percent fractional
energy savings [42].

Solar collector
Domestic hot water

Cold water supply

Auxiliary
heater

Heat exchangers
Pump
Cold water supply

Tank

Figure 5.1: Schematic sketch of a small-scale solar hot water system.

Since heat is usually not simultaneously produced and used a short-term heat
store or accumulator tank is required. This is most often a water tank that can
store hot water for one or two days. A completely stratied tank, where the
colder water with higher density is in the base and the warmer water with
lower density is in the top, would be the most advantageous solution for solar
heating applications. Mixing will, however, always occur in a real tank, both
due to charging, discharging and vertical heat transfer in the tank. Charging
the tank at the height where the water has exactly the same temperature as the
incoming water would reduce the mixing. Different stratication devices have
been developed [43], but the usefulness is still controversial.

Solar collector

Domestic hot water

Cold water
supply

Auxiliary
heater
Electrical
heater Space heating
Heat exchangers system
Pump

Cold water supply

Tank

Figure 5.2: Schematic sketch of a small-scale solar combi-system.

44
By charging solar energy into the base of the tank, even rather low temper-
atures can be used to pre-heat the hot water. The tap water is then heated to
the desired temperature by the auxiliary heater in the top of the tank. A lower
hot water set temperature, but also a low-temperature space heating system,
increase the fractional energy savings. To avoid bacteria in the tank, such as
Legionella, the temperature should, however, exceed 50 C [31].

5.2 The solar heating system market

There have been governmental subsidies for solar collector installations in

Sweden for periods since the late 1970s. The previous subsidy was available in
1991-1996 and was based on the total investment cost. In 2000 a new subsidy
was introduced, based on the energy output of the system [44]. Evaluations of
the effects of the subsidy shows that both Swedish production of solar collec-
tors and import has increased. The impact of subsidies on the installation rate
is also seen in Figure 5.3. The last years increase is mainly for small systems
up to 15 m2 , and to 85 percent combi-systems. One of the main purposes of
the subsidy was to decrease the electricity dependence, and more than half of
the installations, or about 23 GWh 2 replaced electric heating. [38]
Sold solar collector area (1000 m )
2

25

20

15

10

0
1977 1980 1985 1990 1995 2000 2005
Year

Figure 5.3: Sold solar collector area from 1977 to 2007 in Sweden [38, 44]. The rst
peak in installed collector area in 1980 is due to the oil crisis and the other peaks
are mainly due to governmental subsidies for solar thermal installations, 1983 [45],
1987 [45], 1991-1996 and from 2000.

Austria and Germany have similar subsidies, supporting solar heating system
installations. In Israel, as well as in several provinces in Spain and in India, it
is mandatory to use solar hot water systems in new buildings. Moreover, the
goal in China is to supply solar heated hot water to 20-30 percent of the inhabi-

2 The calculations were performed at 50 C average working temperature.

45
tants in 2015, which corresponds to 230 million square meters solar collectors.
The reported sold glazed collectors in China in 2003 correspond to more than
8 000 MWth [38]. The markets in Europe, as well as in the world as a whole,
are found in Figure 5.4. In year 2000 Greece, Austria and Turkey, followed by
Japan, were the leading countries with 113-264 m2 new installed glazed col-
lectors per 1 000 inhabitants. In total there were almost 40 million m2 glazed
collectors installed, corresponding to more than 24 TWh heat per year, in the
member countries of the International Energy Agency (IEA) Solar Heating
and Cooling Programme (SHC). From 1999 to 2000 the installation of at
plate collectors increased by 13 percent and apart from Mexico, the largest
increase was in Sweden [39], most likely due to the subsidy mentioned above.
Installed solar heating (MW/year)

1000 Holland, UK, Sweden & Denmark

Germany
800 Austria
Portugal, Spain, France & Italy
600 Greece

400

200

0
90 92 94 96 98 00 02 04 06
Year

(a) The European market

Installed solar heating (MW/year)

14000
China
12000 EU15
10000 Israel & Turkey
USA, Australia & Japan
8000
6000
4000
2000
0
82 84 86 88 90 92 94 96 98 00 02 04 06
Year

(b) The world market

Figure 5.4: Sold thermal solar collector systems in Europe and the world respectively,
from [38, 39, 46, 47, 48, 49, 50]. Note the different scales.

5.3 Planning new solar heating plants

One prerequisite for solar heating installations is the possibility of performing

realistic estimations of solar energy contribution to a certain building. An ob-

46
vious obstacle to do that is the commercially available planning tools, but also
the knowledge about hot water use in household and how it affects the solar
energy utilization. In the following an evaluation of a building simulation tool
is presented, as well as a model built to generate hot water load proles from
time-use data and the inuence of different load patterns on the fractional en-
ergy savings.

5.3.1 Reliable planning tools

During simulations in the commercial building simulation program VIP+ se-

rious shortages in the solar heating model were discovered. Those problems
resulted in lower than realistic solar contributions and thereby low fractional
energy savings. The consequence of widespread use of this type of simulation
tools may be unwillingness to install solar heating. Instead proper tools to esti-
mate the actual solar contribution, and that are easily handled by the planners,
are required. In the following the evaluation, but also the improvement made
by the programmers after being informed about the shortages, is presented. A
more extensive description of the evaluation is found in Paper II.

Apart from VIP+, two solar heating simulation tools, Winsun Villa3 [51] (re-
ferred to as Winsun below) and Polysun4 [52], as well as measured data from
a study from SP Technical Research Institute of Sweden [42], were used in the
evaluation process. The programs have somewhat different purposes. The cal-
culations in VIP+ aim at estimating the energy demand and the heat balance of
a building and enable easy comparison to the Swedish Building Regulations.
Both Winsun and Polysun, on the other hand, focus on solar heating systems.
More parameters describing the solar heating equipment are therefore open to
the program user to specify. All three tools are, however, based on the physi-
cal description of a solar collector in Equation 3.10 and utilize hourly climatic
data from the global meteorological database Meteonorm [53], although Win-
sun also allows for measured climatic data.

The storage tank for solar heat varies between the simulation tools. In VIP+
the tank model represents a fully mixed tank5 . As soon as the temperature in
the solar collector exceeds the mean temperature in the store solar energy is
fed into the tank. The minimum working temperature of the solar collector is

3 Winsun Villa is based on the simulation tool TRNSYS [19] and constitute a TRNSed or user
interface where only a limited set of parameters are open to the user. The program was devel-
oped by Bengt Perers at Lund University in collaboration with SERC, Dalarna University, based
on the solar heating program Minsun.
4 Polysun 3.3 was used in this study. There are now more advanced versions available.
5 All water in a mixed tank takes the same average temperature.

47
the same as the minimum temperature of the tank. This should, however, also
be the minimum temperature required by the heating system, not to feed too
low temperatures into the hot water system [54]. In both Winsun and Polysun
stratied tanks are modelled 6 . In Winsun there is stratication in three layers
and the solar heat is supplied via a heat exchanger in the base of the tank,
while auxiliary heating is introduced in two steps; pre-heating in the middle
and nal heating in the top of the tank. In Polysun, on the other hand, the tank
is stratied in twelve layers, where the solar heat is fed in to the lowest tank
layer, while the auxiliary heating can be introduced at any level.

Evaluation of the solar heating model

Both a hot water system and a combi-system were studied to evaluate the
solar heating model in VIP+, all with climatic data for Stockholm. The hot
water systems were dimensioned to cover all tap water demand in summer.
The solar energy contribution on a weekly basis from simulations in VIP+
and Winsun 7 are found in Figure 5.5. The main difference between the two
programs is the solar contribution during spring and autumn. In VIP+ no solar
energy is supplied from week 39 to 11, which means the end of September to
mid-March. Winsun, on the other hand, shows an obvious contribution even
during the less sunny seasons.

90
VIP+, 7 m2
80
Solar contribution (kWh/week)

2
Winsun, 6 m
70
60
50
40
30
20
10
0
10 20 30 40 50
Week

Figure 5.5: The solar energy contribution per week from a hot water system in Stock-
holm climate covering the peak hot water demand during summer; 7 m2 solar collec-
tors and 525 litres tank in VIP+ and 6 m2 collectors and 450 litres tank in Winsun.
The collector tilt is 27 degrees.

To further evaluate the solar heating model simulations were compared to

measurements for a solar combi-system, see Figure 5.6(a). Again, there is

6 Ina stratied tank there is a temperature gradient with higher temperature in the top and lower
temperature in the bottom.
7 Polysun 3.3 does not allow weekly time resolution of output data.

48
an evident difference between VIP+ on the one hand and measurements, Win-
sun and Polysun on the other. The total annual solar contribution calculated
in VIP+ was less than half of the total measured useful solar energy. Win-
sun and Polysun yielded somewhat higher total solar contribution compared
to the measurements, which is mainly explained by the lower solar irradiation
the measured year compared to the climatic data used in the simulation.

800 800
Measured Measured
Solar contribution (kWh/month)

Solar contribution (kWh/month)

700 VIP+ 700 VIP+
Winsun Winsun
600 Polysun 600 Polysun
500 500

400 400

300 300

200 200

100 100

0 0
2 4 6 8 10 12 2 4 6 8 10 12
Month Month

(a) VIP+ 4.0 (b) VIP+ 5.0

Figure 5.6: The measured solar energy contribution compared to simulations with
climatic data for Stockholm. The left gure shows the results from VIP+ 4.0 and the
right gure the results achieved with VIP+ 5.0. Measured data is from [42] and all
systems comprise 12 m2 solar collectors tilted 27 degrees and 1 m3 tank.

The deviation between results from VIP+ and measured data, as well as sim-
ulations in the standard solar heating tools, was identied to the tank model
in VIP+. The tank in VIP+ is not only mixed, but is also exclusively used for
solar heat and a certain minimum set temperature has to be reached by all
water in the tank before the heat can be discharged. The advantage of using
solar collectors for pre-heating in spring and autumn is thereby lost, which
also results in the nonexistent solar energy contribution from September to
March in VIP+. On the contrary, in real solar heating systems both solar heat
and auxiliary sources are used to charge the tank, where low-temperature solar
heat is fed to the base of the store and auxiliary heat to the top, which allows
optimized use of the solar heating system.

The identied problems were communicated to the program developer. In the

next version, VIP+ 5.0, the tank model was debugged and improved; partic-
ularly the unrealistic losses from the tank had been taken care of [55]. The
previous tank model was also supplemented by a more detailed model that
is based on temperatures in the heating system rather than energy ows. The
results were again compared to the standard solar heating simulations and
the measurements, which is found in Figure 5.6(b), and shows a signicant
improvement. Figure 5.7 further shows the considerable difference in solar

49
 800
VIP+ 4.0, built env.
700 VIP+ 5.0, built env.
VIP+ 4.0, clear horizon
600 VIP+ 5.0, clear horizon

Solar energy (kWh)

500

400

300

200

100

0
2 4 6 8 10 12
Month

Figure 5.7: The solar energy contribution for a 12 m2 collector area at 27 degrees
tilt and 1 m3 tank, calculated in VIP+ version 4.0 and 5.0 for clear horizon and built
environment (30 degrees horizontal angle) respectively.

energy utilization in spring and autumn with an improved tank model. The re-
sults are given both for a clear horizon (0 degrees) and for a built environment
(30 degrees). The difference only comes down to the tank model, since the so-
lar collector description was not changed in the new version of the program.

5.3.2 Hot water load model

In solar heating simulations the dynamic behaviour of household members is

often reduced to a schematic hot water load prole, usually assuming three
draw-offs per day, not distinguishing between weekdays and weekend days
or taking seasonal variations into account. However, previous studies have
shown that the load distribution inuences the system performance and the
fractional energy savings [56, 57, 58], which indicate the usefulness of im-
proved load descriptions. The most direct method to display energy demand
is to measure the domestic hot water use, but detailed measurements are both
costly and often require severe interference in the apartment. Few detailed hot
water measurements have been performed so far. One exception is a survey
performed by the Swedish Energy Agency [59]. Still, only ten households
were measured in detail, which is too few to be statistically signicant.

An alternative approach is to model load curves. Jordan and Vajen [56]

present a model to generate hot water proles based on probability functions.
The model was developed within IEA-SHC Task 26 and has been used in
several investigations, for example [57, 60]. Stochastic models based on
Swedish measurements have also been developed to predict domestic hot
water load [61, 62]. Existing models, however, seldom focus on the inuence
of the individual household members and their behaviour. In this thesis,

50
on the other hand, such an approach was applied. A method to translate
time-use data (see Section 2.6) into hot water use was developed. The model
aims at presenting the daily distribution of hot water demand of individual
households as well as average Swedish households, based on peoples
statements about their activities. A certain energy demand is thereby directly
connected to a specic activity or user pattern. The model and comparisons
to measured data are described in Paper III and application of the proles
in simulations of a solar hot water system is presented in Paper IV. The
model and outcome has also been described and compared to measurements
in [63, 64, 65].

1.8 1.8
Volume (litres per 5 minutes)

Volume (litres per 5 minutes)

1.6 1.6
1.4 1.4
1.2 1.2
1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0 0
6 12 18 24 6 12 18 24
Hour Hour

(a) Detached houses, weekday (b) Detached houses, weekend day

3.5 3.5
Volume (litres per 5 minutes)

Volume (litres per 5 minutes)

3.0 3.0

2.5 2.5

2.0 2.0

1.5 1.5

1.0 1.0

0.5 0.5

0 0
6 12 18 24 6 12 18 24
Hour Hour

(c) Apartments, weekday (d) Apartments, weekend day

Figure 5.8: Modelled hot water use for an average household member in detached
houses and apartments for weekdays and weekend days respectively. The time resolu-
tion is the lowest possible for the model. Note the different scales for detached houses
and apartments.

The basic data in the model is from a pilot study on time-use performed in
1996 by Statistics Sweden and later coded at Linkping University [22]. A
total of 464 persons in 179 households, both from single-family houses and
apartments, participated in the study. The households were selected by Statis-
tics Sweden and the selection is assumed to be representative for Swedish

51
households. The diaries were written for one weekday and one weekend day
by all household members older than ten years. It should be noted that the
time-use study did not focus on energy use, but on time-use in general. There-
fore, the participants are assumed not to have adjusted the time statements to
be better from an energy perspective.

The hot water load model was developed in the computer program Matlab
by connecting parameters describing the type of tapping (extension in time
and draw-off pattern) and the required volume of hot water to different activ-
ities. Hot water volumes and temperatures can easily be changed to suit other
purposes. The model output is hot water proles, either given in volumes or
energy/power, for individuals, households or groups of people, separated in
apartments and single-family houses. Time resolution from ve minutes in-
tervals, see Figure 5.8, up to one hour can be chosen. The model can also be
applied to any time-use data organised similarly.

Comparison between modelled and measured hot water use

To estimate the validity of the model the hot water use of an average house-
hold member in an apartment was compared to measured data (see further in
Lundh et al., 2008 [65]). Figure 5.9 shows the hourly distribution for an aver-
age household member for a weekday and a weekend day both when averaged
over six measured months and when averaged for one single month. The mod-
elled and measured load follow more or less the same pattern. Compared to
the average measured proles (Figures 5.9(a) and 5.9(b)) the typical pattern
is more clearly distinguished for single months (Figures 5.9(c) and 5.9(d)),
especially in the morning. This is due to the peaks levelling out when aver-
aged, since the peak power demand appears at somewhat different times for
different months of the year.

Not only the load pattern, but also the magnitude of the peaks in Figure 5.9 is
similar. The peaks, however, appear at somewhat different times in the morn-
ing. The reason may be that the two measured apartment buildings are situated
in an area, where most residents commute rather long distances to work, and
therefore may have showers earlier than the average person in the larger time-
use study. Measurements in Malm also show morning peaks at 6-8 a.m. in
weekdays and 9-10 a.m. in weekend days [66], which agrees well with the
modelled proles. Overall, the model has shown to describe hot water use in
apartments rather well.

52
 1400 1400
Modelled Modelled
1200 Measured average 1200 Measured average
1000 1000
Power (W)

Power (W)
800 800

600 600

400 400

200 200

0 0
6 12 18 24 6 12 18 24
Hour Hour

(a) Weekday (b) Weekend day

1400 1400
Modelled Modelled
1200 Measured Jan 06 1200 Measured Jan 06

1000 1000
Power (W)

Power (W)
800 800

600 600

400 400

200 200

0 0
6 12 18 24 6 12 18 24
Hour Hour

(c) Weekday (d) Weekend day

Figure 5.9: Measured and modelled hot water proles for an average household mem-
ber in an apartment for weekdays and weekend days, where the measured data is
averaged over six months and January, respectively.

53
5.3.3 User inuence on solar heating systems

Most studies investigating the inuence of the hot water load prole focus on
the difference between the standard three draw-off prole and high-resolved
proles [56, 60], or similar proles varying the ow-rates or total load [58,
57]. In Paper IV ten households from the time-use study, with characteristic
hot water uses, were selected to investigate the inuence of the prole on the
performance of a solar hot water system. Four different load proles were
generated for each household, one directly from the model, one reducing the
demand for showers, one scaling the demand to 200 litres per household and
day8 , and nally, one prole where the typical pattern of the household is kept,
but seasonal variations are taken into account.
Storage tank
500 litres
Solar collectors
5 m2 Auxiliary
heater
6 kW
Domestic Hot
Water

Pump

Figure 5.10: Sketch of the simulated solar hot water system based on the systems in
Linds Park, Gteborg.

The inuence of hot water use on solar combi-systems is most likely smaller
than on a hot water system, since the hot water demand is completely cov-
ered by solar energy during summer, while space heating dominates the heat
demand in winter. A solar hot water system was therefore chosen for the inves-
tigation. A model was built in the dynamic simulation tool TRNSYS [19] by
modifying the standard solar combi-system from IEA-SHC Task 32 [67, 68].
The system is based on data for the solar heating systems in the low-energy
buildings in Linds Park, Gteborg, see Figure 5.10 and Paper IV.

Figure 5.11 shows the thermal fractional energy savings achieved with the
simulated solar hot water system for ve individual households, as well as the
average household in detached houses and apartments respectively, but also
the standard IEA-SHC Task 32 prole and the standard three draw-offs prole.
Lundh model denotes the proles generated directly from time-use data, while
Lundh in Jordan denotes proles generated by implementing the household
patterns in the statistical model developed within IEA-SHC Task 26 [69].

8A total load of 200 litres per day and household is often assumed to be representative.

54
 1.0
Lundh model
Lundh, 200 l/day
0.8 Lundh red. shower
Lundh in Jordan, 200 l/day
fsav,therm
0.6

0.4

0.2

0
Morning Evening Even Midday Many taps Av.det Av.ap T32 Stand

Figure 5.11: The fractional energy savings achieved for ve individual households
with different use patterns, the average household in detached houses and apartments
as well as the standard IEA-SHC Task 32 prole (200 l/day) and three draw-offs pro-
le (200 l/day), respectively. The hot water proles were generated by four different
approaches.

The variation in system performance with the different load proles is small
as long as the total hot water demand is constant. The deviation between re-
sults for proles generated by the same approach is less than three percentage
points for the constant 200 litres load. For the proles generated directly from
time-use data the variation is larger; up to 20 percentage points. The total hot
water load also varies considerably between those proles. The relation be-
tween total hot water demand and fractional energy savings is close to linear,
see Figure 5.12 .

Furthermore, comparing the results for individual households with the two
200 litres approaches (Lundh, 200 l/day and Lundh in Jordan) shows less than
four percent difference. This indicates that the seasonal variations and short
times of very high ow-rates do not signicantly improve the description of
the household when studying the annual or monthly performance. The high
ow-rates may, however, make a signicant impact on the momentary perfor-
mance of a solar heat store due to the destruction of stratication caused by
the draw-off.

When planning the solar heating systems in the residential area Linds Park
a joint system was discussed, but disregarded due to costs and higher pipe
losses, as well as maintenance issues and questions about individual shares
of the solar heated hot water [70]. One argument for a joint system could be
to level out the individual hot water use patterns. The advantages of such a
joint system could be further investigated by comparing the results from this
study with a model of a large, joint system, superimposing the same hot water
proles. Although the effect of characteristic use patterns was here found to

55
 0.70

0.65

0.60

fsav,therm
0.55

0.50

0.45

0.40
0 100 200 300 400
Hot water load (litres/day)
Figure 5.12: The relation between the fractional energy savings and the domestic hot
water load for the investigated solar hot water system and different load proles. Note
the non-zero origin of the y-axis.

be small for similar total demands, there may also be advantages by levelling
out the magnitude in hot water use.

Apart from well-planned use of the stored solar heat other means may in-
crease the solar fraction. Advanced control systems, that for example applies
a specic strategy to a certain household by learning their habits, could reduce
the auxiliary heating demand. By prognosticating the attendance of people in
the building, as well as weather conditions for the coming day, auxiliary heat
could be supplied during nights only when necessary, while enabling more
solar energy to be fed into the tank during sunny days. A further topic for
investigation would be to study the possible benets from such control strate-
gies.

5.4 Small-scale system solution: Solar and wood pellet

Solar collectors can not supply all heat demanded in a building in temperate
climates; an auxiliary heating system is always required. There are several
possible solutions and in Sweden combinations with electric heaters are the
most common. Another system solution increasingly used in Sweden, but also
in Germany, Austria and United Kingdom, is combined solar and wood-pellet
heating. In the beginning, separate standard solar and pellet systems were
combined, and still few companies offer well-adjusted combined solutions.
The systems can either consist of separate sub-systems connected with a joint
accumulator tank, as in Figure 5.13(a), or have the pellet burner integrated in
the store, as in Figure 5.13(b). Other common solutions can be found in [71].

56
 Solar collector Solar collector

Cold water Cold water

supply supply

Pellet
boiler Pellet burner
Electrical Electrical
heater heater

Pump Pump

Cold water supply Cold water supply

Tank Tank

(a) Two connected systems (b) Integrated pellet burner

Figure 5.13: Schematic sketches of two possible system solutions with solar and pellet
heating; one connecting two separate systems in a joint storage tank and the other with
the pellet burner integrated in the store.

Technically, solar and wood pellet complement each other well. Increased sys-
tem efciency and sustainability advocate combined systems rather than indi-
vidual solar/electricity and pellet heating, respectively. Biomass is a vast do-
mestic natural asset in countries like Sweden, but it is also a limited resource
that should not be wasted (see further in Section 5.7). By adding 10 m2 solar
collectors to a pellet heating system, the pellet consumption may be reduced
by 25 percent according to [72]. Although both solar and pellet technologies
are today well-developed and have reached cost-effectiveness, the develop-
ment of combined systems is slow.

5.4.1 Qualitative interviews

Traditionally, the interview methodology is mainly used within the social sci-
ences and not often in science and technology. In this study the development
towards small-scale system solutions combining solar and wood-pellet heat-
ing was investigated by interviews with persons from the two industries. They
gave their view on the situation; what obstacles they experience and actions
they nd necessary to be taken. The interviews, as well as transcription, analy-
sis and reporting, were mainly performed by the author with supervision from
a social anthropologist. The background in science and technology was con-
sidered an advantage, both in understanding the topic and carrying out deeper
technical discussions with the informants. The results from the study are pre-
sented in Paper V, and has also been partly reported in [73].

The present state of combined solar and pellet heating systems was rst in-
vestigated by a literature and Internet survey in the beginning of 2007. The

57
descriptions and advertisements of solar and pellet systems in trade papers
and magazines as well as web pages of Swedish manufacturers, producers
and retailers were mapped. The outcome of the pre-study constituted the basis
for the interview questions as well as the selection of informants. The inter-
views were planned based on the procedures presented in Merriam [17] and
Kvale [27].

Ten qualitative interviews were then carried out during autumn and winter
2007. Among the informants were manufacturers, retailers, installers and rep-
resentatives for the trade associations, all considered being involved in the pro-
cess towards a combined system solution. The number of informants is con-
sidered to full the purpose of the study. The interviews were semi-structured;
topics and questions were formulated in advance, but the interview guide was
not strictly followed. The interview guide, translated into English, is found in
Appendix A. All interviews were recorded and partly transcribed. The mate-
rial was then categorized and analyzed with inuences from Social Construc-
tion of Technology and Social Anthropological theories.

Participating observations were also carried out at three different meetings;

two gathering company representatives within the wood-pellet industry, while
the third was a joint meeting between the trade associations for solar thermal,
solar power, wood pellet and wind power.

5.4.2 The market as experienced by the informants

Four main actor groups, perceived as preventing the development of combined

solar and wood-pellet systems, but possessing the power to support it, were
identied in the interviews; the solar and pellet companies themselves, the
customer, the installer and the government. The informants describe a close
interaction between those actors, and accordingly, they claim various actions
at different system levels to be necessary.

A fundamental drawback, communicated by the informants, is the small size

of both the companies and the industries. This results in poor economic assets
to be used for marketing as well as inability to compete with larger actors on
the heating market. Marketing and knowledge transfer to the public is, how-
ever, considered a prerequisite for further dissemination. They say that rst
with increased awareness, the customer will feel condent enough to exercise
pressure on market actors such as installers and construction companies.

Not only the customers, but also the installers, are perceived as lacking in
knowledge. This in turn is experienced to make them unwilling to argue for

58
and install the systems. Since the installer is the person inuencing the cus-
tomers choice the most, it is an important actor group to involve. To change
the attitude within the installer corps, it is proposed to simplify both installa-
tion and purchase, as well as making the installations more protable. But the
companies must also supply the installer with the proper knowledge as well
as arguments to be used towards the customer.

Both installers and equipment supply companies need a stable market, and to
create long-term stability the government must collaborate with the industries,
the informants say. They demand clear political standpoints and statements. At
present, they say, customers tend to postpone their purchase due to uncertainty
about future energy policies.

Although an increasing number of people show interest in combined solar

and pellet systems, the focus on the requirements of the customer in the tech-
nical development is experienced as too poor. According to the informants,
customers want simple and convenient, but also well-designed systems. Still,
manufacturers tend to focus on improved technology. The interviewees them-
selves, on the other hand, focus very much on the different market actors
which they rather consider to govern the implementation process.

The market situation as described by the informants is complex with several

mutually inuencing actors that may both promote and prevent further de-
velopment of combined solar and pellet systems. A number of obstacles that
have to be overcome are identied, but they also propose different solutions.
In general they communicate a positive attitude. By mutual consent the infor-
mants claim a change towards general acceptance of solar and pellet heating
and as one retailer puts it; it is not longer taboo to mention solar collectors to
a customer. The companies themselves are also perceived as respecting each
other and the other technology to a greater extent, even though a joint trade
association is suggested to further prevent the feeling of we and them. By
joint effort the informants believe it is possible to reach wide implementation,
and they say that combined solar and pellet systems are already close to a
breakthrough; it is just a matter of time.

5.5 Medium-sized mid-term system solution

Contemporary small-scale solar combi-systems for single-family houses of-

ten comprise about 10 m2 solar collectors and a 750 litres accumulator tank.
The resulting solar fraction is 20 to 25 percent of the total heating demand
for space heating and hot water. There are solutions to reach well-functioning

59
and sustainable heating, for example the combination with solar and wood-
pellet described above, but solar energy is still a minor part of the total en-
ergy supply. The auxiliary heating system is the foundation and solar energy
a complement. By increased fractional energy savings solar energy would be
the basis of the heating system, which would most likely make solar heating
a more attractive solution to the customer, but also to a larger extent justify
the costs of the systems. To reach higher solar fractions the systems have to
be made more efcient or larger, or alternatively, the heating demand could be
reduced, in turn yielding a relative increase in solar energy use.

The International Energy Agency has pointed out the need of nding new
storage solutions to increase the use of small-scale solar combi-systems [74].
Stores utilizing Phase Change Materials (PCMs) are under development (see
for example [75]), but have so far not been commercialized. Another possi-
bility is to increase the size of traditional hot water stores. The technology
is both relatively simple and well-known and does not require advanced re-
search and development to be implemented. Problems may, however, occur
when the stores are to be scaled, due to space constraints in basements and
boiler rooms. New single-family houses can be planned for large solar stores,
but in existing buildings both ceiling height and entrance dimensions limit
the possibility of installing large stores of conventional shape. And changing
the building construction is associated with high costs. If the traditional ge-
ometry of water stores could be foregone in favour of more practical ones,
without signicantly reducing the system performance, the market for solar
heating would increase considerably. Both losses and stratication is highly
dependent on the shape of the store. The inuence of the store dimensions on
the performance of a medium-sized solar combi-system was investigated in
Paper VI.

5.5.1 Inuence of store dimensions

To examine how the store geometry affects the system performance annual
simulations were carried out in the dynamic simulation tool TRNSYS [19].
The validated standard solar combi-system from IEA-SHC Task 32 [67, 68]
was used. The simulated building has a heat demand of 60 kWh/m2 in Zrich
(Switzerland) climate and the solar heating system comprises 30 m2 solar col-
lectors and a 2, 4, 6 or 10 m3 storage tank. Thermal fractional energy savings,
fsav,therm , as dened within IEA-SHC Task 26 [41] was used as a measure of
system performance. See Paper VI for further details.

Figure 5.14 shows the thermal fractional energy savings as depending on the
insulation thickness and store height for a 4 m3 store. The results are sim-

60
 0.55

0.50

fsav,therm
0.45

0.40

0.50 m insulation
0.35 0.30 m insulation
0.15 m insulation
0.07 m insulation
0.30
1 2 3 4 5 6 7
Store height (m)
Figure 5.14: The fractional energy savings achieved by different store heights and
insulation thicknesses for a 4 m3 store. Note the non-zero origin of the y-axis.

ilar for the other store sizes, although the effect of the insulation thickness
increases with the store size. The simulation results for 4-10 m3 stores more
or less coincide. This indicates that the particular solar heating system with
30 m2 collector area does not justify stores larger than 4 m3 for this building
and climate.

The store height yielding the minimum auxiliary demand and maximum solar
fraction differs between the different store sizes, while the optimal height-
to-diameter ratio more or less coincide at about 2, see Figure 5.15. This is
also the height-to-diameter ratio of most commercial storage tanks (2-2.5).
Still, reducing the store height to 1.5 meters (allowing for 30 cm insulation
in a room with around two meters ceiling height) only result in a negligible
reduction in fractional energy savings for the small store and a decrease by
two to six percent for the three larger ones.

One part of the store is usually kept at a high temperature to ensure hot water
and space heating supply even during less sunny days. In smaller stores this
so called auxiliary volume is often situated in the top of the tank, but other
congurations are also possible. In the case of large stores, such as the ones in
this study, an internal auxiliary volume would only constitute a very thin wa-
ter layer, and the lower and broader the store, the thinner the layer. The losses,
both to the surroundings and to the colder water layer underneath, would be
signicant, but in practice it would also be difcult or impossible to keep the
hot volume to such a thin layer. An alternative would therefore be an external
auxiliary unit. The thermal performance of one such solution was compared to
the internal conguration. With few exceptions, such as very large and wide
badly insulated stores, the simulations showed a higher performance of the
store with an internal auxiliary volume, see Figure 5.16. Experimental investi-

61
 8000
V=2m3
V=4 m3
7800 3
V=6 m

Qburn (kWh/year)
7600

7400

7200

7000
0 2 4 6 8 10 12
Heighttodiameter ratio
Figure 5.15: The auxiliary heat demand as a function of the height-to-diameter ratio
for three different store sizes: 2, 4 and 6 m3 . Note the non-zero origin of the y-axis.

gations, though, have to be made to establish how thin the layer may be made
in practice.

0.55

0.50
fsav,therm

0.45

0.40 Internal, d = 0.30 m

insul
External, d = 0.30 m
insul
0.35 Internal, d = 0.07 m
insul
External, dinsul = 0.07 m
0.30
1 2 3 4 5 6 7
Store height (m)
Figure 5.16: The fractional energy savings achieved for an internal and external aux-
iliary volume respectively for a 6 m3 store volume.

Conclusively, according to the simulation results, high fractional energy sav-

ings may be achieved also with large water stores of unconventional geome-
tries. The performance of the store becomes even less sensitive to the dimen-
sions with increasing insulation. The practical advantages of this result are
obvious. Medium-sized stores may be installed even where the space con-
straints are usually the limiting factor. This in turn means that solar energy
may constitute the basis in the heating system even in existing single-family
houses not initially planned for large solar heating systems.

By an increased solar collector area, covering a major part of the roof instead
of the relatively small systems used at present, combined with a medium-sized

62
store, the fractional energy savings for single-family houses may be increased
considerably. Solar heat can in those systems be stored for up to two weeks,
compared to a couple of days in contemporary systems. But to achieve sig-
nicantly higher solar fractions, solar heat must be stored from the summer
to the heating season. This is not economically viable for individual single-
family houses due to the large losses associated with long-term heat storage.
The relative losses, however, decrease with increasing store size. For a larger
demand, such as a residential area, large-scale seasonal storages are possible.
One such system is situated outside Stockholm, presented below. The solar
collector area is on average about 50 m2 per apartment, indicating that the
size of the mid-term systems proposed in this section is feasible also under
Swedish conditions.

5.6 Large-scale long-term system solution: borehole

storage in Anneberg

Several large seasonal storages for heat have been built in Sweden, and so
far water tanks are the most common solutions [76]. One example is Lyckebo
outside Uppsala, where a large solar collector system (4 300 m2 ) combined
with a 100 000 m3 water lled, un-insulated, cavity in the rock was built in
1983 for a new-built residential area. For complementary heating an electric
boiler was used. The heating plant was dimensioned to cover all heat demand
in the area [18], but the amount of solar energy more or less only equalled
the 11-15 percent losses from the store [77]. Furthermore, a heating system
comprising 750 m2 roof integrated solar collectors and a 640 m3 underground
steel water tank supplies heat to 48 apartments in Sr, outside Gteborg,
since 1989. This system was dimensioned to covers 35 percent of the total
heat demand [18].

Another possibility is to store heat in the ground. One example is the solar
heating plant in the residential area Anneberg, outside Stockholm. Paper VII
describes the technical evaluation of the heating plant, while a more exten-
sive description, but also a socio-technical evaluation, is found in Jonsson et
al. [78] and Lundh and Lfstrm [79]. The technical performance of the en-
ergy supply system for 2003-20069 is presented in reference [80].

The residential area Anneberg is situated in Danderyd, north of Stockholm.

It was built in 2000 and 2001 and comprises 50 apartments with
tenant-ownership, either as semi-detached houses or terraced houses, see

9 The operation period after the initial evaluation made within this thesis work

63
 Figure 5.17. The construction is similar to other contemporary residential
buildings regarding insulation, air-tightness and energy demand. On the other
hand, the heating system is not conventional. The heat demand should under
design conditions be covered by solar energy to 70 percent by utilization of
a large solar collector area in combination with a seasonal storage in rock.
Both solar collectors and borehole storages are individually well-known
techniques, but the combination was new at the time10 . The pilot plant was
the rst of its kind, and was part of an EU THERMIE project [82].

(a) Semi-detached house (b) Terraced houses

Figure 5.17: Photos of one type of semi-detached houses as well as the two units of
terraced houses in Anneberg. Photos: M. Lundh.

The expected annual heat demand is on average 11 000 kWh per residential
unit, corresponding to less than 100 kWh/m2 and year [83]. On all south facing
roofs a total of 2 400 m2 , or 40-65 m2 per apartment, roof-integrated at plate
solar collectors are mounted, see Figure 5.17. A rock volume of 60 000 m3 ,
penetrated with 100 boreholes, is used for storage. The boreholes are 65 me-
ters deep and double U-pipes are used as heat exchangers. The mean tem-
perature of the rock volume will vary between 25 and 45 C during the year.
The initial losses from this type of store are signicant, but will stabilize at
about 40 to 50 percent after three to ve years of operation, when the store
has reached full capacity. [83] A borehole storage was chosen mainly due to
the appropriate geological conditions, but also since the combination of solar
collectors and boreholes had never been demonstrated. The ambition was a
pilot plant utilizing the two techniques in combination with a low temperature
heating system, without a heat pump [84].

There are three types of residential units, distributed according to Figure 5.18;
two types of semi-detached houses and two units of terraced houses. The area
is divided into sub-units of two, four or seven apartments sharing a sub-central

10 Atthe same time a similar plant was built in Neckarsulm in Germany, which was taken into
operation in 1999 [81].

64
for heat distribution. They share one or two buffer tanks of 750 litres each,
as well as the connection between solar collectors and the heat distribution
system. Every apartment has its own hot water tank of 100 litres equipped with
an electrical heater, as well as a low-temperature oor heating system with
another electrical back-up heater. Mechanical supply and exhaust ventilation
with heat recovery is also used, and to further increase the temperature of the
supply air an electrical heat battery is integrated in the ventilation system [85].
All electricity is purchased by the individual households.

Figure 5.18: Sketch of the residential area Anneberg; the three sub-areas with varying
roof tilts and solar collector areas. The location of the borehole storage, as well as the
main heat distribution and control building are also seen.

5.6.1 Socio-technical evaluation of Anneberg

The initial technical evaluation for operation during 2003 and 2004 is found
in Paper VII and shows that the system works as intended, charging and dis-
charging the borehole storage, although with lower efciency than under de-
sign conditions11 . The focus here is on the socio-technical evaluation, con-
necting the technical results to interviews also presented in [26] and [78],
i.e. the informants perception of the heating system performance. The tech-
nical and sociological evaluation did not always agree and the results show
the importance of investigating both the technical performance and the users
experiences of a solar heating pilot plant.

First of all, the project has faced several problems along the way, both explod-
ing pipes in the collector loop and leakages in the borehole storage (described
both in Paper VII and [86]). At the time of the interviews in [78] the leakage
in the storage had not yet been discovered, but the exploding pipe in the so-
lar collector loop, and the water leakage it caused in one of the sub-stations,

11 Which is also expected since the studied years are the rst two years of operation.

65
 were recurring issues during the discussions. Several residents communicated
a very dramatic picture of the incident [79], although the perceptions and de-
scriptions varied noticeable between different informants. The damages were
relatively small and the problem was handled professionally. Still, the inci-
dent may have caused both distrust in the heating system and in the property
developer [79].

Problems that occur when facing new technology (solar collectors, oor heat-
ing, technical rooms) is also highlighted in [79]. The informants describe
their inability to affect the heating system; the oor-heating system is per-
ceived as reacting slowly, if at all reacting, and the so called technical rooms
are for experts and not for laymen. A malfunctioning heating system is often
claimed. There have been problems in individual solar collector sub-systems,
see Paper VII, but the experienced malfunction is rather connected to a feeling
of powerlessness to control ones own heating system.

Furthermore, the low contribution of free energy made many informants focus
on the electricity bills [79]. At the time of the study electricity was still the
main heat source, see Figure 5.19. The annual electricity demand varied be-
tween 6 000 kWh and 19 000 kWh per apartment12 , corresponding well to
other contemporary detached and semi-detached houses [87]. The two apart-
ments measuring the lowest demand were more or less uninhabited during
the period and are assumed to represent the heating of the buildings. As com-
municated in the interviews most residents expected a reduced electricity bill
already after one year of operation, and most informants found the electricity
bills too high. They often blame malfunctioning technology (the heating sys-
tem) [79] and although being aware of the correlation, they do not consider
changing their own activities as a means to reduce the energy demand.

The high electricity costs have further caused claims on nancial compensa-
tion from the property developer; the residents do not consider the heating sys-
tem fullling the expectations. The information given to the households has
not been investigated, but the mismatch between the residents expectations
and well-known operational conditions of the system, show that the commu-
nication to the residents have been insufcient. By clearer information, both
the dispute with the property developer and the general disappointment in the
system could perhaps have been avoided.

Although the economic issues engage the informants, the lack of functionality
and comfort causes the largest disappointments, according to Lundh and Lf-
strm [79]. In spite of the residents overall positive attitude to solar heating

12 Total electricity use is measured on a monthly basis in all residential units, while household
electricity and electricity for heating is separated in a few apartments.

66
 25 Electricity
Solar
Measured
20

Energy (MWh/year)
15

10

0
B1 B2 C E
House type

Figure 5.19: The designed solar contribution and electricity demand for heating, as
well as the measured electricity use in 2003, for the different building types in Anne-
berg. B1 corresponds to the smaller terraced houses and B2 to the larger. C represents
the large semi-detached houses and E the smaller.

in general and their unique heating system in particular, it is concluded that

clear information is essential, both regarding the expectations, operation and
what the residents can and can not inuence and control themselves. Conclu-
sively, one important lesson to be learnt is that more effort has to be spent
on informing, but also incorporating, the house owners to a larger extent. The
households, property developer and others involved in the projects have dif-
ferent opinions of to what degree this has been done, but the disappointment
among the residents is a clear measure of the inadequacy. Lundh and Lf-
strm conclude that the heating system may be considered a technical success
to demonstrate new technology, but also that the dissatisfaction of the resi-
dents may be regarded a failure; it all comes down to the methods of measure.

The system principle has been shown to work and at steady state the solar
heating system in Anneberg should supply more than two thirds of the total
energy for heating. The remaining third will be supplied by electric heaters in
the individual households. This is similar to the amount of electricity required
to operate a large-scale heat pump. Many factors have to be taken into account
when choosing a heating system, such as the simplicity of electrical heaters or
district heating, but also the sustainability of the system. Would it then have
been more convenient to use the well-developed heat pump technique instead
of a pilot plant with renewable energy? Maybe, but there are important lessons
to be learnt from pilot projects, and irrespective of system simulations, plants
have to be realised and demonstrated to show the actual technical performance
as well as how technology should and should not be introduced to residents
daily life.

67
5.7 Heating system assessment

Valuation of different heating systems is not trivial and there is no unam-

biguous or completely uncontroversial method to be used. And the outcome
of such an assessment highly depends on the method applied as well as the
system boundaries. Choosing different time perspectives or geographic re-
strictions could, and would, yield other results. In this thesis two aspects are
considered particularly important in an assessment and that is the electricity
production and the use of natural assets.

5.7.1 Electricity production

Carbon dioxide emissions are often used as a measure of sustainability of

power production. Predicting emissions from future production plants is, how-
ever, even more difcult than estimating emissions from present plants. The
energy system is dynamic and no one knows what will constitute future en-
ergy sources. And even if we knew, how should the effect of a certain elec-
tricity consuming activity be assessed as the average power production, the
marginal power or the dirtiest plant in operation? And where should the sys-
tem boundaries be drawn around Sweden, the Nordic countries, Europe or
the world as a whole? Different methods for electricity assessment and the
complexity of valuing power production are discussed in [88], while [89]
shows how the results are affected by the energy system assumptions and
boundary conditions.

Several different approaches for assessing effects of actions within the energy
system are presented in [90]. Two frequently used approaches are the marginal
electricity and the average electricity perspectives. They are used similarly,
estimating the carbon dioxide emissions from power production, but the latter
yields considerably lower CO2 emissions from a Swedish perspective.

Marginal electricity is dened as the electricity from the particular power plant
that increases or decreases the production as the demand changes. Since it will
be the (momentary) last power plant to be taken into operation, it will also be
the one with the highest variable costs. In the public discussion marginal elec-
tricity is often equalled electricity from coal-red condensing plants in Den-
mark, Germany or Poland. But it is not that simple. Although electricity pro-
duction in coal-red condensing plants is expensive, the source of marginal
electricity differs in time. And the import is limited by the transfer capac-
ity, as discussed in [91]. Still, marginal electricity from coal-red condensing
plants is a reasonable assumption in the short perspective [90]. In the long-

68
term perspective it is often assumed to turn towards natural gas combined cy-
cle (NGCC) plants, but the marginal electricity may as well come from other
plants, even renewable sources. Additionally, changes in the energy system
will result in changing conditions for new production plants, which in turn
affect the design of the future energy system.

5.7.2 Natural assets

Biomass and biofuels are sometimes presented as the solution to the energy
supply issues, and in particular to sustainable energy supply. Sweden is
a country with vast supply of biomass and so far mainly residues and
by-products from sawmills, pulp mills and forestry have been used for
heating [92]. The carbon dioxide emissions from biomass are not easily
established, but are often considered low. The carbon dioxide neutrality
that is sometimes stressed is, however, only valid with very narrow system
boundaries. Even with reforesting, energy, which in turn causes CO2
emissions, is required for the harvesting, transport, processing and so on. The
importance of reducing not only the CO2 emissions from fossil fuels, but also
from biomass, is discussed by Grnkvist et al. [93].

A further question is that of reasonable and efcient use of biomass, which is

raised for example by Lindfeldt [94] . At present, but also in the near-future
perspective, the supply of biomass is often expressed as unlimited, but will
it be in the long-term perspective? Several different market areas show in-
terest not only in the available biomass itself, but they also compete for the
same ground (pulp and paper, transportation fuels, electricity and heat pro-
duction, food production and so forth). Increased interest and demand for
biomass will both result in changing consumption and higher costs. Some
applications will be suppressed by the development, as other areas are willing
to pay more. Increased competition for biomass may for example reduce the
possibility of exchanging electricity from coal-red condensing plants. Lind-
feldt and Westermark concludes that efcient use, in the context of biofuels
for transports, is hardly discussed in Swedish Government Ofcial Reports or
government bills. On the contrary, only positive effects are stressed in those
documents. [95]

Small-scale vs. large-scale

If biomass is to be used for heating, and we consider it a limited natural asset,

how should it be used to maximize the benet? Small pellet boilers are often
addressed as a sustainable heating system for single-family houses. They are

69
 considered particularly suitable outside the urban district heating nets, while
the emission of dust in urban areas is still controversial. On the other hand, a
large-scale solution, where the biomass is used in combined heat and power
(CHP) plants may be argued as an even more sustainable solution by some.
The idea is that one unit biomass result in roughly one unit heat in the small-
scale solution. In the CHP plant, on the other hand, one unit biomass result
in roughly two thirds heat, which is supplied to buildings within the district
heating net, but additionally it yields one third electricity13 . This one third
electricity can further be used in contemporary heat pumps to generate one
unit heat in single-family houses in the countryside. Thereby, the same single
unit of biomass would yield almost the double effect; direct heating in the
urban area and indirect heating via a heat pump in the countryside. It could
then be claimed that the benets from using the biomass in a large-scale plant
are larger than when used in small-scale pellet boilers. The issue has not been
investigated within this thesis work, nor is any solution claimed to be better
than the other. The aim is rather to raise the question and stress the importance
of the discussion in heating system assessment.

5.7.3 Assessment from a marginal electricity perspective

Exhaust air heat pumps are more or less standard in contemporary

pre-fabricated houses and the total number of small-scale heat pumps
installed in Sweden is high. There are possible alternatives, but those are not
offered to the customers at present. In this context the potential for combined
solar and wood-pellet systems was investigated in Paper VIII. The electricity
demand connected to the different heating systems were estimated in the
building simulation tool VIP+ 4.014 and the model for assessment was the
marginal electricity approach according to the Swedish Energy Agency
in [97]15 . Costs were not considered in this analysis.

The building model was dened according to specications from a

Swedish manufacturer to represent an average contemporary pre-fabricated

13 Neither the efciency of a small-scale pellet boiler nor heat production in a CHP plant is
100 percent, but this rough estimation could be justied to present the idea in a straightforward
manner. Distribution losses will also occur in both the district heating net (about 14 percent [1])
and the electricity net (about 6 percent [92]).
14 Several different commercial building simulation programs, such as BV2, IDA, ESP-r, Enorm,

all compiled and compared in [96], were considered for the study. VIP+, described in Sec-
tion 5.3.1, was chosen since it included both a solar heating model and a heat pump model.
15 This report was later withdrawn and the approach somewhat modied by the Swedish Energy

Agency.

70
 single-family house; a one-storey building with 143 m2 living area16 . As
a reference case, the building was heated by electric resistance heating.
Calculations were then performed with a combined solar and pellet heating
system and a heat pump system, respectively. The heat pump was assumed
to operate with a constant Coefcient of Performance (COP) over the year,
more or less representing a ground source heat pump or exhaust air heat
pump. Since no boiler model is available in VIP+ the electricity demanded
by a conventional pellet boiler was taken from [99]. Stockholm was assumed
to represent an average location and the same climatic data was used in all
simulations.

The simulation results showed a total annual energy demand for space heating
and hot water of 14 MWh, where 4.7 MWh is for hot water. The electricity
demand associated with the three different heating systems are found in Fig-
ure 5.20(a). It is considerably reduced by a heat pump system and even more
with a solar and pellet system in this particular case. This would give a sig-
nicant impact on the power demand, considering all presently installed heat
pumps.

15.0 2.0
Marginal coalfired
Electricity for heating (MWh/year)

Maximum CO2 emission (Mton)

condensing plant
Marginal natural gas
1.5 combined cycle plant
10.0

1.0

5.0
0.5

0 0
Electric resistance Heat pump Solarpellet Heat pump Solarpellet

(a) Electricity for heating (b) Maximum CO2 emissions

Figure 5.20: Electricity use associated with the three heating systems under investiga-
tion, as well as the CO2 emissions during the hour of highest power demand according
to the applied approach.

It is of further interest to examine the peak power demand, since import of

electricity from coal-red condensing plants is most likely during the most
demanding hours. Figure 5.20(b) shows the CO2 emissions associated with the
most demanding hour of the year for 600 000 households, corresponding to the
number of Swedish single-family houses heated with heat pumps. The peak
power demand, and thereby the CO2 emissions, associated with a combined

16 The specications for the building are the same as in the Appendix of Paper II with the ex-
ception of the solar collector parameters. In this study parameters for a collector from Arcon
Solvarme A/S in Denmark were used [98].

71
solar and pellet heating system is considerably lower than for a heat pump
system.

However, it is not likely that the households that already converted to heat
pumps will change heating systems to solar and pellet in the near future. Fur-
thermore, the analysis only consider one type of building in one climate, op-
timally oriented for solar energy utilization, and generalize from that. An im-
proved investigation would estimate the distribution of convertible houses in
the existing built environment, for example considering the remaining single-
family houses heated by oil boilers or water-borne electric heating. And when
it comes to the CO2 emissions, it has to be considered that the source of
marginal electricity will vary during the lifetime of a building. A thorough
energy assessment could for example be performed according to Figure 5.21.

Perspective Electricity Biomass

- Marginal coal-fired
condensing plants No big issue, surplus
At present
- Average power
production

- Marginal natural gas

Limitation, everybody
Future combined cycle plants
wants it
- Scenarios

Figure 5.21: Proposed focus in a heating system assessment where both electricity
production and biomass supply are discussed in the short and long perspective.

According to the results in Paper VIII, combined solar and wood-pellet

systems could be a suitable alternative, both in pre-fabricated single-family
houses and when converting the heating systems in existing buildings. But
the environmental issues, discussed in this section, may not be decisive
for a customer in the purchase of a heating system. Other aspects, such as
economy, comfort and convenience, constraints, such as construction type
and previous system, as well as recommendations and experiences from
neighbours and friends, most likely inuence the choice as well.

Even though the assessment in Paper VIII is simplied in several ways and
could be further extended, it serves the purpose of widen both the perspective
and the discussion and make small-scale heating systems a part of the larger
energy system.

72
6. Discussion and conclusions

What is the most important conclusion to be drawn from this PhD project? Did
I solve the mystery of the modest dissemination and implementation of solar
heating? Well... no. But I managed to partly widen the system boundaries and
method basis in the study of solar heating, which seems to be a prerequisite
to understand the complexity of the dissemination process and identify barri-
ers that tend to appear in the different implementation phases. By the wider
system perspective, including market actors as system components, the tech-
nical methodology has been widened and from the traditionally very technical
focus in solar heating research, grew a study of technology in a social context.

6.1 Project result overview

A set of obstacles that have to be overcome for solar heating to reach suc-
cess have been identied from the different sub-projects. Other obstacles may
appear, but the ones identied within this thesis work are highlighted and sum-
marized in Table 6.1 and are briey discussed in relation to each other below.

Low fractional energy savings

One of the most important obstacles to be dealt with is the low fractional
energy savings achieved with common solar heating systems. There is a mis-
match between high solar irradiation and high heating demand, but it is also
possible to improve the use of solar heating even in countries like Sweden.
This can be done for example by development of appropriate heat stores
and auxiliary heating systems, but also by improved building constructions to
lower the total demand for space heating. In this study three different projects
show possibilities to overcome the issue of low fractional energy savings. The
rst (Paper VII) is a large-scale pilot plant where solar energy is stored in
a seasonal borehole storage to reach as high as 70 percent fractional energy
savings. The project has met several problems along the way, but those were
more due to organisational issues than technological failure. It can therefore
be concluded that, as long as the inherited knowledge from a demonstration
plant is transferred to future projects, it must be regarded a success.

73
Table 6.1: The main obstacles identied in the papers in this thesis, as well as the dissemination phases where they occur, the main actors affected by
it and the solutions proposed within this thesis work.
Obstacle Dissemination stage Main actor(s) Solution(s) Paper
Low fractional energy Purchase phase Customer 1. Large-scale seasonal heat storages VII
savings Research and development Researcher 2. Medium-sized heat stores for single-family VI
(academia/company) houses
3. Adapted auxiliary heating systems with small V
negative environmental impact (such as CO2 )
Space constraints for Planning, installation Constructor, Architect 1. Flexible geometry of storage tanks VI
installations Purchase phase Customer 2. Preparation for boiler rooms in new dwellings V
Investment cost Research and development Researcher Cheap components produced in industrialized I
Manufacturing Manufacturer processes and large quantities

74
Insufcient planning Planning phase Project planners, Architect, Easily handled and correct simulation tools for so- II
tools Constructors lar heating in different types of buildings
Neglected user Research and development Researcher Include realistic load proles in simulations III, IV
behaviour Planning Installer, Constructor
Engineering focus Implementation phase Installer, User, Trade, Apply a socio-technical systems perspective to V
Government consider both market actors and technology
Sub-optimization of sys- Purchase phase Installer, Customer See solar heating in the larger energy system to VIII
tems enable resource optimization
Complicated systems Installation, operation, Installer, user Standardized systems that are protable and easily V
purchase installed and operated
Un-attractive system Implementation phase User Demonstration projects and good examples VII
designs Planning Architect
Since seasonal storages for small systems are not economically feasible,
other solutions have to be found for individual buildings. The second
solution proposed is to increase the size of the heat store to enable storage
for up to two weeks (Paper VI). The fractional energy savings can then
be increased from about 20-25 percent for an ordinary combi-system to
about 45-50 percent, which is a considerable improvement. Conclusively,
low auxiliary heat demand can be achieved to a reasonable cost even
for single-family houses. Additionally, the auxiliary heat source plays
an important role. The third solution proposed is therefore to develop
auxiliary heating systems that are well-adapted to solar thermal, but also
that cause low environmental impact, such as CO2 emissions, to match the
sustainability mark of solar heating. One possibility is to combine solar
with wood-pellet heating (Paper V). It is, however, concluded that not only
technical performance has to be the focus of development, but also design
and user convenience, to reach wide implementation.

Space constraints for installations

Another obstacle appears when the customer decides to invest in systems with
large heat stores; there is no available space, or if there is, it is limited. Most
contemporary Swedish single-family houses are built without a boiler room,
since most heating systems are based on electricity. The constructors and
architects lack of interest in these problems is considered an important draw-
back by the solar and pellet industries (Paper V). On the other side, in build-
ings with boiler rooms, the ceiling height often constitute a limitation. This
means that the optimal dimensions have to be set aside in favour of practical
ones for large stores. The inuence of the store dimensions has been a hot dis-
cussion topic within the solar industry. In case of a large store for single-family
houses with a ceiling height of about two metres, the reduction in fractional
energy savings due to non-ideal dimensions is, however, relatively small (Pa-
per VI). Conclusively, medium-sized stores can be recommended and installed
in a larger fraction of the existing single-family houses.

Investment cost

The relatively high investment cost is often mentioned by house owners as an

argument against solar heating. They claim it is too high in relation to what
they expect to get from the system, i.e. the pay-back time is too long. Even
though the cost of a complete system solution, comprising solar collectors, a
wood-pellet boiler and an electrical back-up heater, is similar to that of the
popular ground source heat pump, customers do not buy it. In order to reduce
the cost the components can be made more efcient and the manufacturing
be industrialized. There is, for example, a potential in the materials used for
solar absorbers (Paper I), but an improvement would not cause any consider-

75
able, but more likely a rather negligible, effect on the dissemination of solar
heating. Even the interviewed professionals point out that the customer has lit-
tle interest in an additional kilowatt hour. The manufacturing, on the other
hand, is at present a partly manual process. To conclude, standardized systems
produced in large quantities have greater potential than component improve-
ments in reducing the initial cost for the nal customer, and thereby increase
the interest in the systems.

Insufcient planning tools

There are several barriers to be overcome before actually considering invest-

ment in a solar heating system. When reaching this stage, it is, however, essen-
tial that the planners of new residential buildings, as well as retrotting, have
access to reliable tools for estimating the solar energy contribution in different
types of buildings. Still, the present market range of simulation tools allow
either detailed building simulations or detailed simulations of solar heating
systems. One of the few commercial Swedish tools enabling simulations of
solar heating systems in buildings, widely used among consultants, was there-
fore evaluated (Paper II). Serious lacks in the model description were found,
resulting in underestimations of solar contribution, and thereby too low frac-
tional energy savings. Reporting the problems to the programmers, however,
lead to improved model descriptions. It can still be concluded that if this type
of misleading models are used in the planning phase solar heating will most
likely not be installed.

Neglected user behaviour

In general, the energy use in a household is described by very simple load pro-
les in technical simulations. The inuence of user behaviour on the building
or heating system is thereby neglected. Households making use of their own
solar thermal systems, however, conrm that they experience a high inuence
from their hot water use pattern on the auxiliary heat demand. To enable in-
clusion of more realistic hot water load proles a model was developed to
translate time-use data into energy use (Paper III and IV). The model can be
further improved to generate long-term load proles, but the present version
constitutes an important improvement compared to the simplied proles nor-
mally used. Additionally, it shows how methods and data from social sciences
can be incorporated in a technical method (simulations) to achieve realistic
dynamics between user and technology. Conclusively, data from social sci-
ences can, in practice, be used to improve technical methodology and thereby
studies of technology in a social context.

76
Engineering focus

A further problem, which was pointed out in interviews with solar and pellet
professionals, is the lack of system perspective (Paper V). The manufacturers
tend to focus on the technological improvement of the systems, and not on at-
tracting the potential user or involve the installers to promote the technology.
The customer, on the other hand, tend to choose heating systems with low in-
vestment costs, even when the operation cost is rather high. Thus, technology
alone is not decisive for whether a particular heating system is chosen or not.
It is therefore concluded that different actor groups possess a strong power to
inuence the dissemination and implementation of technology positively, but
also to prevent it.

Sub-optimization of systems

Apart from the cost, the environmental impact can not be fully taken into ac-
count until the heating systems are seen in a larger context of the national
or international energy system. Both the electricity supply and resource avail-
ability have to be considered to evaluate different heating systems (Paper VIII)
and prepare the customer with more complete basic data for a purchase. For
instance, a solution that is appropriate in the countryside may not be optimal
for urban applications due to available district heating nets and local combined
heat and power plants. One particular system solution can not be pointed out
as the universal solution, not even solar thermal, but the context and surround-
ing energy system has to be considered as well.

Complicated systems

The installer corps has, by the trade, been identied as a bottleneck in the
dissemination of solar thermal. One reason for their unwillingness to install
solar heating is that they nd the systems complicated, but also that other
installations are more protable (Paper V). A standardization of the systems
would yield compatible components that are easily connected. The fewer de-
tails open to the installer, as well as the user, the easier to handle, and thereby
the more attractive solution. The installer would thereby need to spend less
time on the installation and hence make more money. The user, on the other
hand, would not perceive the system as complicated and would easily be able
to operate it. Apparently, by simpler system solutions the interest and thereby
demand would most likely increase.

Un-attractive designs

With some exceptions, such as reference [100], both house owners and ar-
chitects often lay stress on the un-attractiveness of solar thermal systems as

77
an argument against installation. They often have the traditional solution with
boxes on the roof in mind, where the solar collectors are placed on top
of the roof. They are often unaware of the potential or possibilities with so-
lar collectors; different colours, roof-integration and different architectural
solutions. Well thought out and well-functioning examples and demonstration
projects are therefore highly important to show the aesthetical possibilities.
There are several examples in Sweden, such as the low-energy buildings in
Linds park [70, 101], but also the residential area Anneberg, outside Stock-
holm (Paper VII). The latter project shows how established technology may
be used in a unique system solution, and concludes that attractive, modern
residential buildings with solar heating can be built for ordinary people.

6.2 Overall conclusion

After the overview of all the included sub-projects an overall conclusion can
be drawn; the obstacles should not be seen as isolated issues that can be tack-
led one by one. Instead, they all inuence, interact and support, as well as
suppress, each other. There is not one universal solution to reach increased
use of solar heating, but various actions must be taken at different system
levels and within different actor groups.

In this thesis work, the low fractional energy savings of contemporary solar
heating systems have been identied as a key obstacle. Efforts were made to
identify means to increase the fractional energy savings, such as seasonal stor-
ages and larger stores for single-family houses. The need of sufcient plan-
ning tools that do not underestimate the potential energy savings have also
been pointed out. Increased fractional energy savings would most likely make
the customer willing to pay more for the systems, which in turn could re-
duce the cost due to larger production quantities. A higher demand from the
customers could also encourage the manufacturer to further develop easily in-
stalled and operated systems, as well as adjusted combined systems, such as
the combined solar and pellet systems discussed in this thesis. The installa-
tions could thereby be simplied and the installer make a larger prot. In turn
he would probably start to promote the technology towards the customer, who
would no longer have problems operating the system. The system suppliers
would experience an ever increasing demand and thereby a stronger incentive
to industrialize the manufacturing, which could reduce the costs further. The
implementation and development would start spiralling and the ball would be
set rolling.

78
7. Future work and outlook

Methods traditionally used within other disciplines could promote the devel-
opment of the technical methodology and improve the advance of techno-
logical systems. Both interviews and time-use data, which were used in this
thesis, could be integrated in technical research to a larger extent. The load
model could for example be improved by studying periodic and seasonal vari-
ations in hot water demand to generate long-term load proles. Interviews and
participating observations can also be used in the study and development of
technical system solutions that are convenient to both installer and user. By tri-
angulation, technical results may also be either conrmed or rejected, because
success does not only come down to technically optimized solutions.

The installer corps has, by the professionals in the industry, been identied as
a key bottleneck in the implementation of solar heating. The informants have
also pointed out a possible disparity between different parts of the country. It
would be highly valuable to interview several installers from different parts of
Sweden, but also active within different elds, to thoroughly investigate the
difference in attitude and try to identify the reasons for the inertia within the
installer corps.

Furthermore, integrated, compatible and standardized systems seem to be a

prerequisite to get both installers and customers interested in solar thermal
systems. Focus on these issues in research and development are therefore seen
as important means to speed up the implementation rate.

One of the key issues identied, but not being raised within this thesis, is an
international comparison. Some countries, such as Germany and Switzerland,
are regarded very successful in the implementation of solar energy technology.
It would be of high interest to compare the present and historical conditions in
those countries with the Swedish situation. What can we learn from them and
how did they reach where they are today? But there are also countries, such
as Spain, where solar collector installations are mandatory in new buildings.
How do the different market actors react to such legislations? What bottle-
necks appear on such markets; are there enough installers to carry through all
the installations and is the knowledge and supply of equipment sufcient?

79
8. Final remarks and self-evaluation

It is not trivial to apply a socio-technical perspective to a research topic. The

rst struggle is the complexity of the system; what inuences what and where
should the system boundaries be drawn [11]. Another question that can not be
neglected is how this type of study should be performed to generate relevant
knowledge. It has to be carefully considered what makes a socio-technical
system study more suitable than a pure disciplinary study.

Being a physicist and writing a thesis on socio-technical solar heating systems

requires self-evaluation. I have to ask myself what I achieved by this approach
compared to an intradisciplinary thesis. When I rst started my PhD studies
my aim was to develop absorber materials for thermal solar absorbers, which
would have been more or less an intradisciplinary project. I would then, by
now, probably have a very deep and focused thesis on sputtered nickel-nickel
nitrides. That kind of study is highly important and could supply interest-
ing knowledge about solar heating on a micro-scale system level. The system
boundaries must, however, be very narrow to allow precise results. The nal
conclusion in that thesis would probably have been how my coatings stand up
against existing commercial ones. In the best case it would have been possi-
ble to tell how this inuences the efciency of the solar collector or even the
fractional energy savings on an annual basis.

In this thesis work, on the other hand, I have gradually moved towards higher
system levels to get a more holistic view of the complex system of solar heat-
ing. The questions are naturally not the same in the two types of studies, but
moving the system boundaries and extend the system view opens up for new
research questions that can not be formulated when focusing solely on tech-
nology. To be able to answer the questions about improvement of technical
studies by bringing in the social context, as well as optimization of solar heat-
ing, stated in the rst chapter, the deep and very technological focus has to
be left behind and a broader approach must be applied. To tackle the ques-
tions appropriate methods should be used and the main advantage of this
kind of interdisciplinary approach is the possibility of incorporating meth-
ods from other disciplines into technological studies. Without those methods,
it would not have been possible to perform this type of study. My background
in science and technology has, however, inuenced the formulation of ques-

81
tions, the way in which the methods have been used, the reporting, but also the
projects themselves. The communication with and critical opinions from my
co-workers in the interdisciplinary forums that I had the opportunity to take
part of have been invaluable for my learning process and the nal outcome of
my studies.

I believe that the sub-projects in this thesis are inherently useful and mean-
ingful in the ambition towards increased use of solar heating. The thesis as a
whole as well, but the main product from this type of thesis work is probably
the interdisciplinary doctor that comes out of it...

82
Acknowledgements

The most important person for making this thesis possible is my supervisor
Professor Ewa Wckelgrd. I guess it is time to admit that I did not at all under-
stand the word sputtering that rst time we discussed my project plan during
the Sweden tour 2003. But I am happy you let me extend and change the plan
quite a bit; I am really grateful to your openness towards interdisciplinary
work. As long as you were not worried, I did not worry about the coherence
of my work. And if you were worried; I am happy you did not show it.

I have had several assistant supervisors during these years. Arne Roos sup-
ported me during optical as well as English troubles. I would further like to
thank Jan-Olof Dalenbck for letting me take part of the evaluation of the solar
heating system in Anneberg and Bengt Perers who gladly provided informa-
tion about simulations in Winsun. Tomas Kberger has been a great source
of inspiration and made me reect on my work by all the why questions.
Further inspiration was provided by Annette Henning your endless positive
attitude to our interdisciplinary project! Kajsa Ellegrd is acknowledged for
letting me use the time-use data in my studies of solar heating. Finally, Elisa-
beth Kjellsson is acknowledged for being an excellent discussion leader at my
licentiate seminar!

I would like to thank Claes-Gran Granqvist for including me in the division

of Solid State Physics, but also all present and former co-workers for the open
and positive atmosphere. I would especially like to thank those participating in
the coffee and lunch breaks for making hard days a bit easier! Some persons,
though, have to be mentioned by name. Dr Martin, it means a lot that you
took the time to read and comment my licentiate thesis! Mari-Louise you
were my endless support from day one! Annica what would I do without
you?! Sara your openness, Pia your spontaneity, Andreas J my fellow
traveller, but also Andreas R, Herman, Esteban, Anna, Tobias, Roser, Jonas,
Joakim, Peter W, Magnus ...

I would also like to thank all former and present persons within the Energy
Systems Programme, especially all the D03s for hot discussions, but also for
pleasant social events! A special thank to Erica Lfstrm and Andreas Jonsson

83
for the project in Anneberg hard work, but foremost a good time! Erik L has
been a great support during those years I admire your honesty and courage!

I am very grateful to Professor Klaus Vajen and Junior Professor Ulrike Jor-
dan at the department of Solar- und Anlagentechnik at Universitt Kassel for
letting me be part of their group for half a year. It was a great experience
and all members of the SAT group are acknowledged for being so helpful and
friendly! A special thank to Katrin Za and Claudius Wilhelms for the project
collaboration. Ralf, Bastian and Corry: thanks for all the social adventures it
means so much to me! Philipp: Danke fr deine Deutlichkeit und Geduld mit
meinem Deutsch...

Thanks to Gift Katumba who patiently instructed me on how to apply absorber

paints and Tobias Blom for spending hours helping me making nice SEM im-
ages of the painted layers. Shuxi Zhau was a great support when the optical
measurements caused me troubles, while Micke P has provided me with ex-
cellent LATEX support. Thanks also to Joakim Widn for the collaboration on
load proles and Iana Vassileva for supplying hot water data.

For nancial support I want to thank The Swedish Energy Agency for nanc-
ing this PhD project, but also Anna Maria Lundins stipendiefond for gladly
paying for most of my conferences. I also received funding from Sederholms
stipendiefond for my rst stay in Kassel, while Wallenbergs resestipendium
supported me with parts of a conference trip. Markussens stipendiefond is ac-
knowledged for their private nancial support.

I doubted many times, but thanks to wonderful fellow workers and stimulating
tasks I managed to handle the sometimes struggling periods. But it demanded
patience, both from all the friends I did not have enough time for, but fore-
most from my family... Thanks Pappa for nding the advertisement about this
position, but also for your love and support. Mamma, thank you for always
being there and having the time to listen when I feel confused. The best big
sister ever, Misse if it was not for you and your engagement in environmental
issues... Robin, Anton and Isabella thank you for being my family! Familjen
Boork is the best second family. Och Mormor. Fr din ndlsa uppmuntran
och omtanke och att du tycker att jag r duktig vad jag n gr. But there is one
person that had to deal with me every day: Pr. How did you manage? I was
not always happy, I was not always easy to be around, but you were always
there. That means everything to me.

84
APPENDIX A: Interview guide

Interview questions to investigate the development of combined

solar and wood-pellet heating systems in Sweden

Pre-questions

- What does your company/organisation do?

- What is your professional role/position in the company?
- What is the size of the company? What is your annual turnover?
- For how long have you been active within this industry?

Questions

1. What do you think about solar heating systems? Pellet heating systems?
a. What do you think about combined solar and pellet heating?
2. Why do you think solar heating and pellet should be combined? (if positive re-
sponse to question 1a)
a. What are the possibilities with combined systems with solar and
pellet (in a wider perspective)?
b. Are there conditions for such combined systems?
c. What do you think about the present governmental subsidies?
Previous subsidies?
d. Are there any drawbacks with combined systems?
e. Do you consider combined solar and pellet systems prioritised? Who
invest in this development?
f. Why is this combination prioritised (and not other technical solu-
tions)? (if positive response to question 2e)
g. What do you consider a good alternative technical solution?
3. If we go back in time, how do you remember the development of combined solar
and pellet systems? How did the combination evolve?
a. Is there any development of combined solar and pellet systems at present?
What? Why (not)?
b. Do the industries show interest in these systems/development of systems?

85
 c. Is any industry more active/inuential than the other in the development?
Why?
4. What are the prerequisites for the technological development of combined systems
to get going/continue?
a. What is demanded for combined solar and pellet systems to reach the
Swedish market to a larger extent?
b. What are the obstacles for solar and pellet systems? What driving forces
are missing?
c. Is there enough knowledge about solar and pellet heating systems? Who
has knowledge about the systems? Who needs increased knowledge about the
systems?
d. How available are solar and pellet systems to the customers?
e. Are the customers interested? Are there purchasers?
f. Who advertise combined solar and pellet systems?
g. Would you/your company be willing to pay for testing combined systems
to get them P labelled?
5. What does the procedure look like when customers want to buy combined systems?
Who do they contact?
6. What are your thoughts about the solar industry/the pellet industry?
a. How do you experience the general attitude between the solar and pellet
industries?
b. How does the collaboration between the solar and pellet industries work?
Why do you think the solar companies/pellet companies are interested in
collaboration? Why does the collaboration work like that?
c. What is required to make the solar and pellet industries collaborate more?
d. How are the industries located? How do they meet? What forums are used
to communicate across the trade boarders?
e. Do you/your organisation/company collaborate with anybody? In what
way?
f. What is required for you to collaborate with the solar/pellet industry?
g. How active is your company/organisation in the development of combined
solar and pellet systems?
7. How could the work on combined solar and pellet systems continue?
a. What is required for solar and pellet systems to reach success?
8. Have you noticed any changes in your industry during the time you have been
active?
9. Would you like to add something?

86
APPENDIX B: Word list and abbreviations

absorptance (A, ) The fraction of the incoming radiation that is

absorbed by a medium.
AR Antireection
artefact Man-made object, in contrast to naturally existing
objects.
auxiliary volume The part of the storage tank that is kept at a certain
set temperature to always allow hot water draw-offs.
CHP Combined Heat and Power
CO2 Carbon dioxide
combi-system A solar heating system that supplies both space
heating and hot water.
convection Movements in gases or liquids due to differences in
density within the gas or liquid. Losses through
convection means that heat is transported or lost
when for example air is moving (wind).
COP Coefcient of Performance, a measure of the efciency
of a heat pump, tells how much heat is supplied by the
heat pump compared to the energy supplied to it.
DHW system Domestic Hot Water system, a solar heating system
that only supplies hot water.
electromagnetic radiation Radiation consisting of both electric and magnetic
elds, for example light.
emittance (E, ) Energy radiated from a surface relative to
what is radiated by an ideal black body.
fractional energy savings, The ratio between the auxiliary heat required in the
( fsav,therm ) solar heating system and with conventional heating.
IEA-SHC International Energy Agency-Solar Heating and
Cooling Programme
intrinsic Inherent or natural.
mixed tank All water in the tank has the same average temperature.
reectance (R, ) The ratio between reected and incident light
intensity.
social factors Political, economical, organisational or cultural
factors.

87
socio-technical system A system comprising both social and technical
components.
solar fraction The fraction of the total heat demand covered by
solar energy.
spectral selectivity Different optical properties in different wavelength
intervals.
sputtering A method to deposit a material on a substrate.
stratication The formation of a temperature gradient in a storage
tank due to the difference in density, which depends
on the temperature of the water. The higher the
temperature, the lower the density.
substrate The underlying or base material used for application
of thin lms.
transmittance (T, ) The fraction of the incident light that is let through
a medium.
TRNSYS TRaNsient SYstem Simulation Program, a common
simulation tool for dynamic solar heating system
simulations.
TSSS paint Thickness Sensitive Spectrally Selective (absorber)
paint.
SEM Scanning Electron Microscope
U value Heat transfer coefcient, describing the heat losses
through for example a building element.
zero loss efciency (0 ) The optical efciency of the collector, which
mainly depends on the transmittance of the cover
glass and the absorptance of the absorber.

88
9. Summary in Swedish

Uppvrmning med solvrme i bostder

Studier av teknik i ett socialt sammanhang och sociala komponenter i

tekniska studier

Solenergi kan anvndas fr mnga olika ndaml, ssom belysning,

uppvrmning och elproduktion. Den hr avhandlingen fokuserar p
uppvrmning av bostadshus och varmvatten med hjlp av s kallade termiska
solfngare. I en solfngare omvandlas solstrlningen till nyttig vrme i
absorbatorn, som bestr av en metall belagd med ett hgabsorberande
material. Den alstrade vrmen verfrs sedan till en vtska och frs till
vrmelagret. Vrmen kan antingen anvndas enbart till varmvatten, ett s
kallat tappvarmvattensystem, eller till bde varmvatten och uppvrmning
av byggnader, ett s kallat kombisystem. I Sverige r kombisystem
vanligast, medan tappvarmvattensystem dominerar i vrlden som helhet.
Eftersom perioder med hg solinstrlning oftast inte sammanfaller med
uppvrmningsssongen r lagring av vrme, liksom lmpliga kompletterande
vrmekllor, essentiellt fr solvrme. Bda dessa frgor genomsyrar samtliga
forskningsprojekt i denna avhandling.

Traditionellt r fokus inom solenergiforskning p att utveckla och frbttra de

tekniska komponenterna och p s vis ka effektiviteten hos systemen. Bde
laboratorieexperiment och systemsimuleringar i avancerade datorprogram
r vanliga metoder inom solvrmeforskning idag. I ett strre perspektiv
ska dock tekniken anvndas av mnniskor. Aktrer av olika slag, ssom
arkitekter, installatrer, byggherrar och brukare, har avgrande betydelse
fr om solvrme installeras eller inte, men glms ofta bort i forsknings- och
utvecklingsarbetet. I den hr avhandlingen breddas perspektivet p solvrme,
frn att endast inkludera det tekniska vrmesystemet till att p olika stt
koppla samman teknik och anvndare. Fr att gra detta anvnds traditionella
tekniska metoder i kombination med metoder som vanligtvis inte anvnds
inom teknisk forskning. Solvrme gr frn att vara ett rent tekniskt system
till att ses som ett socio-tekniskt system, bestende av bde sociala och
tekniska komponenter, dr de sociala komponenterna kan vara bde aktrer,
lagstiftning, politik och kulturella aspekter.

89
Tekniken fr solvrme har utvecklats markant sedan marknadsintroduktionen
p 1970-talet. Trots detta r det fortfarande frhllandevis f som installerar
solvrmesystem i Sverige, medan framgngen har varit betydligt strre
i andra europeiska lnder, exempelvis Tyskland, sterrike och Spanien.
Ibland nmns priset som ett hinder. Mjligheterna att frbttra ett billigt,
kommersiellt material fr solabsorbatorer undersks i ett av delprojekten
i denna avhandling. I just detta fall r frbttringen inte tillrcklig fr att
processen ska vara ekonomiskt genomfrbar i stor skala. Men frutom
utveckling av komponenter kan systemutfrandet pverka kostnaden och
drmed gra solvrme mer attraktivt fr konsumenten. Genom att ka den
andel av det totala vrmebehovet som tcks av solvrme r det mjligt
att snka den relativa kostnaden. Samtidigt blir solen den huvudsakliga
komponenten i systemet, istllet fr att som idag utgra ett tillskott p
20-50 procent.

Fr att uppn en hg andel solvrme i ett system mste vrmen kunna lagras
effektivt, idealt frn sommar till vinter. Vrmelager av varierande storlek
och fr olika lng tid har varit en av huvudfrgorna inom detta arbete. Fr
smhus r det inte ekonomiskt genomfrbart med enskild ssongslagring,
eftersom frlusterna blir fr stora. Dremot r det mjligt att ka lagrets
storlek, s att vrmen kan lagras ett par veckor istllet fr korta tidsperioder.
Ett problem med att ka storleken p vrmelagret, som lyfts i denna
avhandling, r de fysiska begrnsningarna i bentliga hus, ssom takhjd
och ppningar till kllare och pannrum. Hr visas dock att det r mjligt
att frng de optimala dimensionerna p ett vrmelager och nd uppn
hg tillfrsel av solvrme. Det innebr att en strre del av smhusen skulle
kunna utrustas med stora solvrmesystem med bde kad solfngararea och
lagringskapacitet. Fr bostadsomrden, andra sidan, r det mjligt att bygga
ssongslager fr att n tckningsgrader fr solvrme p uppt 70-80 procent.
En sdan pilotanlggning nns utanfr Stockholm och dess frsta tid i drift
har utvrderats inom detta avhandlingsarbete. Systemlsningen har visat sig
fungera som planerat, men brist p kommunikation och information under
bde byggnation och initial drift har visat sig medfra svl hga kostnader
som frustration bland de boende.

Uppfrande av svl smskaliga som storskaliga solvrmesystem fregs ofta

av ett omfattande planeringsarbete. Ibland anvnds simuleringsmodeller fr
att uppskatta hur mycket solvrme som kommer att tillfras. Fr att sdana
modeller ska vara till nytta krvs dock att uppskattningarna r realistiska. I
den hr avhandlingen visas ett exempel dr solenergi underskattas markant
p grund av en felaktig modellbeskrivning. Om sdana brister inte uppmrk-
sammas nns risken att solvrme vljs bort av arkitekter och installatrer p
felaktiga grunder.

90
En ytterligare faktor som pverkar andelen anvndbar solvrme r hur
varmvatten anvnds. I simuleringsmodeller r varmvattenanvndningen, och
drmed hushllets beteende, ofta reducerat till en mycket enkel lastprol
med ett ftal tappningar varje dag. Fr att ta hnsyn till dynamiken mellan
anvndare och teknik utvecklades en metod fr att modellera varmvattenan-
vndning i svenska hushll. Den r baserad p tidsanvndningsdata, dr
ett stort antal hushll har frt dagbok ver sina samtliga aktiviteter under
ett par dagars tid. Simuleringar av ett tappvarmvattensystem med olika
varmvattenproler tyder dock p att mngden varmvatten har en strre
inverkan p systemet n nr p dagen det anvnds.

Solenergi ses som en ren och miljvnlig energiklla, men ett komplett
vrmesystem mste ocks ses i ett strre sammanhang; som del av det
regionala, nationella eller internationella energisystemet. Den vanligaste
svenska systemlsningen fr solvrme r kombinationer med el, men
kombinerade sol- och pelletsystem marknadsfrs allt mer intensivt som ett
hllbart alternativ. Problematiken kring elanvndning svl som anvndning
av biobrnsle lyfts i denna avhandling. Fr att utreda hur en ny systemlsning
formas och utvecklas tillfrgades dessutom olika aktrer inom sol- och
pelletbranschen om hur de upplever marknadssituationen fr kombinerade
system. Inte bara tekniken i sig, utan i nnu hgre grad de olika aktrerna p
marknaden, utpekades som avgrande fr ett systems framgng eller fall.

Sammanfattningsvis nns det ett antal hinder mot solvrmeanvndning i

Sverige, men ocks lsningar. Dremot visar era av projekten i denna
avhandling att lsningarna inte bara handlar om frbttrad teknik, utan
snarare att f samspelet mellan teknik och aktrer att fungera. Att lyfta
perspektivet ter sig drfr ndvndigt fr att solvrme ska n framgng. Den
hr avhandlingen utgr ett steg i den riktningen.

91
Bibliography

[1] Energimyndigheten. Energilget 2007, 2007. ET 2007:49. In Swedish.

[2] Statistics Sweden (SCB). Summary of energy statistics for dwellings and non-
residential premises for 2001-2006, 2008. EN 16 SM 0704, Corrected version
2008-04-23, ISSN 1404-5869.

[3] Boverket (The National Board of Housing, Building and Planning). Kon-
sekvensutredning: Revidering av avsnitt 9 i Boverkets byggregler (BFS
1993:57). Available at www.boverket.se, August 2008. In Swedish.

[4] The Swedish Government. Regeringens proposition 2005/06:145, March 2006.

Available at www.regeringen.se. In Swedish.

[5] Conny Ryytty. Utveckling av tgrderna fr minskad elanvndning i bostder

och lokaler (N1999/12982/ESB). Swedish Energy Agency, Dnr: 00-00-423,
August 2000. In Swedish.

[6] SP Technical Research Institute of Sweden. Provning & forskning No 2, 2006.

In Swedish.

[7] Annette Henning. Ambiguous Artefacts: Solar Collectors in Swedish Con-

texts. On Processes of Cultural Modication. PhD thesis, Department of
Social Anthropology, Stockholm University, Sweden, 2000. ISBN 91-7265-
034-6.

[8] Jane Summerton. Den konstruerade vrlden, chapter Stora tekniska system:
En introduktion till forskningsfltet. Brutus stlings Bokfrlag Symposion,
1998. In Swedish.

[9] Bodil Jnsson and Nina Reistad. Experimentell fysik. Studentlitteratur, Lund,
1987. In Swedish.

[10] Lars Ingelstam. System att tnka ver samhlle och teknik. ISBN 91-
89184-24-6. Energimyndigheten, 2002. In Swedish.

93
[11] C.West Churchman. Systemanalys. Rabn & Sjgren, Stockholm, 1973.
ISBN 91 29 41833X. In Swedish. Original title: The Systems Approach, 1968.

[12] Thomas P. Hughes. Networks of Power: electrication in Western society,

1880-1930. John Hopkins University Press, Baltimore, 1983.

[13] Wiebe E. Bijker, Thomas P. Hughes, and Trevor J. Pinch, editors. The social
construction of technological systems: new directions in the sociology
and history of technology. MIT Press, Cambridge, Massachusetts, 1987.

[14] Charles Edquist, editor. Systems of Innovation: Technologies, Institutions

and Organizations. Pinter, London, 1997.

[15] Boel Berner. Perpetuum Mobile? Teknikens utmaningar och historiens

gng. Arkiv frlag, Lund, 1999. ISBN 91 7924 122 0. In Swedish.

[16] Robert K. Yin. Case Study Research: Design and Methods. ISBN
0-7619-2553-8. Thousand Oaks: Sage Publications, Inc., Third edition, 2003.

[17] Sharan B. Merriam. Fallstudien som forskningsmetod. ISBN

91-44-39071-8. Studentlitteratur, 1994. In Swedish. Original title: Case Study
Research in Education.

[18] Jan-Olof Dalenbck. Solar heating with seasonal storage: some aspects
of the design and evaluation of systems with water storage. PhD thesis,
Department of Building Service Engineering, Chalmers University of Technol-
ogy, Gteborg, Sweden, 1993.

[19] S. A. Klein et al. TRNSYS 16 a TRaNsient SYstem Simulation program.

Solar Energy Laboratory, University of Wisconsin-Madison, USA, 2006.

[20] Torsten Hgerstrand. Om tidens vidd och tingens ordning: Texter av

Torsten Hgerstrand. Byggforskningsrdet, 1991. ISBN 91-540-5363-3. In
Swedish.

[21] Kajsa Ellegrd. A time-geographical approach to the study of everyday life of

individuals a challenge of complexity. GeoJournal, 48:167175, 1999.

[22] K. Ellegrd and M. Cooper. Complexity in daily life a 3D-visualization

showing activity patterns in their contexts. electronic Journal of Time Use
Research, 1:3759, 2004.

[23] K. Vrotsou, K. Ellegrd, and M. Cooper. Exploring time diaries using semi-
automated activity pattern extraction. In IATUR XXVIIII Conference, Wash-

94
 ington DC, USA, 2007.

[24] K. Vrotsou, K. Ellegrd, and M. Cooper. Everyday life discoveries: Mining

and visualizing activity patterns in social science diary data. In Proceedings
of Information Visualization 2007, pages 130138, 2007.

[25] Helena Krantz. Matter that matters: A study if household routines in a

process of changing water and sanitation arrangements. PhD thesis, De-
partment of Water and Environmental Studies, Linkping University, Sweden,
2005. ISBN: 91-85297-65-8.

[26] Erica Lfstrm. Visualisera energi i hushll: Avdomesticeringen av so-

ciotekniska system och individ- respektive artefaktbunden energianvnd-
ning. PhD thesis, Department of Technology and Social Change, Linkping
University, Sweden, 2008. In Swedish.

[27] Steinar Kvale. Den kvalitativa forskningsintervjun. Studentlitteratur, Lund,

1997. ISBN: 91-44-00185-1. In Swedish. English title: InterViews, Sage Pub-
lications, Inc., 1996.

[28] E. Wckelgrd, G. A. Niklasson, and C. G. Granqvist. Solar energy the

state of the art: ISES position papers, chapter 3. James & James, London,
2001.

[29] John A. Dufe and William A. Beckman. Solar Engineering of Thermal

Processes. John Wiley & Sons, Inc., New York, 1991. ISBN 0-471-51056-4.

[30] Graham L. Morrison. Solar energy the state of the art: ISES position
papers, chapter 4. James & James, London, 2001.

[31] German Solar Energy Society (DGS). Planning and Installing Solar Ther-
mal Systems: A guide for Installers, Architects and Engineers. James &
James, 2005. ISBN 1-84407-125-1.

[32] F. A. Peuser, K.-H. Remmers, and M. Schnauss. Solar Thermal Systems:

Successful Planning and Construction. Solarpraxis AG and James & James,
Berlin, 2002. ISBN 1-902916-39-5.

[33] C. M. Lampert. Coatings for enhanced photothermal energy collection, I. Se-

lective absorbers. Solar Energy Materials, 1:319341, 1979.

[34] Z.C. Orel. Characterization of high-tempearture-resistant spectrally selective

paints for solar absorbers. Solar Energy Materials and Solar Cells, 57:291
301, 1999.

95
[35] Solec Solar Energy Corporation. www.solec.org/solkotehome.htm, Accessed
2006-04-07.

[36] M. Lundh, T. Blom, and E. Wckelgrd. Anti-reection treatment of TSSS

paints for thermal solar absorbers. In Proceedings of Eurosun 2006, Glas-
gow, Scotland, 2006.

[37] B. Hellstrm, M. Adsten, P. Nostell, and E. Wckelgrd. The impact of optical

and thermal properties on the performance of at plate solar collectors. In
Proceedings of Eurosun 2000, Copenhagen, Denmark, 2000.

[38] Arne Andersson. Delredovisning av erfarenheter frn det statliga bidraget till
investeringar i solvrme (SFS 2000:287), verksamheten 2004. Statens En-
ergimyndighet, February 2005. In Swedish.

[39] Werner Weiss and Gerhard Faninger. Solar thermal collector market in IEA
member countries. International Energy Agency, Solar Heating & Cooling
Programme, December 2002.

[40] Lars Andrn. Solenergi: Praktiska tillmpningar i bebyggelse. AB Svensk

Byggtjnst, Stockholm, 1999. In Swedish.

[41] Werner Weiss, editor. Solar Heating Systems for Houses A design hand-
book for solar combisystems. ISBN 1 902916 46 8. James & James, 2003.

[42] Peter Kovcs and Ulrik Pettersson. Solvrmda kombisystem: En jmfrelse

mellan vakuumrr och plan solfngare genom mtning och simulering. SP
Technical Research Institute of Sweden, Bors, 2002. SP Rapport 2002:20. In
Swedish.

[43] E. Andersen, S. Furbo, M. Hampel, W. Heidemann, and H. Mller-Steinhagen.

Investigations on stratication devices for hot water heat stores. International
Journal of Energy Research, 32:255263, 2008.

[44] Energimyndigheten (Swedish Energy Agency). Lpande rapportering av

erfarenheter frn det statliga bidraget till investeringar i solvrme (SFS
2000:287), October 2008. Dnr: 00-06-6705. In Swedish.

[45] Boverket (National Board of Housing, Building and Planning). Bostadspoli-

tiken svensk politik fr boende, planeringe och byggande under 130 r.
Boverket, Karlskrona, 2007. In Swedish.

[46] Arne Andersson. Lpande rapportering av erfarenheter av det statliga bidraget

till investeringar i solvrme (SFS 2000:287). Nringsdepartmentet, October

96
 2008. Dnr: 00-06-6705. In Swedish.

[47] ESTIF. Solar Thermal Markets in Europe (Trends and Market Statistics 2004),
June 2005. Available at www.estif.org.

[48] ESTIF. Solar Thermal Markets in Europe (Trends and Market Statistics 2005),
June 2006. Available at www.estif.org.

[49] ESTIF. Solar Thermal Markets in Europe (Trends and Market Statistics 2006),
June 2007. Available at www.estif.org.

[50] ESTIF. Solar Thermal Markets in Europe: Trends and Market Statistics 2007,
June 2008. Available at www.estif.org.

[51] Bengt Perers and Chris Bales. A Solar Collector Model for TRNSYS Sim-
ulation and System Testing. Solar energy research center SERC, Hgskolan
Dalarna, December 2002. Report of IEA SHC Task 26.

[52] SPF. Users manual for Polysun 3.3, December 2000.

[53] Meteotest. Meteonorm, version 5.0 edition 2003. www.meteotest.ch.

[54] Sune Hggbom, StruSoft, 2006. Private communication.

[55] Sune Hggbom, StruSoft, 2008. Private communication.

[56] Ulrike Jordan and Klaus Vajen. Inuence of the DHW load prole on the
fractional energy savings: a case study of a solar combi-system with TRNSYS
simulations. Solar Energy, 69:197208, 2000.

[57] R. Spur, D. Fiala, D. Nevrala, and D. Probert. Inuence of the domestic hot-
water daily draw-off prole on the performance of a hot-water store. Applied
Energy, 83:749773, 2006.

[58] Chris Bales and Tomas Persson. External DHW unit for solar combisystems.
Solar Energy, 74:193204, 2003.

[59] . Wahlstrm, R. Nordman, and U. Pettersson. Mtning av kall- och varm-

vatten i tio hushll,. Technical Research Institute of Sweden and the Swedish
Energy Agency, ER 2008:14, ISSN 1403-1892, 2008. In Swedish.

[60] Sren Knudsen. Consumers inuence on the thermal performance of small

SDHW systems theoretical investigations. Solar Energy, 73:3342, 2002.

97
[61] S. Holmberg. Flow rates and power requirements in the design of wa-
ter services. PhD thesis, Department of Heating and Ventilation Technology,
Royal Institute of Technology, Sweden, 1987.

[62] J. Wollerstrand. District heating substations: performance, operation and

design. PhD thesis, Department of Heat and Power Engineering, Lund Institute
of Technology, Sweden, 1997.

[63] Magdalena Lundh, Ewa Wckelgrd, and Kajsa Ellegrd. Design of hot water
user proles for swedish households based on time diaries. In Proceedings of
the Scientic Conference on Energy saving and Green energy at Alvsjo
fair, Stockholm March 12-13, 2008 in connection with the "Energitinget
2008", ISBN: 978-91-977493-2-9, 2008.

[64] Magdalena Lundh, Ewa Wckelgrd, and Kajsa Ellegrd. Design of hot water
user proles for Swedish conditions. In Proceedings of ISES Solar World
Congress 2007, Beijing, China, 2007.

[65] M. Lundh, I. Vassileva, E. Dahlquist, and E. Wckelgrd. Comparison between

hot water measurements and modelled proles for Swedish households. In
Proceedings of Eurosun 2008, Lisbon, Portugal, 2008.

[66] Hans Bagge. Energy use in multi-family dwellings measurements and meth-
ods of analysis. Licentiate thesis, Department of Building Physics, Lund Uni-
versity, Sweden., 2007.

[67] R. Heimrath and M. Haller. Project Report A2 of Subtask A: The Reference

Heating System, the Template Solar System. Solar Heating and Cooling Pro-
gramme, International Energy Agency, 2007.

[68] S. Kuethe, C. Wilhelms, K. Zass, R. Heinzen, K. Vajen, and U. Jordan. Mod-

elling complex systems within TRNSYS SIMULATION STUDIO. In Pro-
ceedings of Eurosun 2008, Lisbon, Portugal, 2008.

[69] U. Jordan and K. Vajen. DHWcalc: program to generate domestic hot water
proles with statistical means for user dened conditions. In Proceedings of
ISES Solar World Congress 2005, Orlando (US), 2005.

[70] Wiktoria Glad. Aktiviteter fr passivhus: En innovations omformning

i byggprocesser fr energisnla bostadshus. PhD thesis, Department of
Technology and Social Change, Linkping University, Sweden, 2006. In
Swedish.

[71] F. Fiedler, S. Nordlander, T. Persson, and C. Bales. Thermal performance of

98
 combined solar and pellet heating systems. Renewable Energy, 31:7388,
2006.

[72] Tomas Persson. Combined solar and pellet heating systems for single-
family houses. PhD thesis, Department of Energy and Environmental Tech-
nology, The Royal Institute of Technology, Stockholm, Sweden, 2006.

[73] M. Lundh, E. Wckelgrd, and A. Henning. Qualitative investigation of the

development of combined solar and pellet heating systems in Sweden. In Pro-
ceedings of Eurosun 2008, Lisbon, Portugal, 2008.

[74] C. Philibert. Barriers to Technology Diffusion: The Case of Solar Thermal

Technologies. OECD/IEA, October 2006.

[75] M. Kenisarin and K. Mahkamov. Solar energy storage using phase change
materials. Renewable and Sustainable Energy Reviews, 11:19131965,
2007.

[76] Jan-Olof Dalenbck. Some views from Sweden related to the workshop "Sim-
ulating growth of solar collective systems in European housing markets"
Barcelona, October 23-24, 2003.

[77] Gunnar Brms. Solvrmeanlggningen i Lyckebo, frslag p tgrder. Tech-

nical Report U 01:93, Vattenfall Utveckling AB, 2001. In Swedish.

[78] Andreas Jonsson, Magdalena Lundh, and Erica Lfstrm. Hushll med
solvrme ett svenskt pilotprojekt i Anneberg. Arbetsnotat Nr 30, ISSN 1403-
8307, Program Energisystem, Linkping, 2005. In Swedish.

[79] Magdalena Lundh and Erica Lfstrm. Large scale pilot project with so-
lar heating for a residential area success and disaster. In Proceedings of
ECEEE 2007 Summer Study, Cte dAzur, France, 2007. ISBN: 978-91-
633-0899-4.

[80] Anders Bernestl and Jenny Nilsson. Brf Anneberg: Kv Vindkraften, Kv

Solvrmen, Kv Solfngaren Utvrdering av energianvndning. Ingenjrs-
byrn Andersson & Hultmark AB, Arb. nr 3805.00, Gteborg 2007-03-06. In
Swedish.

[81] J. Nussbicker, D. Mangold, W. Heidemann, and H. Mller-Steinhagen. Solar

assisted district heating system with seasonal duct heat store in Neckarsulm-
Amorbach (Germany). In Proceedings EuroSun 2004, Freiburg, Germany,
pages 509517, 2004.

99
[82] Lundin S.-E. et al. Solvrme med ssongslagring i borrhl i berg och
lgtemperatur fr bostadsomrdet i Anneberg, Danderyd, Frprojektering
1998-03-31. Statens Energimyndighet, 1998. SOLV-98/1. In Swedish.

[83] Jan-Olof Dalenbck. Anneberg Solar Heating Plant with Seasonal Storage in
Rock. CIT Energy Management AB, 2002. REB 0061/97.

[84] Jan-Olof Dalenbck. Private communication, 2004.

[85] Lundin S.-E. Solvrmesystem och borrhlsvrmelager fr bostadsomrde An-

neberg, Danderyd. Rapport 23340, 2002-12-04, 2002. In Swedish.

[86] Jan-Olof Dalenbck. Utvrdering av solvrmesystem Brf Anneberg. CIT En-

ergy Management AB, March 2006. In Swedish.

[87] Statistics Sweden (SCB) and Swedish Energy Agency (Energimyndigheten).

Energistatistik fr smhus 2003. Statistiska meddelanden, EN 16 SM 0403,
December 2004. In Swedish.

[88] Elforsk. Miljvrdering av el med fokus p utslpp av koldioxid, 2009. Avail-

able at www.elforsk.se, accessed 2009-03-20. In Swedish.

[89] Hkan Skldberg and Thomas Unger. Effekter av frndrad elanvnd-

ning/elproduktion. Energimyndigheten och Elforsk, Elforskrapport 08:30,
April 2008. In Swedish.

[90] Hkan Skldberg, Thomas Unger, and Mattias Olofsson. Marginalel och
miljvrdering av el. Elforsk rapport 06:52, Augusti 2006. In Swedish.

[91] Jrgen Sjdin and Stefan Grnkvist. Emissions accounting for use and supply
of electricity in the Nordic market. Energy Policy, 32:15551564, 2004.

[92] Tobias Persson. Koldioxidvrdering av energianvndning: Vad kan du gra fr

klimatet? Underlagsrapport, Statens Energimyndighet, September 2008. In
Swedish.

[93] Stefan Grnkvist, Kenneth Mllersten, and Kim Pingoud. Equal opportunity
for biomass in greenhouse gas accounting of CO2 capture and storage: a step
towards more cost-effective climate change mitigation regimes. Mitigation
and Adaptation Strategies for Global Change, 11:10831096, 2006.

[94] Erik G. Lindfeldt. A trinity of sense: Using biomass in the transport sector
for climate change mitigation. PhD thesis, Department of Chemical Engi-
neering and Technology, KTH Royal Institute of Technology, Stockholm,

100
 Sweden, 2008. ISBN 978-91-7415-107-7.

[95] Erik G. Lindfeldt and Mats O. Westermark. Arguments in Swedish Policy

Documents for Using Biomass for Production of Bio-based Motor Fuels: Why
an Energy Engineer Could Feel Like Being in a Novel Written by Kafka. In
Proceedings of World Renewable Energy Congress X and Exhibition,
Glasgow, Scotland, 2008.

[96] Bengt Bergsten. Energiberkningsprogram fr byggnader en jmfrelse

utifrn funktions- och anvndaraspekter. Effektiv, 2001. ISBN 91-7848-851-6.
In Swedish.

[97] Gran Andersson. Miljvrdering av el. Swedish Energy Agency (Statens

Energimyndighet), 2006. In Swedish.

[98] ARCON SI-350-N Solar collector. Data sheet from ARCON Solvarme A/S.

[99] F. Fiedler, C. Bales, T. Persson, and S. Nordlander. Comparison of car-

bon monoxide emissions and electricity consumption of modulating and non-
modulating pellet and solar heating systems. International Journal of En-
ergy Research, 31:915930, 2007.

[100] Christian Roecker, MariaCristina Munari-Probst, Estelle de Chambrier, An-

dreas Schueler, and Jean-Louis Scartezzini. Facade integration of solar thermal
collectors: A breakthrough? In Proceedings of ISES Solar World Congress
2007. ISES, Springer, 2007.

[101] T. Bostrm, W. Glad, C. Isaksson, F. Karlsson, M.-L. Persson, and A. Werner.

Tvrvetenskaplig analys av lgenergihusen i Linds Park, Gteborg. Arbetsno-
tat Nr 25, ISSN 1403-8307, Program Energisystem, Linkping, Sweden, 2003.
In Swedish.

101
  -   /

  
 
  
 

, * 0 1) 2 0,  2

,  ,   0

 1) 2 0,
  23 // - 23))2)

2 0
)
+ 0//40 / 0 
/ ,   
5/  
6 ,+3 
)

2 , +) ,   2  )

 7
/-)

 0 //
   0
 1) 2 0,  24
89  :)23#..3 /)+,), 
 ;7
/-)

 0 //   
0
 1) 2 0,  2<4=




  
 +)  */)+  4))4  
)*+**))*,-"."(#