Life-Cycle Analysis and Optimisation of Solar Home Systems

Life-Cycle Analysis and Optimisation of Solar Home Systems
21 June 2000
ECN-C--00-047
Life-Cycle Analysis and Optimisation of Solar Home Systems
Interim report over 1999 of ENGINE project 74513
J.D. Dogger, J.C. Jansen, M.C.C. Lafleur, P.E. Lasschuit ,
N.H. van der Linden, F.D.J. Nieuwenhout, M.R. Vervaart
Abstract
This report describes the activities and outcomes of the first six months of the three-year
ENGINE project Life-Cycle Analysis and Optimisation of Solar Home Systems (project
number: 74513).
The whole project is divided into four main activities that cover different perspectives of the use
of solar PV equipment by households: monitoring of solar home systems, conducting a
household survey, socio-economic and institutional analysis and lifetime tests of PV equipment.
In this phase of the project we focussed on hardware development for a data acquisition system,
on preparation of a survey questionnaire and on general socio-economic and institutional
aspects of solar home system programmes.
An analysis was made of experiences with data loggers for solar home systems. Furthermore, a
prototype was developed of a small data logger. After testing the prototype it was concluded to
postpone further development activities and rely for the near future on equipment that is readily
available on the market.
Several versions of an advanced user interface were tested. Improvements were discussed with
the manufacturer, AME. Until now the modifications have not yet resulted in a version that can
be tested in the field.
As preparation for the household survey, which will take place parallel to the monitoring, a first
draft questionnaire has been formulated. This needs to be expanded to include questions
regarding lifecycle, socio-economic and institutional aspects. Zimbabwe has been selected to
conduct the household survey and the monitoring.
The socio-economic impacts discussed in this study deal with improvement of the quality of
life, the stimulation of commercial activity and associated employment and getting acquainted
with electricity.
In the first phase of this project, the preparations have been made for the experimental work in
the next phase. We are ready to start survey and monitoring work soon.
2
ECN-C--00-047
CONTENTS
1.
INTRODUCTION ............................................................................................................................. 4
1.1 BACKGROUND.................................................................................................................................. 4
1.2 OBJECTIVES AND SCOPE OF WORK .................................................................................................. 4
1.3 PROJECT PROGRESS IN 1999............................................................................................................. 5
2.
DATA ACQUISITION IN A SOLAR HOME SYSTEM............................................................... 7
2.1 EXPERIENCES OF OTHERS WITH MONITORING OF SOLAR HOME SYSTEMS ......................................... 7
2.1.1 Monitoring of solar home systems with the help of surveys ................................................... 7
2.1.2 Complement survey information with in situ measurements .................................................. 8
2.1.3 Experiences with the use of data loggers ............................................................................... 9
2.2 FORMULATION OF SPECIFICATIONS FOR THE DATA LOGGER ........................................................... 10
2.3 BUILDING AND TESTING OF THE FIRST PROTOTYPE......................................................................... 11
2.4 SECOND PROTOTYPE OF THE DATA LOGGER ................................................................................... 12
2.5 CONCLUSION ................................................................................................................................. 13
3.
ADVANCED USER INTERFACE ................................................................................................ 14
4.
PREPARATION FOR THE HOUSEHOLD SURVEY ............................................................... 16
5. SOCIO-ECONOMIC AND INSTITUTIONAL ASPECTS OF INTRODUCTION OF SOLAR
HOME SYSTEMS ................................................................................................................................... 18
5.1
5.2
5.3
5.4
6.
THE ROLE OF SHS IN NATIONAL AND REGIONAL ENERGY PLANNING ............................................. 18
THE SOCIO-ECONOMIC IMPACT OF THE INTRODUCTION OF SHS .................................................... 19
FINANCING SOLAR HOME SYSTEMS ............................................................................................... 20
INSTITUTIONAL ARRANGEMENTS................................................................................................... 22
PROPOSED ACTIVITIES FOR 2000 .......................................................................................... 25
ANNEX 1. DRAFT SHS PERFORMANCE QUESTIONNAIRE
26
ANNEX 2. DATA LOGGER HARDWARE
31
ANNEX 3. REFERENCES
34
ECN-C--00-047
3
1.
INTRODUCTION
1.1
Background
Solar home systems for rural electrification in developing countries and grid-connected PVsystems in industrialised countries are the backbones of the global PV-market and the
corresponding renewable energy policy. Grid-connected solar PV systems are primarily sources
of electricity, while solar home systems can supply services for the two billion people, which
are not connected to an electricity grid. Apart from appropriate financing mechanisms, the
price/quality ratio is still a major barrier for large-scale introduction. Only with a substantial
decrease in costs and improvement of quality one can expect that solar home systems will have
a rapid breakthrough. When solar home systems have to contribute to sustainable development
it is essential to spend sufficient attention to socio-economic and environmental aspects.
Because relevant and representative data about the use of solar home systems in households are
very rare, the pace with which solar home system components are being improved is slower
than desirable.
There is a serious lack of knowledge world-wide about the following topics:
• Reliability of solar home systems under field conditions;
• Lifetime of PV-system components in relation to their use;
• Environmental effects of solar home systems over the complete product cycle;
• Effects on the socio-economic development of the areas where PV-systems are introduced;
• Preferences and wishes of (potential) end-users.
November 1999 ECN and University of Utrecht started a Novem project that complements this
ENGINE study. The project ‘Monitoring and Evaluation of Solar Home Systems’ is intended to
target these issues by conducting a large literature survey into experiences with solar PV for
households in developing countries. Lessons learned from projects documented in the (grey)
literature will be summarised and analysed. May 2000 is the deadline for the Novem project. Its
outcomes will be useful in focussing the scope of the ENGINE project to those issues where the
lack of information is most serious.
1.2
Objectives and Scope of Work
The main objective is to improve the price/quality ratio of solar home systems. Quality is
defined here as the extent to which users are satisfied with the services provided by the system
in a sustainable way. The project has to contribute to increased knowledge of how solar home
systems are used by households in developing countries. Manufacturers can use these insights to
design more appropriate components and systems and to improve system sizing.
Intended results are the following:
• Information about the causes of failures of PV systems and their components;
• Insight in the influence of the feed-back of user information on actual use of SHS;
• Overview of the life-cycle of SHS which will lead to recommendations for new products
and product improvements;
• Insight into the strengths and weaknesses of the institutional framework;
• Knowledge about how solar home systems are used over longer periods of time (years);
• Information about preferences of users of solar home systems;
4
ECN-C--00-047
•
•
•
A methodology for duration tests of solar home system components;
Conclusions about the lifetime of a number of solar home system components;
Insight into the effect of climate circumstances on performance of a solar home system.
The project is divided into four major activities:
1) Monitoring
In a representative group of households, data loggers will be integrated into their solar home
systems. In some households in Indonesia a new user-interface will be tested. This userinterface has been developed by ECN and the Indonesian company PT Cilengka. A number of
other households act as control group. Outcomes will be used to formulate recommendations for
modifying charge regulators and the user-interface and for sizing of the different components.
2) Survey
Information from the monitoring activity will be linked to results from a household survey. A
number of households will be visited a few times over a number of years. With the help of a
survey questionnaire the following issues will be assessed among others: failure rates of the
different components, maintenance and waste disposal.
3) Socio-economic and institutional analysis
An analysis will be made of the social and economic circumstances of the users in relation to
the solar home system, and conclusions that can be drawn regarding productive applications.
Furthermore we will assess how solar home systems fit into regional and national energy and
environmental planning.
4) Duration tests
Lifetime tests will be conducted for a number of components: charge regulators, batteries and
lights. Based on the monitoring outcomes, a selection of components for the duration test will
be made.
1.3
Project progress in 1999
Start
In July we received formal permission to start work on the first phase of this project. However,
some preparatory activities were conducted already earlier in 1999. We were explicitly asked to
take the following three remarks into account:
1. Beware for an overly ‘high-tech’ approach, especially focussing on demands of the user and
owner;
2. What is new with respect to monitoring in Swaziland?
3. This project focuses only on Indonesia and that makes the project vulnerable. Spread the
risks. See if it can be applied in other countries for example Swaziland or Botswana.
The first remark will be taken care of through the survey questionnaire. As can be seen in
Annex 1, a number of questions explicitly enquire about user experiences (especially questions
8-13). Also indirectly, via the monitoring results, we intend to learn more about household
demands.
ECN was involved in evaluation of solar homes systems in Swaziland with a survey as the
major instrument to obtain the required information. In this Engine project, additional to the
surveys, is the data gathering via data loggers, which provide detailed information about the
actual use of the PV-system. This information is limited in scope and quantity, but can be much
more reliable than behaviour distilled from survey questions. By the parallel use of the tools of
data loggers and surveys we intend to obtain a more complete picture of household demands
than we have at the moment.
ECN-C--00-047
5
The third remark is very well taken. For the activities in 2000 we propose to apply the research
both in Indonesia and in Zimbabwe. The background for the selection of Zimbabwe is discussed
in chapter 4.
Monitoring module
Due to the circumstances where the monitoring takes place, an ideal solar home system data
logger is small, reliable, has a large data storage capacity, a low electricity consumption and has
a low cost. There is no data logger available in the market that is really suitable and meets all
these requirements simultaneously. Therefore, a large effort was made to develop our own data
logger for this specific purpose of monitoring solar home systems. At the moment we have a
version which is currently being tested, but which still has a number of flaws. For the second
phase of this project in 2000, we propose to postpone development of our own data logger and
use off-the-shelve products. We propose to continue development of a data logger only after the
use of existing data loggers have demonstrated that they are less suitable for monitoring solar
home systems.
User interface
A prototype of an advanced user interface was produced by AME based on a concept developed
by PT Cilengka in Indonesia in cooperation with ECN. Via a small LCD screen the user
receives information about the use of their solar home system in the last week. This prototype
was tested by ECN, and a number of problems were found. Suggestions for improvements were
formulated, and these were implemented by AME. Additional tests have been conducted, which
show that the user interface still needs further improvements. For field tests in Indonesia we
intend to use an older version of the user interface.
Preparation of the survey
Some first preparations were made for the survey that will be conducted parallel to the field
monitoring. An inventory of important issues has been made, and a first draft questionnaire has
been formulated.
Socio-economic and institutional aspects of introduction of Solar Home Systems
A global review was conducted regarding the role of solar home systems in regional and
national energy planning. Furthermore, we looked into socio-economic impacts of the
introduction of (solar) electricity, the financing mechanisms and the different institutional
arrangements.
6
ECN-C--00-047
2.
DATA ACQUISITION IN A SOLAR HOME SYSTEM
2.1
Experiences of others with monitoring of solar home systems
Parallel to this ENGINE project, a study is being conducted for Novem called ‘Monitoring and
evaluation of solar home systems’. It aims to assess world-wide experiences with solar home
systems, solar lanterns and battery charging stations. Final results of the Novem study, which is
due May 2000, will provide more detailed insights into the actual requirements for field
monitoring. However, in the framework of this ENGINE study some preliminary findings are
outlined in this section 2.1.
By conducting literature research that was supplemented with contacts with a number of
institutes involved in monitoring, the following picture emerges of existing, world-wide
monitoring experiences. Three modes of data gathering can be distinguished: surveys with only
questionnaires, surveys supplemented with one-time measurements and the use of data loggers,
which regularly measure a number of relevant system parameters.
2.1.1 Monitoring of solar home systems with the help of surveys
Kenya
Most of the publicly available information on the performance of solar home systems comes
from surveys. One of the most useful publications is by Acker and Kammen, describing a
survey conducted in Kenya1 in about 40 households. One of their findings was that 40% of the
smaller solar home systems (<25Wp) is only partly operational and 13% is inoperational. Large
systems fare somewhat better: 25% are partial operational and only 8% are inoperational.
Amorphous silicon modules mainly power small systems. According to the authors, the higher
failure rate is “.. doubtless due to the small size of the systems, which often cannot produce
enough electricity to satisfy the demand of the household. When the family tries to use more
energy than the panel can supply, it leaves the battery in a continuously low state of charge,
resulting in a damaged battery with a shorter life.”2 However, according to ECN experience it is
very unlikely that the size itself is the main reason for the lower performance. A common
practice in smaller systems is to leave out the battery charge regulator, which saves on the firsttime investment costs but is detrimental for the quality of the system. This discussion will
remain unsettled before more monitoring data from data loggers become available. Only by
analysing the link between the user behaviour and performance of the system, one can trace the
real reasons for these high failure rates.
Swaziland
ECN has conducted surveys in Swaziland as described in a report by Petra Lasschuit3. Survey
results showed that 86% of the respondents were happy with their systems and 96% would
recommend a PV system to others. A much lower percentage than in Kenya (25%) had
problems with their systems. However, the lifetime of the batteries is shorter than the expected
three years. Of the people who had to replace their battery once or more, 73% had to do so
within 2 years and 42% even within one year! Also in this case, monitoring with data loggers
will provide more insights into the determinants of the short battery lifetime.
ECN-C--00-047
7
Indonesia
Monitoring with surveys only provides a good overview of the technical and non-technical
problems occurring with the use of solar home systems. However, to be able to analyse these
problems and identify solutions, one needs to obtain more detailed information. Part of this can
be obtained by adding more specific questions to survey questionnaires. But there is a limit to
the extent to which one can gather information via surveys. Ideally, it should be complemented
with on site measurements, as was also concluded by Angele Reinders in an article regarding
the experiences in Sukatani4: “ The combination of an analysis of monitoring data, a field
survey and interviews of SHS-users show some contradictions. For instance, on the basis of
monitoring data we could not conclude whether users had to adapt their electricity consumption
in the rainy season. In interviews, however, they told us that they did have to do this.” “By
comparing the number of installed lights with the lights actually used according to the villagers,
we noticed that their answers were influenced by the information given in the instruction sheet.
Due to deviations between real and narrated experiences, we conclude that a field survey
that comprises only interviews may not be sufficient to assess an SHS-project.”
2.1.2 Complement survey information with in situ measurements
Brazil
Solar home systems have been installed in Brazil by a number of utilities5. One of these,
CEPEL, has developed a performance evaluation methodology that can be easily applied by the
utilities’ technicians, using only simple instrumentation. Periodically, or after a fault has been
reported, a performance evaluation form is filled out. In situ measurements are made with a
digital multimeter, a current probe, a portable solar radiation meter and a temperature probe.
The following measurements are made:
• Battery voltage and current during charge and discharge;
• Battery open circuit voltage;
• PV panel short circuit current;
• Solar radiation at measuring interval;
• Ambient temperature.
No PV-panel open circuit voltage measurements have been made because of reliability and cost
reasons according to the authors.
FhG-ISE
Fraunhofer-ISE developed a ‘Solar Home System Tester’, a cheap, easy to use, hand-held
device which can assist in checking the performance of a solar home system in a short time. The
tester can measure all the parameters as measured by CEPEL in Brazil plus the module open
circuit voltage. For these measurements the tester is connected between the charge regulator on
the one hand and the module, battery and load on the other hand. This requires some time and
effort, and a short measurement is only possible if the solar home system is already adapted for
the use of the tester. January 2000, one of the developers of the tester, J. Kuhmann of
Fraunhofer-ISE stated that the tester is not yet commercially available.
Conclusions
From the monitoring results in Brazil it can be concluded that the extra on site measurements
provide useful quantitative information about the performance of the systems. This can be very
helpful for maintenance purposes, which is one of the main objectives of the utilities in Brazil.
However, with this type of monitoring, one still lacks information on time development of
performance (e.g. degradation of the battery). What is even more important, one still does not
know how the systems are actually used by households: during what part of the day the energy
is consumed, how often is the battery empty, and what is the extent of shading of the module
etcetera. To obtain this type of information one has to use data loggers.
8
ECN-C--00-047
2.1.3 Experiences with the use of data loggers
FhG-ISE
Fraunhofer Institut fur Solare Energiesysteme has experience in using data loggers in solar
home systems in a number of developing countries. In Balde de Leyes in Argentina, data
loggers were installed in two solar home systems6. Information that was obtained from the data
loggers was:
• 95% of the electricity consumption occurs during the night;
• energy demand over the year is proportional to the lenght of the nights. Where the seasonal
irradiation varies between 3 kWh/m2/day in winter and 8 kWh/m2/day in summer, the
average energy consumption per household was only between 200 Wh per day in winter and
120 Wh per day in summer.
• There was not much difference in energy consumption between larger and smaller families;
• There was no energy cut-off caused by a deep discharged battery!
FhG-ISE usually applies data loggers that measure three currents and three voltages. A reason
given for having separate load and module voltage measurements is that it provides information
about the low-voltage and high-voltage disconnect. For example, if the measured load current is
zero, but at the same time the load voltage is unequal to zero, it shows that the user switched off
the load and not the low voltage disconnect.
NREL
Different groups at the National Renewable Energy Laboratory NREL have experience with
data loggers for remote power systems. Some of these can also be relevant for monitoring of
solar home systems. Despite years of efforts they are still looking for an ‘ideal’ logger for
remote power systems, that is reliable and does not consume too much energy. They usually
apply Campbell Scientific data loggers that are of high quality. However, the own energy
consumption of the Campbells is high, which makes independent operation difficult.
Alternatives such as the weather loggers of NRG have lower own consumption, but have fewer
channels for the same costs. A few years ago a proposal was formulated to develop a data logger
especially for monitoring of small PV systems. However, no funding was obtained.
One of the problems that occurred in the field was that tampering with the battery bank resulted
in damaging the data logger. Furthermore, incorrect installation and accidental disconnection of
wires have resulted in loss of data. Lessons learned from a workshop with FhG-ISE and Sandia,
held at NREL on October 19th, 1999 were the following:
• It costs time to check and process data;
• Collecting data can be difficult;
• Take time for installation and checking;
• Include redundancy in measurements (even when a parameter can be calculated from others,
still measure it);
• Total Watt-hour input and output to the batteries is required;
• The Data Acquisition System (DAS) needs its own power supply: the interesting things
happen when the system is dying;
• Data cards are the most practical method to gather data;
• Review of accurate log books is essential;
• Simple is best.
Utrecht University
Angele Reinders formulates the most detailed description of monitoring of solar home systems
with data loggers in an article7. In Sukatani, Indonesia, four-channel data loggers were used of
the Squirrel type, manufactured by Grant. Measurements were made of battery voltage, current
from the array, current to the load, and irradiance in the plane of the array or temperature in the
ECN-C--00-047
9
battery box. Sample interval was 18 seconds and data were stored in half-hourly averaged
values.
Based on an extensive analysis of the monitoring data over the period 1988-1993, Reinders
arrived at the following conclusions:
1. For more accurate analyses and better insight into battery performance, measurements are
recommended by JRC or IEC of a) power instead of current, b) power into the battery and
from the battery, and c) non-availability of power to the load.
2. For determining the state of charge of the battery, the recording interval needs to be
decreased to less than one minute.
3. Irradiance sensors (and modules?) need to be cleaned once every two days.
4. Use of a portable reference cell gave problems with correct orientation of the sensor and
having simultaneous array current and voltage measurements. Measurements of irradiance
taken during site visits were therefore not used in the analysis.
2.2
Formulation of specifications for the data logger
In an early stage of the project, a number of research questions were formulated regarding the
use of solar home systems by households in developing countries, which could possibly be
handled by obtaining monitoring data:
1. Is there (partial) shadowing of the modules during the day?
2. What part of the electricity is used during the day and what part during the night?
3. What is the power consumption of the equipment used (the load)?
4. What is the state of charge of the battery at the beginning and at the end of the evening?
5. At what time of the day does low voltage disconnect occur?
6. What is the actual capacity of the battery?
7. What are the losses in the battery?
8. What are the different components of the system losses due to: a) non-optimal position and
orientation of the module, and b) losses in cables, charge regulator and battery?
9. Does it happen that the battery is also charged in a battery charging station?
Based on these monitoring demands the following parameters were chosen for data logging (in
brackets the number of the questions in the above list):
a) System voltage stored as minimum voltage and maximum voltage per hour (4,6,9)
b) Module current stored as average module current per hour (1)
c) Load current stored as average load current per hour (2,3)
d) Battery charging current stored per hour as Ampere hours in and Ampere hours out (4,6,7)
e) Flags per hour for low voltage disconnect and high voltage disconnect (5)
Answering question 8 about the different system losses can not easily be achieved just with a
data logger alone. This requires additional on-site measurements. Furthermore, it requires
measurement of irradiation that is not planned by us, due to the fact that it requires additional
equipment and extra wires, thereby increasing the expected chances of malfunctioning of the
data logger.
In this stage, only the parameters that need to be measured were formulated. No other
specifications were formulated such as environmental conditions, level of accuracy and cost
limits.
10
ECN-C--00-047
2.3
Building and testing of the first prototype
As far as we knew at the start of this project, there is no suitable data logger available in the
market place that is small, cheap and reliable, has a large memory and a low own consumption.
Therefore we decided to develop a customised data logger for monitoring solar home systems.
The first design ideas for the data logger were based on a charge controller with an Ampere
hour balance for the battery, which was developed earlier by Mark Vervaart. For the Ampere
hour balance data acquisition and storage is required. We expected that this could be expanded
to a data logger measuring the specifications under a) to e) above. Data are stored in an
EEPROM, which can be taken from the data logger and can be read via a special computer
interface module. In a relatively short time the prototype of the data logger and the computer
interface were built.
Tests were conducted on this prototype in a small solar home system located in Alkmaar. This
solar home system is meant for test purposes and consists of a 19 Wp multi-crystalline PV
module, an old 76 Ah battery, two fluorescent lights (Suntec of 7 Watt and PT LEN of 5 Watt)
and a timer for switching the lights on and off.
Some minor problems were detected in the tests. The data of the first few hours were stored in
one single location, and only after an unknown number of hours the actual data logging started.
This problem was solved.
Results for a three-day period in June 1999 are presented in figure 1, which shows hourly
averages of system voltage, the current from the module and the current to the load. The first
day was a completely clear day. The low module currents before noon are due to shading. Direct
sunlight only reaches the module after about 13.00 hours (MEST). Especially during the second
day, the energy consumption of the load was larger than the energy generated by the module,
resulting in a decrease of the system voltage due to a lower state of charge of the battery.
Figure 1 Monitoring results for three days in the summer: hourly averages of system voltage and
currents from module and to the load.
Monitoring results SHS Alkmaar
3 days in June 1999
16
14
Voltage or current
12
10
Module current (X 100 mA)
Load current (X100 mA)
Voltage (V)
8
6
4
2
73
70
67
64
61
58
55
52
49
46
43
40
37
34
31
28
25
22
19
16
13
7
10
4
1
0
Hours after start
ECN-C--00-047
11
Figure 2 Monitoring results for the same days as in figure 1: peak system voltage, hourly
average load current and peak load current.
Monitoring results:
peak system voltage, average load current and peak load current
25
Voltage or current
20
15
Peak power of load (Watt)
Peak voltage (Volt)
Load current (X100 mA)
10
5
73
70
67
64
61
58
55
52
49
46
43
40
37
34
31
28
25
22
19
16
13
7
10
4
1
0
Hours from start
2.4
Second prototype of the data logger
With the successful tests of the first prototype it was illustrated that in principle the data logger
can operate well. However, being integrated in a charge controller that is not commercially
available is a barrier for normal use of the data logger. Ideally you want to be able to use the
data logger in all different types of solar home systems. Therefore, the next logical step is to
develop a version that can operate independent from the charge regulator. Given the relative
ease with which the first prototype was developed, it was assumed that modifying the design to
make it independent would be a simple, straightforward job which should be easy to
accomplish. In reality it turned out to be a long and tedious process.
A disadvantage of having the data logger separate from the charge regulator is that the
information about the status of the charge regulator (e.g. a disconnected load due to low voltage
disconnect) is no longer easily available. This limits the parameters which can be measured to
four: system voltage, PV-module current, battery current and load current.
A new circuit diagram was drawn for the second prototype. In the summer, a student was asked
to draw a diagram for a PCB. Inadvertently, a number of errors were introduced in the PCB
design. This resulted in serious delays because the errors had to be repaired manually.
Laboratory tests of the second prototype showed problems with the data. It took a considerable
amount of time before some of the causes were traced. The following problems were
encountered and solved:
• The sample time was not accurate enough. Load and insolation pattern shifted in the order
of one hour per week. This was corrected by using a slightly different crystal oscillation
frequency.
• A separation byte between the seven data bytes was accidentally overwritten due to a
change from eight to seven data bytes.
12
ECN-C--00-047
•
•
Measurements of the Ah of the battery were wrong. This was caused by a mistake in a
counter.
The load current measurements were wrong. This was solved.
Since early December 1999 the second prototype is being tested in the solar home system in
Alkmaar. However, there are still a number of unsolved problems, which require further
research:
• There were problems with EMC. The data logger module appears to be sensitive to all kinds
of spikes. Additional capacitors have been placed in the data logger, but it is unlikely that
all the EMC problems have been solved.
• Measurements in the first hour show values that may be too low. Causes are still unknown.
• Voltage measurements are 0.1 to 0.2 Volt lower than the actual values.
• Current measurements for battery and load current are not linear. With a simple software
correction some improvement have been achieved, but measurements below 1A are still not
accurate enough.
• The data from the measurements in the solar home system have not yet been analysed. It is
therefore unclear what the influences of positive and negative switching are on this data
logger.
2.5
Conclusion
The problems with designing a data logger module do not seem insurmountable. However, they
probably require substantial changes in the current design. Given the uncertainty in time
requirements for these modifications, it was decided to postpone the development of an ECN
data logger and rely instead on data acquisition equipment that is readily available on the
market. When there is more experience at ECN with the use of data loggers in solar home
systems we are better suited to formulate the right specifications.
ECN-C--00-047
13
3.
ADVANCED USER INTERFACE
First prototype
Users of solar home systems have limited information available about the state of charge of the
battery. Sometimes only two or three LEDs show if the battery is empty or full. For users it can
be helpful to be informed beforehand that they are depleting their battery too fast. To assist
users in determining if their demand is line with supply an advanced user interface has been
designed. The user interface shows in the form of histograms the highest and lowest voltage
reached by the battery in the past week. With the help of the trends shown by the display the
user can adjust its electricity consumption.
A first prototype for a user interface for solar home systems was developed in an earlier project
(ENGINE project Autonomous Systems 74433). This is one of the solar home system
components that will be included in the process of design and modification by taking into
account life cycle aspects. Before field testing can start, the prototype was tested in the
laboratory to see if it meets the specifications and if it can operate under developing country
conditions. Within the current preparatory phase of this ENGINE project, only the laboratory
tests were conducted. Field testing has to start in the second phase of this project.
The user interface shows the actual voltage of the battery in the range of about 11.5 to 13.5 volt
in steps of 0.2 Volt. Displayed is a histogram with bars consisting of a number of small blocks
of 0.2 Volt each. It displays minimum and maximum battery voltage for the past 6 days. This
provides the users with information about how their electricity use is related to daily electricity
supply.
Laboratory tests
A number of laboratory tests have been conducted to test accuracy and reproducibility of
voltage indications and to determine if the values are shifted correctly from day to day.
The first laboratory test was to check if the value of the voltage as shown by the user interface
differs from the actual voltage as measured with a separate voltage meter. The average
difference found was 0.1 volt, which is equivalent to half a block. The maximum difference was
0.4 Volt, occurring in 7% of the 42 measurements.
Furthermore, tests were conducted in a small solar home system. 14 user interfaces were
connected parallel in groups of 2 or 4 for periods of a few weeks each. Because the interfaces
were working in parallel, they are expected to display the same patterns. This was not the case.
The main problems that were encountered are the following:
1. A large discrepancy between voltage displayed and the actual voltage occurred with one
sample;
2. Large differences between displayed voltages of one or more days ago and actual voltages
occurred twice;
3. In one sample the display of the previous days did not shift every day;
4. With symbols (thumbs up and down) the users are informed about the status of the system.
However, three samples showed at the same time the two different symbols that users
behaved well and not well.
AME, the producer of the interface, offered to add a software filter to reduce the first two
problems and software modification for the fourth problem. 50 interfaces were returned to AME
and reprogrammed.
14
ECN-C--00-047
Second prototype
ECN tested the interfaces with the adjusted software. They were connected to a solar home
system in two groups of 16 each. In a period of three days the battery was discharged from 14.5
Volt to 11 Volt. This resulted in the following findings:
• When the input voltage is within the range of 11.5 to 11.7 Volt the interfaces all displayed
level 12 instead of level 2. Depending on the length of the measurement interval, these
values also show up in the history.
• Twice the voltage accidentally dropped to below 4 Volt. No problems occurred due to the
low input voltage.
The interfaces were returned to AME for the second time. According to the manufacturer a
software error was traced and repaired.
Third prototype
Testing of the latest version showed the following:
• When the voltage drops below a certain threshold, all levels in history become equal to 1.
After this happens, the history can not be updated anymore, not even by using the reset
button. Consequently, when the voltage drops below a certain level (which could not be
determined precisely) the interface becomes permanently destroyed. In practice, the
interface will usually be connected directly to the battery. There is a real chance that low
input voltages would occur.
• The ‘old’ problem with the voltage range of 11.5 to 11.7 volt still occurs.
Conclusions
The problems that occur are too severe to have this third prototype of the user interface tested in
the field. Therefore we decided to test user reactions on the first prototype before further
changes to the design will be made.
ECN-C--00-047
15
4.
PREPARATION FOR THE HOUSEHOLD SURVEY
The technical monitoring process will be complemented with interviews among the end-users.
The aim of the interviews is to get answers on questions that can not be derived from the
technical monitoring process. The information thus collected will include the following issues:
- the quality of installations;
- the maintenance of the systems;
- use of appliances;
- lifetime of components;
- disposal of redundant system components;
- impact of PV system on social and economic development;
- user satisfaction.
The survey will consist of one comprehensive interview and biannual follow-up visit. Trained
enumerators will accompany the technicians installing the data loggers and conduct the initial
interviews using a standard questionnaire (see draft questionnaire annex 1). After the first
comprehensive interview, a shorter questionnaire will be prepared to update the collected data
especially on the technical issues. Follow up interviews will be combined with the data readings
(data loggers) and carried out by the technicians for 2 times a year, viz. once in summer, once in
winter.
The country selection has long been debated. Initially one country, viz. Indonesia was selected
as a study case for this project. From a representative point of view, the selection of a second
country on another continent has been envisaged. With most of ECN’s PV implementation
activities being carried out in Africa the choice for this continent seems obvious. Below the
pro’s and con’s of three potential countries are presented.
Swaziland
ECN-IDE carried out several energy surveys in Swaziland and through its long presence in the
country has established a substantial network, making the organisation of the survey rather easy.
In addition, ECN-IDE together with its local partner Swazitronix has installed about 500 solar
homes system and has access to another 500 solar systems identified in a previous survey.
Despite the relative easy of organisation, Swaziland is a rather small country and has a higher
per capita income than most other countries in Africa.
Kenya
Kenya has a substantially developed PV market, probably the largest among the developing
countries. From this perspective Kenya would be an interesting and representative case study.
Discussions with local Kenyan organisation took place early December and suitable partners
have been identified for the organisation of the survey.
A disadvantage of Kenya as a case study is that ECN has no ongoing activities in Kenya in the
field of PV. As such the organisation of the organisation of the survey will be more time
consuming and costly.
Zimbabwe
Zimbabwe is a country that overcomes both problems. It has the second most developed PV
market on the continent and ECN-IDE has been actively involved in the country. ECN has
established a large network in the PV sector and will have permanent ECN staff in the country
for the next 2 years.
16
ECN-C--00-047
Through one of our local partners, viz. PV supplier, ECN will get easy access to a large
customer data base and can make use of a team of qualified technician for the installation and
readings of the data loggers.
ECN-C--00-047
17
5.
SOCIO-ECONOMIC AND INSTITUTIONAL ASPECTS OF
INTRODUCTION OF SOLAR HOME SYSTEMS
As set out in Chapter 1 the study will also entail a socio-economic and institutional analysis of
the SHS market. That analysis will be further explored in this chapter. Its objective is to analyse
social and economic circumstances of the users in relation to the solar home system, and to
draw conclusions regarding productive applications. Furthermore it will be assessed how solar
home systems fit into regional and national energy and environmental planning.
Unfortunately, the implementation of this analysis was delayed by a number of factors that
affected also the overall implementation of the project. First, the late date of approval of the
project diminished the availability of persons to implement the project. Secondly, the approval
resulted in a reduced budget available for the whole project, which led to redefining of priorities
and subsequently a re-division of activities to be implemented for the socio-economic and
institutional analysis. Thirdly, and foremost, the selected target area suffered from severe
political and violence problems during the course of the year. Therefore, no field work could be
undertaken to actually conduct the survey and identify and evaluate the main socio-economic
and institutional aspects relevant for the introduction of solar home systems.
Therefore, the analysis presented in this chapter is of a more general nature and provides a
framework for the actual analysis, which is envisaged in the second phase of this project.
The analysis presented in this chapter draws heavily on work done by S. Dlamini who was
seconded at ECN for a period of three months. Mr Dlamini’s research is reported in a separate
document (ECN-I-99-005). In this chapter, a number of main issues resulting from that study
will be presented. First, the role of SHS in national and regional energy planning is outlined.
Next, the socio-economic impact of rural electrification in general and SHS in particular is
explained and finally the financial and institutional aspects of the introduction of SHS is
elaborated upon.
5.1
The role of SHS in national and regional energy planning
In developing countries, urban households consume the largest part of residential electricity.
Although since 1970 approximately 800 million people in rural areas gained access to
electricity, the problem still remains that, of the approximately 3.2 billion people living in rural
areas of developing countries in 1990, 1.8 billion are still without access to electricity.
This means that the development options of the vast majority of rural households are seriously
hampered and production and service establishments in these areas are disadvantaged. As a
result, the already existing social gap between urban and rural communities will further increase
and will stimulate the migration to urban areas. This situation is regarded by the national
government as highly unsatisfactory and usually the energy policy formulated in developing
countries is addressing this problem by initiating various activities to promote rural
electrification and to analyse the options to increase the rural connection rate in the most costeffective manner.
18
ECN-C--00-047
However, the decentralised character of the population of rural areas and their small per capita
commercial energy consumption has made these areas less attractive for grid extension. The
extension of the grid to rural areas is in many cases less cost-effective as compared to urban
areas. Therefore, the present emphasis in the energy policy with regard to supplying electricity
to small and scattered loads is on decentralised generation of electricity. The main options are:
•
•
•
•
•
minihydro
photovoltaics
wind
biomass
diesel
Facing the social pressure to address the energy needs of the rural population, governments in
developing countries are increasingly considering off-grid PV systems as an attractive means to
address some of the energy problems in rural areas and to improve the quality of life. A good
example in this regard is Kenya with the highest penetration rate of household photovoltaic
systems in the world. To date, more than 80,000 systems have been installed and current annual
sales amount to approximately 20,000 systems. Some 50 local and 15 international companies
import, assemble, install and provide after sales in this market.
Although the example of Kenya clearly shows the potential role SHS systems can play in
regional energy polices, there is still a tremendous need for strengthening the local PV industry
in many developing countries through studying the credit arrangements and standardisation of
equipment.
5.2
The socio-economic impact of the introduction of SHS
The socio-economic impact of the introduction of SHS is related to the improvement of the
quality of life, the stimulation of commercial activity and associated employment and getting
acquainted with electricity.
Improvement of the quality of life of rural population is brought about by the fact that electric
lighting enables people to undertake a range of additional activities in the evening hours at
home and in public facilities such as schools, clinics and community buildings. Outdoor
lighting may also bestow a perception of improved security. Electricity for radio and TV gives
people access to mass media for entertainment, but also for extension services and distance
education. Electric pumps powered by PV may facilitate access to safe drinking water.
Electricity for small local health facilities enable conservation of medicines and emergency
distance communications.
Stimulation of commercial and agro-industrial activity and associated productive employment is
brought about by the establishment of a PV market consisting of dealers, assemblers and after
sales providers of the PV systems. In addition, improved lighting is an inherent benefit for shop
owners stimulating their trade.
Electrification by means of PV can be seen as an interim measure to provide electricity services
until the national grid will have reached the area. However, PV acquaints the consumer with the
use of modern electricity usage. This means that when a household eventually gets connected to
the grid, it is used to electric devices such as the radio and the TV and, consequently, electricity
consumption after grid connection will be higher compared to the situation whereby electricity
is introduced for the first time. Higher consumption, of course, contributes positively to the
financial performance of the rural electrification programme.
ECN-C--00-047
19
5.3
Financing Solar Home Systems
Two main potential consumer groups can be identified for SHS. The first group comprises of
potential consumers that do not have access to grid electricity. The second group involves
households connected to the grid but, because of the frequent power interruptions, uses the SHS
as a backup device.
Most SHS however are meant for rural households which are not connected to the grid These
households usually belong to the low income group of households which cannot afford the high
initial investment that is required for the purchase of a SHS. This raises the important question
of how to finance a SHS programme ?
In general, the following constraints and impediments in financing frequently encountered in
SHS programmes can be distinguished:
• access to finance:- this is claimed to be a cardinal pre-requisite for SHS projects, and
renewable energy projects in general;
• cost of finance:- when financing involves loans, raising and servicing the financial
arrangements comes with a cost;
• perceived risks:- all projects are exposed to risks, which are often considered high when the
project involves relatively new technologies. Avoiding, reducing and sharing risks amongst
key players is essential for success.
Three finance-related concerns can be identified that stand out as being critical if the SHS
industry is to grow successfully, namely
• consumers need to obtain credit from banks or distributors
• suppliers and retailers must be able to secure working capital if they are to be able to
provide in-turn credit to their customers
• investors need credible financing opportunities to sway capital towards the solar industry
Close inspection of literature reveals three main categories in the finance process arranged in a
fixed vertical structure viz; international financiers, several (institutional) intermediaries who
handle the funds and the end user of the funds. In practice one can see a pattern which further
breaks down these categories into five levels of actors, again vertically integrated. These are
illustrated by figure 1.
The level 1 actors are involved in supplying of capital, usually as a grant, subsidies and/or loans
with “strings” attached. These may be organisations like aid agencies, UN related bodies or
multinational banks.
20
ECN-C--00-047
INTERNATIONAL FINANCIER
• Grants and subsidies from aid agencies
• United Nations related bodies e.g UNDP, GEF
• Multinational banks e.g World Bank, regional
development banks
FIRST INTERM EDIARY
• National government
SECOND INTERM EDIARY
National
development banks
•
• Commercial banks
• Electricity utility
•
•
•
•
THIRD INTERM EDIARY
Rural cooperatives
NGOs
Private companies
Other community based organisations
END-USER
• Households
• Small businesses
Level 1
Level 2
Level 3
Level 4
Level 5
Figure 1 Typical finance delivery chain
The next levels involve intermediaries, whose goal is to see that the Level 1 funds end up with
the consumer in Level 5. It’s worth noting that the longer this intermediary chain, the more the
administrative overheads compile. Often these intermediaries are necessary, or at least
perceived to be, for successful PV market development. They fulfil necessary services in the
total market structure by creating economies of scale and hedging risk through the bundling of
projects.
National governments often intercept the funds as is the case for GEF projects, although this is
not always necessary for example, the Solar Development Corporation may skip level two
directly into level 3. Actually there is also a possibility of a loop directly from Level one to four,
as is with the Solar Investment Fund.
Last in the chain is the consumer himself. Although the literature contains diverse views on the
exact details of how funds have to filter down to the consumer, there are also remarkable
similarities. These exact details of flow of funds will be called financial mechanisms in this
paper. Example of the interactions between the levels is given in the section dealing with
institutional arrangements.
Several financial schemes have been developed and applied to overcome the barrier of high
initial investment, namely:
• Revolving Loan Fund (RLF) : the concept of RLF is simple. Initially, a fund is made
available to finance the SHS programme. The costs of the SHS are paid back in monthly
ECN-C--00-047
21
instalments that flow back in to the fund. This money can be used again to finance new
SHS programmes.
• Credit Financing: in this financing scheme the consumer deposits a down payment, and then
pays the balance in regular payments.
• Leasing schemes: this scheme draws similarities to the system of how grid electricity
consumers pay for electricity. The SHS is leased to the consumer for a monthly fee, the
utility (or Energy Service Company) will provide O&M services and hand-over the system
after a number of years.
5.4
Institutional Arrangements
Institutional arrangements are usually meant to facilitate the processes of the financial
mechanism(s), the product delivery mechanism(s), training and the after sales services. For this
reason, it is difficult to disentangle finance issues and those that are institutional. The following
key players can be identified in this respect:
•
•
•
the public authorities;
the utilities; and
the co-operatives and associations.
Public authorities
The conventional attitude of public authorities that SHS “doesn’t work”, “its too expensive” is
on its own an institutional constraint. This situation is encouragingly changing though and many
developing country governments have already included, RE technologies, including SHS in
energy policy objectives. The Solar Summit Process, in Harare 1996, gave a much required
impetus to the change of attitude of public authorities, resulting in their commitment to “ join in
the development and implementation of the World Solar Programme 1996-2005”8. The
declaration made during this summit includes inter alia three grave commitments in favour of
renewable energies in general, and solar energy in particular. Here commitment two is quoted,
which reads: “We commit ourselves to work towards policies and effective mechanisms that
will speed up and facilitate the use of solar energy avoiding duplication and administrative
delays, and the encouragement of international co-operation, including participation in regional
and international bodies, scientific and technical organisations”. The other two commitments are
of similar seriousness. Much work is ongoing and has already been done towards the fulfilling
these commitments. However, there is still a great scope for improvements and this remains a
challenge of authorities; to ensure that commitments made are effected and not just remain a
mere will.
On the other hand over-enthusiasm on SHS can lead into ill-planned projects yielding disastrous
results. The best role of public authorities seems to be one of market catalyst without excessive
interventionism.
Public authorities can also play an important role in removing or reforming fiscal and policy
regimes, which are in disfavour of alternatives, like SHS. In the Senegal case, authorities
waived duties on solar energy equipment, making systems more affordable and promoting the
solar business
Utilities
The attitude of utilities towards alternative energy, SHS for example, is changing for the better.
Although not funding the project, the utility is often responsible for quality assurance and
standards issues, which is a real achievement. The South African Utility, ESKOM, has been
involved with active research and implementation of alternative energies, including solar, since
22
ECN-C--00-047
19919. It has also recently launched an ambitious electrification project, in conjunction with the
Renewables division of Shell, which seeks to disseminate 50,000 system on a leasing basis.
The engineering background of utility engineers and technicians, and the infrastructure utilities
already possess, could be extremely useful if available also for SHS programmes. In fact
utilities with a commitment to rural electrification should find SHS an attractive cost-effective
option in many instances. The examples of utilities in many countries that are already active in
SHS should serve as a reassurance that it is possible to integrate SHS in electrification
programmes.
An interesting example is CRE, perhaps the largest electric co-operative in the world, which has
the electrical concession for the state of Santa Cruz in Bolivia. CRE has set up a “solar electric
utility” within the co-operative structure. CRE has developed a tariff collection and maintenance
system, and has over 1000 customers. Another distribution utility in Bolivia, ELFEC,
considered setting up a similar electrification unit to CRE, but later rejected it based on financial
analysis10.
Private sector initiatives
Another venue is the entering of private sector in the provision of electricity services. Such
companies are usually referred to as Energy Service Companies (ESCOs) and work as a “Solar
Electric Utility” which basically acts like grid electricity utilities. This type of company sells
electricity and retains ownership of the SHS. The SEC in Kiribati is one example. The principle
advantage of utility-based service is that it can provide electricity at low cost to the consumer. It
also transfers the maintenance responsibility from the user, to trained technicians, thus
guaranteeing proper care of systems. It however, requires a capable technical and administration
infrastructure in the area to be served and often needs access to long term credit at modest rates.
The Kiribati experience also highlights the importance of establishing a “critical mass” of
demand in the service territory, to improve economic viability (this requirement is also echoed
by Kalumlana et al,11 and Aguilera et al,12).
Co-operatives and Associations
Experience in many countries` shows that using existing co-operatives and associations with a
good reputation, is more effective than trying to establish new structures. These groups have
more on the ground experience, are in better contact with the people and so know the socioeconomic situation of their areas quite well.
However, care should be taken to assess the capabilities of their officers, and provide training
where necessary. In a successful project in Honduras, Enersol worked with an existing coffee
co-operative, COMARCA, after the former had conducted training13. This approach can lead to
prejudice though towards communities with strong organisation, leaving out potential customers
or deserving recipients, in other communities. In cases where using existing associations is not
possible, new organisations can be set up. In the Pacific Islands, the Tuvalu Solar Electric Cooperative Society was established after failed attempts to involve the utility14.
Similarly, NGOs may be very effective in programmes since they are normally familiar with the
communities they serve on more than just a monetary basis. NGOs have the further advantage
of having established extension networks, which could be used for training and information
services. Stone et al15 reports another institutional possibility; religious missions. The US
Renewable Energy Laboratory (NREL) co-financed a project with the Indian government in
West Bengal. The Ramakrishna Mission reliably managed the project that is now completed. So
reports Stone about the Mission as a project partner, “The Ramakrishna Mission has been
perfect in this respect”.
ECN-C--00-047
23
Conclusion
Experiences differ from country to country but some conclusions can be drawn for the
conditions for successful project design. The best role of public authorities seems to be one of
market catalyst without excessive interventionism. Experience in many countries` shows that
using existing co-operatives and associations with a good reputation, is more effective than
trying to establish new structures.
24
ECN-C--00-047
6.
PROPOSED ACTIVITIES FOR 2000
In general, we propose to continue with the activities as described in the proposal, formulated
medio 1999 and accepted on July 5th 1999. In this proposal, four activities were stated:
1. monitoring with data loggers;
2. Field survey with questionnaire;
3. Socio-economic and institutional analysis; and
4. Laboratory tests in relation to lifetime aspects of solar home system equipment.
The fourth activity will be postponed to 2001, to benefit from the Novem project: “Ontwikkelen
en uitvoeren van levensduurtesten van Solar Home Systemen”, which will be conducted from 11-2000 to 31-5-2001.
For 2000 the following activities are planned:
1. The draft survey questionnaire will be expanded with questions relevant to life-cycle
analysis and socio-economic aspects;
2. An updated version of the questionnaire will be tested in Swaziland and Indonesia in a few
tens of households;
3. Parallel to the pre-survey, data loggers will be installed in rural households in Swaziland
and Indonesia;
4. A household survey will be conducted in Swaziland in about 300 households;
5. Analysis of socio-economic impacts and institutional arrangements for SHS-dissemination
in South Africa, Indonesia and possibly in Zimbabwe.
ECN-C--00-047
25
ANNEX 1 DRAFT SHS PERFORMANCE QUESTIONNAIRE
A. Location: ..................................................
F. Questionnaire no. : .......
B. Enumerator no .......
G. Date of interview :..........
C. Name Household ................................................................................................
D. Interviewee :
1.
head
2.
wife of head
E.
Number
empty huts)
3.
4.
son
daughter
of occupied
structures/huts
no. structures...........................
5. other.......................
..............................
on
the
homestead
(exclude
storage,
Specifications Solar system
1.
When was your solar system installed?
year...............
month................
2.
From whom did you buy the solar system?
Name supplier
.................................................
City, town
.................................................
3.
What does your solar system consist of ?
component
1. Solar panel
2. Battery
3.
4.
5.
6.
7.
8.
9.
Battery regulator
Battery box
Inverter DC/DC
Inverter DC/AC
Cabling
Mounting material
Lights
10.Light switches
11.Connection for TV
12.Connection for Radio
13.Other___________
4.
5.
26
no.
type / brand name
capacity
Wp
solar bat......................
car bat........................
Ah
Ah
Watts
Watts
Fl. tube......................
PL tube......................
..................................
toggle / pull switch
Watts
Watts
Did you buy:
1. a complete solar system at once
2. first bought a battery and at a later stage added a solar panel and other
components
3. don’t know
How was the system paid for?
1.
cash
2.
credit from the solar supplier
3.
loan from the bank, indicate name of bank...........................
4.
loan from friends, relatives
ECN-C--00-047
5.
other, indicate…………………………………………………….
6.
Who installed the solar system for you:
1.
installed by the supplier
2.
installed by a technician (other than the supplier)
3.
did the installation yourself
7.
What are the solar panels mounted on:
1.
roof (grass thatched)
2.
roof (corrugated iron)
3.
roof (roof tiles)
4.
pole next to the house
5.
outside wall of the house
6.
other...................................
Use of a solar system
8.
2.
3.
4.
5.
6.
7.
What do you use the solar system for ?
1.
Lighting
Radio
B&W TV
Colour TV
VCR
Hifi
other………………………………..
9.
Do you use a battery regulator?
1.
do not have a battery regulator (go to question 11)
2.
use the regulator all the time
3.
bypass the regulator every now and then
4.
by passes the regulator all the time
5.
don’t know
10.
Do you regularly look at the indicator lights of your regulator?
1.
don’t have indicator lights
2.
never look at the lights
3.
frequently look at the lights
4.
don’t know
11.
Are you happy with the performance of the solar system?
1.
yes
2.
no, why ...................................................................................................
.............................................................................................................................
............ .................................................................................................................
12.
Are you planning to expand your system soon?
1.
yes
2.
no
ECN-C--00-047
27
13.
What would you do if grid electricity would become available in your area?
1.
apply for a grid connection and re-sell the SHS
2.
apply for a grid connection and keep the SHS (use both of them)
3.
not apply for a grid connection and keep on using the SHS
4.
other.................................................................................
Battery Performance
14.
1.
2.
3.
4.
5.
15.
If it is flat, how long does it take the battery (on average) to get charged by the
solar panel?
1.
charged the following day
2.
longer, indicate...............................................................................
16.
1.
2.
3.
17.
Do you often have a flat battery?
almost daily
once or twice a week
once or twice a month
a few times per year
never (go to question 18)
Have you ever charged your solar battery other than with the solar panel?
yes, at a charging station
yes, other .......................................................................................
no (go to 18)
How often do you charge the battery other than with the solar panel?
1.
once or more per week
2.
once or twice per month
3.
a few times per year
4.
once every one to two years
18.
3.
19.
1.
2.
3.
4.
Was the battery ever topped-up with water
1.
yes, topped-up by you or somebody else in the family
2.
yes, topped-up by a technician
no (go to 20)
If yes, what kind of water was used?
water from the tap
boiled water
distilled water
don't know
20.
Did you ever replace the battery?
1.
yes
2.
no (go to 24)
21.
If yes, after how much time did you have to replace the battery?
1.
less than 6 month
2.
within one year
3.
within two years
4.
within three years
5.
within four years
6.
within five years
7.
after more than five years
22.
From whom did you buy the new battery?
1.
from the supplier you bought the solar systems from
2.
from another battery supplier
3.
other.....................................
28
ECN-C--00-047
23.
What did you do with the old battery?
1.
returned it to the supplier
2.
threw it away
3.
other...............................
Appliances Performance
(Check question 8 whether lights are being used, if no go to question 28)
24.
How many lights are powered by your solar system: no. ……..
25.
Did you ever replace one or more tubes/bulbs of you solar lamps
1.
yes, no of lights replaced: ..............
2.
no (go to 27)
26.
From where did you obtain the replacement tubes/bulbs
1.
spare tubes/bulbs were provided by the supplier of the solar system
2.
bought it from the supplier of the solar system
3.
bought it somewhere else.
27.
1.
2.
3.
Are you satisfied with the brightness of the lamps
yes
no, not bright enough
no, too bright
Repair and Maintenance
28.
1.
2.
Did you every had a problem with your solar system?
yes
no
29.
If yes, please indicate what kind of problem(s)?
1.
2.
3.
4.
5.
6.
7.
8.
9.
faulty battery
faulty lights
low power output
maintenance is expensive
no spare parts available to repair system
system struck by lighting
theft
vandalism
other, …………………………………….
30.
What did you do to solve the problem(s)?
……………………………………………………………………………………………………
……………………………………………………………………………………
31.
How often did you call in a technician to check/repair your system in the past 12
months?
no. of times:……………….
32.
Did you ever clean the solar panel?
1. yes, did that once
2. yes, do that regularly
3. no, never cleaned it before
ECN-C--00-047
29
6.1.1.1 Information
1.
2.
Did you receive a user manual from your supplier?
yes
no (go to question 35)
1.
2.
Do you consider the manual useful?
yes
no, why not...............................
33.
34.
35.
Did the installer explain the use of the solar system to you?
1.
yes
2.
no
36.
1.
2.
37.
Would you recommend a solar system to other people?
yes, why................................................................................................
..............................................................................................................
no, why not...........................................................................................
..............................................................................................................
Have other people shown an interest in your solar system?
1.
yes
2.
no
END OF QUESTIONNAIRE
30
ECN-C--00-047
ANNEX 2: DATA LOGGER HARDWARE
At this time there are three options to obtain the required data logger for Solar Home Systems.
The possibilities are listed below:
1) Buy a complete unit (commercial of the shelf, COTS).
2) Develop the current prototype to a production ready model.
3) Develop an upgraded version of the current prototype.
Sample interval
Storage interval
Average battery voltage
Average module current
Average load current
Average battery current
Memory data retrievel methode
Memory capacity (registrations)
Power supply via solar home system (W)
Battery life included battery
Dimensions hxlxd (mm)
Electronics
Sentry of Big Ben
of ECN logger
Squirrel of Grant
ECN Logger
Upgraded version
Comparison table data logger features
Current version of
While developing the prototype of the data logger we realised the need for further
improvements such as the use of more than four channels. At the same time an answer is formed
to the question if data loggers available on the market are applicable to suit our needs. In the
table ‘Comparison table data logger functions’ below the functions per data logger are listed.
0.1 seconds
0.1 seconds 1 sec - 250 days 1 sec - 250 days
0.1 sec - 24 hour 0.1 sec - 24 hour 1 sec - 250 days 1 sec - 250 days
yes
yes
yes
yes
yes
yes
internal shunt external shunt
yes
yes
internal shunt external shunt
yes
yes
internal shunt external shunt
Memory module Memory module RS-232/PCMCIA
RS-232
86000
172000
256000
40000
0.2
0.2
no
no
no
no
2 years
5 years
40x100x60
40x100x60
40x150x80
25x104x56
As shown in the table, there are functional differences between commercially available data
loggers and ‘our’ data loggers. When we do not make the requirements regarding dimensions
and memory too strict, commercially available data loggers are suitable for our purposes.
In first instance option 2 (Develop the current prototype to a production ready model) and
option 3 (Develop an upgraded version of the current prototype) are investigated to estimate the
costs per unit. Additional, the costs from COTS (commercial of the shelve) components is under
investigation (option 1). The amount of money paid for the COTS components is set to HFL
1440,- based on the quotations of CaTeC for 20 pieces of the Grant 401. It is very likely the
price per COTS unit can be further reduced. The cost breakdown for the ‘current version‘ and
the ‘upgraded version’ are listed in the table on the following page.
The figure ‘cumulated costs for data loggers’ shows the different options in one picture, based
on rough cost estimates. From the figure it’s clear that a student is (as expected) by far the
cheapest, however considering the factor time a student is not the best option. The quickest way
to obtain a ready to use data logger is to use COTS components instead. A big advantage is the
risk due to the make-the-prototype-production-ready does not exists for COTS components.
ECN-C--00-047
31
For small numbers of ‘sites to monitor’ COTS components will be the quickest and cheapest.
For larger quantities it is cheaper when our prototype monitor module is made production ready.
The break-even point between COTS and ‘home-made’ units is considering HFL1440,- for
COTS approximately 70 units but negotiations are expected to bring the price down to
HFL1000,- to HFL1200,-.
Figure Cost comparison of different options for producing or buying data loggers (for cost
assumptions, see table on the next page).
Cumulated costs for data loggers
140000
120000
Total costs in Guilders
100000
80000
Current version
Upgraded version
COTS
Upgraded by student
60000
40000
20000
0
0
10
20
30
40
50
60
70
80
90
100
Units
Conclusion:
At this time the best way to obtain the required data loggers for monitoring of Solar Home
Systems is to buy COTS components. With COTS components the risks are minimised for both
financial risk and delivery time risk. If monitoring on larger scale is required (e.g. more then
150 data loggers) continuing development of the upgraded version might be an interesting
consideration taking into account the lessons learned from a smaller project with COTS
components.
32
ECN-C--00-047
Kosten raming monitor module vervolg
Ontwerp+prototyping
Prijs per uur (gulden)
Current version Upgraded version
180
180 gulden/uur
Componenten invoeren cad
Ontwikkelen/aanpassen ontwerp
Schema tekenen
Review schema
Ontwikkelen/aanpassen software
Pcb preplacement
Aanmaken shapes
Bestukken prototypen
Testen protypen
Aanmaken bedradings-schema's
0
40
24
4
0
16
8
16
16
8
Totaal uren
TOTAAL KOSTEN MANUREN ONTWIKKELING
Calibreren en testen per print
KOSTEN CALIBRATIE PER PRINT
16 uur
64 uur
24 uur
4 uur
160 uur
16 uur
8 uur
16 uur
16 uur
8 uur
132
23760
+
332 uur
59760 gulden
1
180
1
180 gulden
Produktie kosten
Uitbesteden bestukken
EXTERNE PRODUKTIE AANLOOP KOSTEN
KOSTEN EXTERNE PRODUKTIE PER PRINT
7500 gulden
25 gulden
Bestukken print intern
Per print
AANLOOP KOSTEN INTERNE PRODUKTIE
KOSTEN INTERN BESTUKKEN PER PRINT
4 uur
1500 gulden
720 gulden
Integratie printen in behuizing
Prijs per uur (gulden)
Componenten
Behuizing
TOTAAL KOSTEN ONDERDELEN PER PRINT
Totaal kosten vervolg huidig ontwerp
Machinale produktie
Vaste kosten
Variabele kosten per print
Handmatige produktie
Vaste kosten
Variabele kosten per print
Totaal kosten vervolg upgraded ontwerp
Machinale produktie
Vaste kosten
ECN-C--00-047
Golf-solderen
MTS'er extern
3 uur
KOSTEN MONTAGE per print
PCB
AANLOOP KOSTEN PCB PRODUKTIE
Produktie kosten per print
Eventueel stagiair laten uitontwikkel
en print ontwerp laten maken
Uur per unit
100 gulden/uur
Montage print in kast
Onderdelen kosten
Opmerkingen
(In elkaar zetten kast, aansluitkabel
(drie per print (aan de ruime kant))
300 gulden
Prijs
5 weken lead time
incl. Print ontwerp
Volgens offerte eerste prototype
5000 gulden
50 gulden
35 gulden
20 gulden
105 gulden
Ontwerp
23760
23760
Ontwerp
59760
Schatting Mark F35,00 per print
produktie integratie pcb+onderdelen
calibratie
Totaal
7500
25
300
5000
105
=
180 =
36260
430
1500
720
300
5000
105
=
180 =
30260
1125
produktie integratie pcb+onderdelen
7500
5000
calibratie
=
72260
33
ANNEX 3. REFERENCES
1
R.H. Acker and D.M. Kammen, The quiet (energy) revolution, Analysing the dissemination of
photovoltaic power systems in Kenya, Energy Policy, vol. 24, NO.1, PP. 81-111.
2
Op. Cit. Page 99.
3
P.E. Lasschuit, Review of the PV market in Swaziland, Evaluation of Government PV Demonstration
Project, Report: ECN-CX—98-018, January 1999.
4
A. Reinders et al. Sukatani revisited: on the performance of nine-year old solar home systems and street
lighting systems in Indonesia, Renewable and Sustainable Energy Reviews, 3 (1999) 1-47.
5
C.M. Ribiero et al., Performance evaluation of about 800 PV systems in the Northeast of Brazil after
one year of operation, paper presented at the 13th European Photovoltaic Solar Energy Conference, Nice,
France, 23-27 October 1995, pp. 1081-1084.
6
K. Preiser, P. Schweizer and O. Parodi, “Balde de Leyes – the integrated way to electric light”, 13th
European Photovoltaic Solar Energy Conference, 23-27 October 1995, Nice, pp 1787-1790.
7
A. Reinders et al, op. cit.
8
UNESCO, “World Solar Programme 1996-2005”, September 1997.
9
See web page of Shell, 1998.
10
Smith, P. “RETSs for Rural Electrification – Some reflections”, Renewable Energy for Development,
Vol. 11, No. 1, April 1998.
11
Kalumiana, A. Arvidson, “Establishing Photovoltaic Energy Service Companies in Rural Areas”,
Renewable Energy for Development, Vol. 11, No. 2, November 1998.
12
J. Aguilera, E. Lorenzo, “ Rural Photovoltaic Electrification Programme in the Bolivian High Plateau”,
Progress in Photovoltaics Vol.4, pp. 77-84, 1996.
13
CADDET webpage: http://www.caddet-re.org/html/techpv.htm.
14
A. Cabraal, M. Cosgrove-Davies, L. Schaeffer, “Best Practices for Photovoltaic Household
Electrification Programs”, The World Bank Technical Paper No. 324, 1996.
15
J. Stone, H.S. Ullal, “The Ramakrishna Mission PV project”, published on PV resources website.
34
ECN-C--00-047
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement