The establishment of a dedicated UESP test site at Fraunhofer ISE has started. Beside the demonstration of UESP systems it will be used for further development of components and system tests. At a later stage, components will here undergo intensive compliance testing before they are able to carry a UESP conformity label. Within the project six different types of power supplies will be set up and tested, many of them in the field. The following table shows the various applications with information on power and energy:

3. The Consortium

Eight German companies have teamed up with Fraunhofer ISE in this challenging project (funded by the German Ministry for the Environment, Nature Conservation and Nuclear Safety, BMU) to define the previously described protocol and interfaces and to elaborate the various implementations. The applications range from village power supply systems to highly reliable telecommunication stations and signalling systems. The business‘ interest ranges from component manufacturers to potential users or operators. The research

institute’s role lies in the development and implementation of the concept and the central software. It is foreseen to establish UESP as standardised interface in the future.

A. Abete, R. Napoli, F. Spertino

Politecnico di Torino, Corso Duca Abruzzi 24, 10129 Torino (Italy)

Tel. +39-11-564.7105 Fax +39-11-564.7199 Email filippo. spertino@ polito. it

Introduction — PhotoVoltaic (PV) plants are already economically viable in off grid sites (mountains, islands and rural areas), as for solar home systems, or where the grid is far some kilometres, as for road lighting. Technical problems are still present and different standards have been proposed for quality assurance, i. e. the “Universal Technical Standard for Solar Home Systems”: it is a design guideline for manufacturers and installers, which gives requirements classified as compulsory, recommended and suggested; especially they deal with batteries, identified as the most costly component during the system lifetime (operation and maintenance), and charge regulators [1]. The simulation of battery behaviour in PV systems is an usual topic and several mathematical models have been developed with different degrees of complexity: the challenge is to find a good compromise between complexity and precision [2]. This question is still existing, because often models consider battery charges and discharges only at constant current, contrary to what occurs in the stand alone PV systems with various types of load (household appliances, motors, lamps, etc.). With reference to a stand alone PV system for road lighting, this paper deals with firstly the experimental characterisation of batteries by a cycle of discharge-charge and then the simulation of yearly system behaviour by the battery parameters for determining reliability and black-out hours.

Solar Tergo is a charger for small "personal products," such as mobile telephones and portable CD — or MP3 player, for use in combination with a Boblbe-e backpack. It is designed by Bernadette Weitjens in co-operation with Energy research Centre of the Netherlands. When the Solar Tergo is exposed to light it charges a battery pack which is used, at its turn, to charge the product that is to be charged in the first place. A prototype of the Solar Tergo (see figure 6) has been realised, based on a Unisolar US 3 flexible triple junction cell, 8 NiCd AA cells, a SAA1501T Battery charge controller and a set of LEDs for battery level indication. In order to determine the performance of the Solar Tergo it is equipped with a Grant Squirrel 800 data-logger, a Kipp & Zonen SP-lite pyranometer as well as thermo-couple. In February 2004 the first data have been gathered with this "personal laboratory." The Solar Tergo is suitable for outdoors conditions.

During these experiments data is gathered simultaneously with the portable Solar Tergo and with a stationary installation that is placed on a roof within a range of five kilometres of the Solar Tergo. In these experiments the Solar Tegro’s PV-cell has been isolated from the rest of the system. During the experiments the irradiance and temperature on the Solar Tergo are measured every 5 or 10 seconds. Simultaneously the open voltage and the voltage over a 0,1 Ohm resistor that are generated by the Solar Tergo’s PV-cell are measured alternately. Switching between the two modes takes place every 30 seconds, while a measurement is done every 5 or 10 seconds. The stationary installation that is laced on the roof consists of a Kipp & Zonen SP-lite pyranometer, a Unisolar US-3 flexible triple junction cell and a Grant Squirrel 800 data-logger. Here the irradiance and the voltage over a 0,1 Ohm resistor that is generated by the PV-cell are being measured every 10 seconds in a horizontal plane. The shortcut current is calculated from the voltage that is generated over the 0,1 Ohm resistance. The output power and efficiency of the PV-cell is calculated from the specified operating voltage of the PV-cell, the calculated shortcut current and the surface area of the PV-cell.

The samples that are presented here are taken on 18 and 31March 2004. Although the Solar Tergo was logging all day long, for both days a shorter sample is isolated and edited. Both samples cover a period of time in which the Solar Tergo is used while biking in an urban environment in the Netherlands. The 18 March sample is taken from 10.11 to 10.41 (winter time) under heavy overcast conditions. The 31 March sample is taken from 15.00 to 16.00 (summer time) under practically cloudless conditions. During this activity the panel inclination is about 90°.

It is clear that reliable figures can not be derived from two samples. Nevertheless, the results of the experiments are presented in order to give an idea of the output that is generated. Average values of each of the quantities that are measured by the Solar Tergo are calculated per minute of time. Table 1 shows the mean value and the standard deviation of those figures. Moreover, mean value and standard deviation of the ratio 0ST/0Roof of irradiance values and the ratio IST/lR00f of the shortcut currents are displayed. These ratio are defined as the "portable irradiance” over the "stationary irradiance,” respectively "portable shortcut current” over the "stationary shortcut current.” The ratio IST/IRoof is in fact the same figure as the "Power Ratio” that is calculated in Experiment one.

Figure 7 and 8 show the irradiance as measured on the roof and on the Solar Tergo during the two samples. The mean value of the ratio 0ST/0Roof is smaller under cloudless conditions than under heavy overcast conditions, while the standard deviation is larger. Logically this difference is the result of variation in shadow and variation in the angle of the PV-cell towards the sun in the horizontal plane, which occurs while biking around. Figure 9 and 10 show the shortcut current that was generated by the cell on the roof as and on the Solar Tergo during the two samples. The identity of shape of the figures 7 and 9, and 8

and 10, in combination with the figures of table 1, suggest that the output of both cells can simply be deducted from the irradiance.

In order to verify whether the relation between the output and the irradiance is equally clear for portable applications as for stationary applications, a scatter diagram is drawn. Figure 11 displays the relation between the irradiance and the shortcut current, for both the Solar Tergo (portable) and the roof installation (stationary). This diagram is based on both the data of 18 March and 31. The diagram suggests that there is not much difference in this relation between the data that is produced by the Solar Tergo and the roof installation. Variation might be a result of factors such as spectral distribution or cell temperature.

Discussion

When comparing the two experiments that are described above, one notes that they differ in the equipment used and in the samples that are studied. The shoulder bag compares the output of PV-cell on a portable device with the output of a stationary PV-cell. The Solar Tergo does the same, but adds information on the conditions under which this output is generated. One can also imagine a piece of equipment that only logs conditions. If the relation between irradiance and shortcut current proves to be equal for portable

applications as for stationary devices, than data on the conditions on product level in combination with data on performance on component level can theoretically serve as input for sizing procedures and energy balance simulations. This is attractive since irradiance data is universally applicable while data gathering on component level takes place under laboratory conditions. In some cases data on component level can be taken from the PV — cells specifications. Therefore it is useful to study the relationship between irradiance and shortcut current for different conditions and activities. The Solar Tergo is equipped to study this relation for the Unisolar cell. The first experiments with the Solar Tergo suggest that, for this cell, the relation between irradiance and shortcut current is similar for portable applications as for stationary applications. Note that an approach as sketched above, that leaves the spectral distribution of light out of consideration, might be valid for outdoors application. For indoors applications spectral distribution is likely to play an important role, particularly in relation with multi-junction cells.

The shoulder bag experiment shows that there is a very large variation in output when a 24-hour sample is taken, for example in terms of the "Power Ratio." This implies that figures based on such a sample provide little support if it comes to taking design decisions, for example based on calculation of probability. For a group of products it will be possible to use more specific scenarios as basis for design decisions, for example because they are usually used in a specific environment and/or under specific conditions. Therefore data that is gathered on product level while controlling some variables, as is done in the Solar Tergo experiment, will prove useful.

Frank Fiedler, Svante Nordlander, Tomas Persson, Chris Bales
Solar Energy Research Center SERC, Dept. of Mathematics, Natural Sciences
and Technology, Dalarna University College, S-7188 Borlange,

Phone: +46 (0) 23 77 87 11, Fax +46 (0) 23 77 87 01, ffi@du. se

Abstract — Various pellet heating systems are marketed in Sweden, some of them in combination with a solar heating system. Several types of pellet heating units are available and can be used for a combined system. This article compares four typical combined solar and pellet heating systems: System 1 and 2 with a pellet stove, system 3 with a store integrated pellet burner and system 4 with a pellet boiler. The lower efficiency of pellet heaters compared to oil or gas heaters increases the primary energy demand. Consequently heat losses of the various systems have been studied. The systems have been modelled in TRNSYS and simulated with parameters identified from measurements. For almost all systems the flue gas losses are the main heat losses except for system 3 where store heat losses prevail. Relevant are also the heat losses of the burner and the boiler to the ambient. Significant leakage losses are noticed for system 3 and 4. For buildings with an open internal design system 1 is the most efficient solution. Other buildings should preferably apply system 2 or 3. The right choice of the system depends also on whether the heater is place inside or outside of the heated area. A large potential for system optimisation exist for all studied systems, which when applied could alter the relative merits of the different system types.

KEYWORDS: Pellet heating systems, heat losses, flue gas losses, leakage losses

1. Introduction

A variety of system concepts for solar heating systems for new one — and two-family houses can be found on the European market. In Sweden electrical heaters are often used as the auxiliary heat source but the use of wood pellets in pellet stoves and pellet boilers is becoming more and more popular. Sweden is already today the most developed European market for pellet heating systems for one-and two-family houses. Approximately 30000 pellet heating units have been installed by the end of 2001 (Hadders, 2002). There are several Swedish manufacturers active on the market but also a number of manufacturers from other countries. For this study two Swedish, one Finnish and one German product, all marketed in Sweden, have been investigated.

The efficiency of pellet heaters varies significantly between different manufactures and designs and is generally lower than for comparable gas or oil boilers (Fiedler, 2004). For this reason the heat losses from these units are important parameters to evaluate the complete heating system.

2. Method

The systems investigated in this work are considered to represent typical solutions within a wide range of design variants. The systems are for the most part taken as they can be found on the market whereas system 4 is somewhat an exception. This system is
not available as a complete system but at least the pellet boiler and the store are standard products.

This study uses the results of comprehensive measurements on two pellet stoves, one pellet boiler and a storage integrated pellet burner at the Solar Energy Research Center SERC. Several tests have been performed to evaluate the performance of these units and to acquire data enabling an identification of characteristic parameters (Nordlander and Persson, 2003). The investigated system designs have been modeled in the simulation environment IISiBat/TRNSYS (Klein et al., 2002) based on a simulation model created for IEA task 26 Solar Combisystems (Bales, 2003). The same house model (Streicher and Heimrath, 2003) and other standard boundary conditions have been applied.

For the modeling of the pellet heating units TRNSYS type 210 that has been recently developed by Nordlander (2003) has been used. This dynamic model can be used to simulate pellet stoves, pellet burners and to a certain extent also pellet boilers and gives flue gas losses during operation and in standby mode (leakage losses), as well as heat supplied to water in a mantle and to the surroundings. The applied parameters are obtained from the parameter identification and have been validated by simulations comparing simulated and measured data. The models for the pellet heating units have been integrated into the system models and yearly simulations have been performed for a building with an annual heat demand for the Stockholm climate of approximately 12200 kWh (87 kWh/m2) and a hot water demand of 3100 kWh. The results from these simulations have been analyzed and used to compare the four systems.

Solar powered water pumping has the potential to bring sustainable supplies of potable water. [1]

For irrigation of a large number of small agricultural areas, remote from power grid, solar cell systems ranging from 300 W to a few kW may be used. In our estimation we calculate with an average electric power of solar pumps of 500 Wp. For example, an experimental 500 Wp solar "drop by drop” water pumping system, has been working near to Becej (North Serbia), for more than 10 years. [2]

Considering existence of necessary stimulating bank-credit terms for private persons, realization of 400 of these solar systems with total capacity of 200 kW can be expected already in 2006. year.

With an annual growth rate of about 60, 39, 6.5 and 5.3% respectively, the following dynamics of solar pumps application in next four years is anticipated:

Rapid increase of solar pumps application for irrigation in 2007. and 2008. years, would be result of initial market euphoria of most of domestic consumers, mainly individual farmers (truck farmers). Severe slowing-down of this dynamics in the years to follow, would be caused by relative saturation of domestic market.

5 kWp will be installed on this drug rehabilitation clinic that is run by the Vest- Agderklinikkene (a leading institution for alcoholic rehabilitation). It has 2.200 m2 area with room for 38 tenants. The PV modules will be integrated in an aluminium faqade in the southeastern corner of the building and as sunscreens in the southwestern faqade.

3.5 PV-parking, the Netherlands

30 kWp of amorphous PV modules will be integrated in the southwest and southeast facing faqades of a parking building. The colour and the relatively low price per sqm have led to the choice of using a-Si modules (thin-film amorphous silicon) in this project. This gives an important contribution to PV-NORD, as all other demonstration buildings will be using crystalline silicon cells.

3.6 Shell office building, the Netherlands

40 kWp of multi-crystalline PV modules are integrated in a "PV Pergola" — a glass cover over an open space — of a new Business and Technology Centre in Rijswijk. The project gives valuable input through the architectural integration in the building design.

3.7 Kollektivhuset, Denmark The Danish Kollektivhuset building is used as an example of the principles discussed above, notably the functional and aesthetical building integration of PV modules. It merits its own heading, see below.

4.1 Consultation with Stakeholders: Consultations with a wide range of stakeholders in South India, which included SHS vendors, banks and other small scale financing institutions, governmental agencies, experts, NGOs etc., were carried out during the planning phase of the programme. The findings from stakeholder discussions indicated that there is a strongly felt need for SHS in areas with non-existent or erratic electricity grid. Strong motivators for the use of SHS were children’s education, other lighting needs, and TV. A lack of availability and the cost of credit emerged as two major barriers to the wide scale adoption of SHS by potential buyers. It was also observed that although banks had enough credit available and were seeking new loan products, they were not yet ready to treat SHS as a standard technology. Since it is not a standard product in banks’ lending portfolios, and the loan size is small, transaction costs for the bank are thus high, reducing its attractiveness further. But the banks were ready to experiment and explore the possibility of scaled up lending for SHS. It also became apparent from the consultations that the poor, who have least access to electricity, could benefit from a lending facility through SHGs.

Given that for renewable energy technologies access to finance has been widely considered a barrier, discussions with stakeholders also included feedback on a variety of finance support mechanisms including front-end / back­end interest rate subsidies, credit default guarantees, loan term extensions, beneficiary margin support, and subsidised transaction costs. According to both vendors and banks, an interest rate subsidy would help address many of these barriers and in so doing would accelerate SHS adoption by providing bank loan managers the incentive to promote the

product and by offering consumers more attractive terms to purchase the product.

The stakeholder consultations indicated the need to address risk perceptions in the financing community regarding SHS loans. The high up-front cost of SHS, the high cost of credit and a lack of awareness among potential users were other barriers identified during the consultations.

4.2 Selection of Programme Partners: The stakeholder consultations were carried out in three southern states of India but Karnataka was finally selected to implement the programme in the initial phase. Karnataka has a strong rural banking system with a ready­made platform for credit delivery. Several vendors are based, and have manufacturing facilities, in Karnataka, including three of the most reputed names in the renewable energy industry in India — Tata BP Solar, Shell Solar and Selco India. Vendors already have after­
sales service networks in Karnataka, an important requirement for market sustainability and growth.

Canara Bank and Syndicate Bank, two large Indian retail banks with networks of branches in most parts of South India, were chosen as the partner banks. The banks support Regional Rural Banks (RRBs or Grameen banks) and hence have considerable coverage in the more rural areas. They are at the forefront when it comes to launching innovative products. They provide credit to poor households through SHGs. Malaprabha Grameen Bank, one of the rural banks promoted by Syndicate Bank, has already been lending for the purchase of SHS in association with Selco, although this experience is still nascent. The banks also have the ability to replicate their success in other parts of India thanks to their national reach.

4.3 Financing Mechanism : Interest Rate Buy-Down: A number of market development models were considered during programme preparation, including providing capital cost subsidies to solar vendors, end-user subsidies directly to customers, or financing subsidies through one or two partner banks. It was determined that direct links with vendors or customers were not necessary, and that working through the banks would be the most effective approach. Experience from other programmes has shown that capital subsidies have a tendency to stick and distort the market. Bankers were also not very enthusiastic about capital subsidies and preferred interest rate subsidies, which enable them to offer preferential banking terms to their customers. Since banks were willing to take the full credit risk under the programme, the option of a guarantee facility was also ruled out. An interest rate subsidy also reduces the programme risk, whilst the possibility of a gradual reduction in subsidies ensures that it can be planned properly and withdrawn fully without significantly damaging the market.

One of the attractive features of an interest rate subsidy is that it does not distort the market, either in terms of the capital cost (i. e. the ticket price) that the customer associates with a solar PV system, or the risk that a banker associates with a solar loan. As against a new product, interest subsidies can distort the market if offered for a mainstream financial product. The provision for reviewing and revising the subsidy in the programme is to ensure that the interest rate subsidy is gradually eliminated.

The interest subsidy approach allows the partner banks to offer loans to customers at

concessional rates of interest, initially about 7% below their prime-lending rate of 12%. A corpus of USD 0.9 million can fund interest subsidies for loans to finance approximately

18,0 to 20,000 SHS. As these subsidies phase out over time, the actual number of SHS financed might in fact be larger, depending on the timing of the phase-out.

The programme, by providing loans with an interest rate buy-down, addresses the ‘high up-front cost’ and high credit cost, the barriers identified by stakeholders. The programme is expected to help increase awareness and confidence in SHS technology, bring down the financing costs of the technology in India, and widen the market. Worldwide advances in PV technology and decreasing costs are expected to usher the SHS market in India to levels where it can be sustained without further support.

4.4. A Multi-Vendor — Free Market Approach: Since a number of experienced solar rural electrification companies already existed in Karnataka, it could have distorted the market to choose, or tender for, one vendor over the others. Furthermore, working with a single vendor would require the use of narrowly defined and monitored system specifications to ensure that the ‘chosen’ vendor actually delivered a quality product. This heavily regulated approach could restrict the vendor/customer relationship, leaving little room for vendors to innovate in product/service offerings and for consumers to choose the system most appropriate for their needs and budgets. It was therefore decided to work with as many vendors as could meet the product quality and after-sales service criteria. The vendor panel has been kept open and new vendors will be included as and when they are able to meet the criteria. The partnership with only two banks was considered in view of the size of the subsidy corpus — a large number of banks would have left hardly any incentive for the banks due to the small amounts of business. Thus, the ‘two banks — multi vendor’ approach was deemed to be the best possible, free-market oriented approach for the programme, making use of competitive forces to ensure quality products, competitive pricing and reliable after-sales support.

4.5 Technical Support and Awareness Raising: An important component of the strategy was to provide support to the banks by providing standard specifications for SHS equipment and inputs for appraisal of the product at branch levels. Since the technology and the products were new to the banks, a further cushion was provided through the vendor qualification process so that banks can ensure quality products and reliable after­sales service. This addressed the problem of risk perception and also reduced transaction costs in processing the loans.

During the product launch, training programmes were organised by the banks for their branch managers, and some support and inputs to the training were provided. The window for training support has been kept open for banks, and they have already indicated the need for such support whilst they incorporate their Grameen banks into the
programme. In addition to training, support for village level meetings between banks, vendors and potential customers has also been provided. Both the banks have been organising village meetings regularly to increase awareness among potential customers. In addition to this, banks and vendors have been carrying out other promotional activities to raise awareness. Support has also been provided to banks to meet expenses related to programme promotion activities, which reduces the cost for promotion by the banks. However, these support activities are quite small, on average 300 rupees per loan approved (or roughly 2 per cent).

4.6 Reaching the Poor: The programme focuses mostly on rural and semi urban populations and also aims to reach the poorer households. To deepen the reach in rural areas, the extension of the programme through their RRBs (Grameen banks) is in progress. Efforts are also underway to reach the poor through the involvement of SHGs and NGOs working with the poor. One such initiative is in the pipeline to be supported by the Small-Scale Sustainable Infrastructure Development Fund (S3IDF).

4.7 Feedback: Periodic customer surveys are used to gain feedback from customers, and periodic audits for relevant checks are part of the programme design. These are used to take corrective actions. A recent audit indicated the need for some corrections, which have now been initiated.

It seems that all operating system with densely packed cells rely on active cooling. Verlinden et al. [16] describe a monolithic silicon concentrator module with a fully integrated water cooled cold plate. The module consists of 10 cells and is supposed to act as a "tile" in a larger array. The design is further described by Tilford et al. [17]. However, details are not given on the way in which the water flows through the cold plate. Lasich

[18]

has patented a water cooling circuit for densely packed solar cells under high concentration. The circuit is said to be able to extract up to 500 kW/m2 from the photovoltaic cells, and to keep the cell temperature at around 40 °C for normal operating conditions. This concept is based on water flow through small, parallel channels in thermal contact with the cells. The cooling circuit also forms part of the supporting structure of the photovoltaic receiver. It is built up in a modular manner for ease of maintenance, and provides good solutions for the problem of different thermal expansion coefficients of the various materials involved. Solar Systems Pty. Ltd. has reported some significant results from their parabolic dish photovoltaic systems located in White Cliffs, Australia [4, 19]. They work with a concentration of about 340 suns. An average cell temperature of 38.5 oC with a corresponding cell efficiency of 24% is maintained. If all of the thermal energy extracted were used, the overall useful energy efficiency in this system would be more than 70%. This demonstrates clearly the benefits of active cooling if one can find uses for the waste heat.

Vincenzi et al. [21, 22] have suggested integrating the cooling function in the cell manufacturing process by using a silicon wafer with microchannels circulating water directly underneath the cells. The system under consideration is run at about 120 suns. Microchannel heat sinks will be presented in more detail later. A system is patented by Horne [20] in which a paraboloidal dish focuses the light onto cells that are mounted vertically on a set of rings, designed to cover all of the solar receiving area without shading (Figure 4). In this system, the water both cools the cells and acts as a filter by absorbing a significant amount of UV radiation that would otherwise have reached the cells. It would also absorb some of the low energy radiation, resulting in higher cell efficiency and a lower amount of power converted to heat in the cells. The Horne patent incorporates a phase — change material in thermal contact with the cells, which works to prevent cell damage at "worst-case scenario" temperatures. Koehler [23] suggests submerging the cells in a
circulating coolant liquid, whereby heat is transferred from two cell surfaces instead of just one. In this way the coolant also acts as a filter by absorbing much of the incoming low — energy radiation before it reaches the cells. The coolant liquid must be dielectric in order to provide electrical insulation of the cells. By choosing the right coolant fluid and pressure, one can achieve local boiling on the cells, which give a uniform temperature across the surface and a much higher heat transfer coefficient.

Creation of large-scale manufacture of polycrystalline silicon for PV converters demands not only uses of more powerful sources of solar radiation and application of more powerful concentrating systems.

There is an imperative need of search of the new technical decisions providing an effective supply of a concentrated sunlight to a silicon for melting. The important factor is also a necessity of maintenance of maximum active interaction between volume of melt and gas — liquid border which provides active formation of volatile impurities contained substances and their intensive removal into a gas phase.

It is possible to apply for these purposes the big solar furnaces (BSF) adapted to the technological aims, such as, for example, BSF of Weizman Institute ( Rehovot, Israel) [5], BSF of the National Centre of Scientific Researches of 1000 kW (Odejo, France) [6] and BSF with 1000 kW thermal capacity of the Academy of Sciences of Republic of Uzbekistan (Parkent, Uzbekistan) [7].

In all specified installations technological operations are carried out in a special technological tower. A furnace in Odejo and a furnace in Parkent represent a design executed with a horizontal optical axis. It consists of parabolloid mirrors and heliostates systems..

One of primary goals of the BSF in Uzbekistan, in particular, is the possibility of performance of basic researches of physical and chemical processes at heat treatment of various objects by the concentrated light flow in a spectrum of a sunlight, i. e. studying of the mechanism of
influence of photons of a spectrum of a sunlight (Е = 0,4-5,0 eV, Solar radiation up to 1,7.103 Wt/cm2) on various objects.

At the same time, the structure of the available equipment provides an opportunity of manufacture of large portions (up to tens tons) materials as powders, granules or billets. It allows to use such installations both for industrial production of silicon, and for researches. Transmission of a solar energy along horizontal optical axis of BSF does not allow in this case to create a stable bath of melt at an end of an ingot as it was described in [2]. In this case also it is difficult to provide stable removing of the fused zone from focus of the solar furnace for crystallization of silicon. Another than described in [2] technical decisions for increase of the area of contact of melt volume with its surface with the purpose of an intensification of process of silicon purification by removal of impurities into a gas phase are required.

In this case the following scheme of technological process performance can be represented (Fig.2).

Fusion of silicon is carried out in the rotating drum-type furnace-crucible. The supply of a solar energy is carried out through a special aperture in the end face of a drum.

Silicon melted in the rotated furnace-crucible under solar energy influence may form shapes presented at fig.3 [8].

The most active mixing and liquid volume-surfase interraction take place for condition «b”. But in this case direct concentrated sunlight can reach open surface of internal furnace isolation and destroy it. So the case «c» is cosidered as the most appropriate for practical use [8].

The fused silicon from the rotating furnace after proccesing goes to a crystallizer.

In a crystallizer with special screens the gradients of temperature providing formation of silicon billets with columnar structure are created. Seeds at the bottom of a crystallizer provide given crystallograf orientation of big monocrystalline grains.

It is possible to expect, that the material obtained by these means will meet SOG-Si requirements.

c

Fig.3. Possible shapes of melted silicon within furnace-crucible (section across it axis) during rotation.

a — n> (K/2 n)(Vg/R); b — n<(K/2 n)(Vg/R); c- n&(K/2 n(Vg/R), where n — speed of rotation; K-coefficient; g-gravity acceleration coefficient; R-radius of internal surface of melted silicon.

The specified process equipment can be applied as well to obtaining of silicon directly from quartz by carbothermic reduction with the subsequent purification before casting in a crystallizer.

Work in this direction requires participation of the organizations having at their disposal necessary solar installations (BSF), possessing experience in creations and operation of the process equipment on silicon manufacture and on fabrication of solar cells on its basis. Realization of research projects possible on a multilateral basis within the framework of the international programmes would allow to bring in the significant contribution to development of a solar industry.

The systematic monitoring of plant parameters and the comparison of monthly yields is an essential basis for the evaluation of every PVS as well as for technical improvements for future concepts and constructions. Since few PVS in Saxony have the capacity for remote readouts, obtaining any testable data often depends mainly on the persistence and correctness of the owner of the plant.

The following results and discussions are mainly based on measurements of the PV energy fed into the grid of the regional energy suppliers. Here it is important to consider whether and how the connection to the net was established. Some PVS from the 1000- roofs-program feed also today only the excess energy to the grid (as originally requested in this program), whereas all new PVS feed all the generated energy into the grid. Usually one time per year the generated energy is registered by the energy supplier, but some plant operators register the monthly yields also.

The fundamental basis of the evaluation of the PVS results is the solar irradiation at the location, which can, however, be influenced considerably through local factors (shading, generator orientation, etc.).

The DWD irradiation measurements in Dresden were proved to be representative for Saxony [3]. In Figure 2 the monthly course of the mean value of the irradiation in Dresden is shown together with the data for the years 2002 and 2003. It is evident that in the year 2002 irradiation corresponded exactly the annual average (meteorological regular year). By contrast, this value was exceeded in the year 2003 ("century summer”) by 15%.

To compare the capacity of PV-plants, generally the normalized output power (yield, measured in kWh/kW) is used. The performance ratio PR better characterizes the design and the component properties of a PVS, but the needed irradiation measurement in the module plane is only at few PVS available.

Figure 3 illustrates the classified yields (as far as they are accurate by the data basis) of about 170 PVSs in Saxony in the years 2002 and 2003.

The depicted PV-systems clearly reflect the changed conditions of irradiation in both years. The average yield in 2003 is about 15% higher than the yield of 2002. About 30 PVS reached a yield of more than 1000 kWh/kW in 2003 (compared with zero systems in 2002). The highest yield was 1105 kWh/kW. Nevertheless, the figure also shows a number of PVS with unacceptable low yields.

Figure 4: Monthly yield in 2002 for about 60 PVS depending on the commissioning date

Different influences on the monthly and thereby also annual yields are known, and it is difficult to select any from individual output measurements. From the experiences with PVS in Saxony, the following is noteworthy: [16]

be reached. In [5] and [6], about the same amount of yield increase depending on the year of construction was reported, without known or detectable selective influencing factors.