Although tanks were included in the testing program in 1999, testing effectively started only in early 2000. The program tests tanks with nominal capacities of 100, 200, 300, 400, 500, 600, 800 and 1000 liters. Table 3 presents the tests that are performed. At the beginning only the three first items listed were tested. In 2003, however, the last two items, related to the safe operation of the electrical resistance, were added to the program. The safety tests were introduced due to the fact that almost all tanks sold in the residential market have electrical auxiliary heating.

One important difference between the label for collectors and the one for tanks is that, in the case of tanks, there is no performance ranking of the equipment tested. The tanks are simply classified as approved or not approved according to the standards specifications. Figure 3 shows a sample tank label. It has only basic information as model, serial number, manufacturing date, volume, maximum operating pressure and electrical specifications. At the bottom, the label states that the product has been approved by INMETRO, and it has the logos of the Brazilian electricity savings program (PROCEL) and INMETRO.

In the case of the volumetric capacity, the tank is approved if the measured capacity is neither 10% higher nor 5% lower than its nominal capacity.

For static pressure, a testing pressure 1,5 higher than the declared maximum working pressure is applied for 15 minutes. The tank is then inspected for leaks and deformations.

In the case of the thermal efficiency the standard used, ISO 9459 — Part 2: item 9.9, was adapted to a situation where the tank is tested separately, and there is no forced convection around the tank. After the cooling test is performed, the heat loss coefficient is calculated. With this value it is then possible to estimate the monthly losses for the tank. The heat losses calculations are executed assuming the hot water temperature in the tank at 50 oC and the ambient temperature at 21 oC. The maximum acceptable losses, which were revised in August 2003, are listed at Table 4.

RESULTS

The labeling program was initiated by the main SWH manufacturers in Brazil. After the release of the first results, these companies started including information about the label in their advertisements and brochures. That was compounded by a small campaign in national television by the government agency INMETRO, urging consumers to buy only tested and labeled products. These actions brought, in 2000, the first increase in the number of companies participating in the market, from the initial 8 companies in 1998 to 12 companies in 2000. The year 2001, with the electricity crisis in Brazil and a booming market for SWH, marked the beginning of an even greater expansion in the number of participating companies.

By the end of 2001, 15 companies were participating. The crisis prompted the Brazilian government to offer financing for SWH through one of its banks, but only for equipment that had already been labeled. Moreover, the publicity gained by the sector through many media reports during the crisis was another opportunity to reinforce the importance of the label, and in 2002 the number of companies participating reached 25. In March 2004 there

were 35 companies participating in the program. Figure 4 shows the evolution of the number of participating companies.

The growth in the number of participating companies brought the challenge of testing a significant number of collectors. Up to the end of 2003, 96 collectors had been fully tested. As a way to give the companies a label before the full tests are completed, a mechanism called a pre-label was introduced. In order to obtain a pre-label, the collector is first tested for thermal performance (Group 2, Table 1). If the company already has a model using the same materials in terms of the glazing and fins surface finishing, the label is first released based on the information collected by the thermal performance tests and the behavior of the similar model previously fully tested. When the tests are finalized, the label is corrected, if necessary. Figure 5 shows the number of collectors fully tested for each year and the cumulative number until the end of 2003.

On top of the 93 models already fully tested, another 9 have received the pre-label and are waiting to complete the tests. Another 51 collectors are still waiting to be tested. As

these numbers indicate, the smaller number of collectors tested in 2003 was not caused by lack of demand, but by poor weather conditions.

1998 1999 2000 2001 2002 2003

Figure 6 shows the percentage of collectors in each category for each year of the program, to the end of 2003. Although there is not a strong trend, it is interesting to note that the number of collectors classified as A first increased from 0 to 33 % and then gradually decreased until 2003, before increasing again. This is believed to be due to the fact that once the collectors are tested, the manufacturers improve their collector models. The appearance of new participants in the market brought a higher number of collectors with lower classifications, such as D and even G.

In the case of the tanks, there was a similar growth in the number of models tested in 2001 and 2002, with a decrease in the number of models tested in 2003, as can be seen at Figure 8. The reason for this decrease, however, is different in relation to tanks than for collectors. The decrease in the number of tanks tested was caused by the introduction of the new standards related to electrical safety. Since the manufacturers that had already tested their models had until January of 2004 to get new tests performed, most preferred to wait, while adjusting and developing their products to meet the new standards.

At the beginning, most of the tanks that failed had problems with the volumetric capacity test. The percentage of models approved rose from 42% in 1999 to 78% in 2002. In 2003, with the introduction of new tests, the percentage of approved models fell back to 50%. It is expected that this number will increase again in 2004, with the testing of models already adapted to the new standards.

CONCLUSIONS

It is possible to conclude that the labeling program, although not a complete quality assurance program, has had a significant impact on the Brazilian solar water heating market. Most of the collectors commercialized in the country have been labeled and the main manufacturers have all developed better products as a result of the testing process.

The labeling has also enabled the participation of solar water heaters in government tender processes and created a minimum standard to be used in government programs.

At the same time, consumers have been given simple and reliable information to help with purchasing decisions.

At the present time, a sizable backlog in collector testing remains. In response, the Brazilian government has sponsored the installation of a solar simulator at GREEN SOLAR. The simulator is expected to be operational in the second half of 2004.

As is the case with many labeling programs, the program has been gradually improving the quality and performance of the equipment evaluated. This so-called "ratcheting” dynamic is already evident in relation to both collectors and tanks. The program would be significantly improved with some kind of testing and certification of SWH installers, since in Brazil the installation process has a particularly strong effect on the final efficiency of the system.

The authors would like to acknowledge the support of the Companhia Energetica de Minas Gerais (CEMIG) and Centrais Eletricas Brasileiras (ELETROBRAS).

REFERENCES

[1] ASTM E823-81(2001) Standard Practice for Non-operational Exposure and Inspection of a Solar Collector, American Society for Testing and Materials, 2001.

[2] FSEC-GP-5-80, Test Methods and Minimum Standards for Certifying Solar Collectors, Florida Solar Energy Center, USA, 1985.

[3] ASHRAE 93-86 Methods of Testing to Determine the Thermal Performance of Solar Collectors, American Society of Heating, Refrigeration, and Air-Conditioning Engineers, Inc., Atlanta,1986.

[4] ASHRAE 96-1980 RA 1989, Methods of Testing to Determine the Thermal Performance of Unglazed Flat-Plate Liquid-Type Solar Collectors, American Scoiety of Heating, Refrigeration, and Air-Conditioning Engineers, Inc., Atlanta,1989.

[5] INMETRO/GT-SOL, Regulamento Especifico para o Uso da Etiqueta Nacional de Conservagao de Energia — ENCE — Sistemas e Equipamentos para Aquecimento Solar de Agua, Brasilia, 2003.

[6] ISO 9459 Part 2, Solar Heating — Domestic Water Heating Systems; Performance Testing for Solar Only Systems, CEN, 1994.

[7] NBR NM IEC 335 -1, Seguranga dos aparelhos eletrodomesticos e similares: Parte 1 — Requisitos Gerais, Associagao Brasileira de Normas Tecnicas, 1998.

[8] NBR 14016 — Aquecedores instantaneos de agua e torneiras eletricas — Determinagao da corrente de fuga — Metodo de ensaio, Associagao Brasileira de Normas Tecnicas,1997.

[9] NBR 14013 — Aquecedores instantaneos de agua e torneiras eletricas — Determinagao da potencia eletrica — Metodo de ensaio, Associagao Brasileira de Normas Tecnicas,1997.