Quando uma ciclo de refrigeração por compressão de vapor é analisado, uma das variáveis mais importantes é a vazão de fluido refrigerante que circula pelo sistema. Ela resulta do equilíbrio entre a operação do compressor e do elemento de expansão. O tubo capilar é o elemento de expansão normalmente utilizado em sistemas de refrigeração de pequeno porte como os do refrigeradores, freezers e aparelhos de ar condicionado de janelas domésticos. Muito embora seja muito mais simples e muito mais barato do que o compressor, o tubo capilar tem praticamente a mesma importância do que aquele na determinação da vazão de refrigerante. O tubo capilar é ainda hoje dimensionado por métodos empíricos e semiempíricos, apesar dos grandes avanços havidos nos campos da mecânica dos fluidos, teoria de escoamento bifásico, transferência de calor e cálculo numérico. O método ASHARAE de seleção foi desenvolvido para tubos capilares adiabáticos com R12 e R22. O método pode ser adptado de forma grosseira para tubos capilares não adiabáticos considerando que na região de troca de calor existe apenas escoamento de fluido refrigerante em fase líquida. Até 1991 havia muito poucos trabalhos tratando de tubos capilares não adiabáticos. Quando a presente pesquisa foi iniciada, não havia nenhuma publicação analisando o tipo de trocador de calor tubo capilar- linha de sucção utilizado no Brasil (trocador de calor de tubos concêntricos). Um programa de simulação matemáticado escoamento de refrigerantes através de tubos capilares não adiabáticos e adiabáticos é desenvolvido no presente trabalho. Ele permite tanto o projeto como a simulação do desempenho (cálculo de vazão mássica) de tubos capilares. A versão de simulação é adequada para ser utilizada em um programa que simula globalmente o desempenho de refrigerado ou freezer. O modelo matemático utiliza o modelo homogêneo para o escoamento bifásico. ) A questão da estabilidade de cálculo foi considerada com cuidado, uma vez que trabalhos anteriores mencionavam alguma dificuldade no tratamento deste item. Para validação do modelo matemático foi construído uma unidade experimental de avaliação do desempenho de tubos capilares adiabáticos e não adiabáticos. Ela não utiliza um ciclo de refrigeração mas, ao contrário, foi concebida para permitir ensaios em lotes. Como o compressor não é utilizado, pode-se realizar ensaios com fluido puro, sem óleo.

One of the most important variables to be investigated in vapor compression refrigerant cycles is the refrigerant mass flow rate through the system. It is mainly determined by the balance3 between the compressor and an expansion device. Small systems often use a capillary tube (CT) as expansion device. Despite its lower cost and simplicity when compared with the compressor, CT is essential for determining the mass flow rate of the system. Since refrigerating CT were conceived, their design has employed empirical or semiempirical methods. This procedure is still applied, in spite of the progress achieved in the fields of fluid mechanics, two-phase flow theory, heat transfer and numerical calculus as well as the recent spread of computational resources. ASHRAE curves for CT sizing are suitable only for adiabatic flow. Such a graphical method was developed for Just two fluids: R 12 and R 22. For non-adiabatic CT, ASHRAE sizingmethod is roughly accomplished, because it assumes only liquid phase in CT flow through the heat exchanger region. Very few works dealing with non-adiabatic CT were published until 1991. At the time this research begun there were no work dealing with the kind of heat exchanger (HX) manufactured in Brazil, e.g., concentric tubes HX. In the present work a computer program for simulating the flow through adiabatic and non-adiabatic CT, was developed. The program may be used for CT sizing as well as for CT simulation (e.g., mass flow calculation). The later version is suitable to be included as a routine of a more general refrigerator/freezer simulation program. Homogeneous two-phase flow model is assumed. In addition, special care was taken to ensure the stability of calculation, since previous works had reported some difficulty to accomplish it. To validate the theoretical model an experimental testsection was built. Basically, it is a blow-down apparatus, where refrigerant, stored in a high pressure vessel, flows through a CT and is releases in another vessel at a lower pressure, That design avoids the use of a vapor compressor which could lead to refrigerant contamination with oil. Vapor refrigerant that flows through the suction line in an actual refrigerator or freezer was substituted by air. The air flow rate was set in such a way that no modifications of thermal and dynamic characteristics of the flow were introduced. As thew experimental apparatus runs in a batch process (not through a cycle) and because air is the substance used in the suction line, all the independent variables can be finely controlled. About 250 tests for adiabatic and non-adiabatic CT were performed. Several operating conditions combining different CT inlet pressures and temperatures as well as suction line inlet temperatures were tested. Refrigerant mass flow rate was the main dependent variable. Its values were continuously recorded by a data acquisition program. In addition, CT and suction line temperatures throughout the device, and CT inlet and outlet pressures were also recorded by the same device. All the experimental results for adiabatic and non-adiabatic CT with R 12 and R 134ª fluids were compared to the respective numerical results and a close agreement were obtained. In some tests, theoretical and experimental temperature profiles were also compared leading to relatively low deviations. Experimental results have shown that adiabatic and non adiabatic CT present different behavior regarding to the stability of the flow. This hás been related to thedelay of vaporization. This phenomenon is often present in CT lows and a physical model for this process is proposed. Finally, a numerical sensitivity analysis for CT performance was performed. Among other variables, the simulation program allowed the evaluation of mass flow rate dependence on the CT inlet pressure, subcooling degree, CT diameter, length and type of heat exchanger, and heat transfer to the ambiance.