J. Jaramillo, E. Mas de les Valls, C. Oliet and J. Cadafalch

Centre Tecnoldgic de Transferencia de Calor (CTTC)

Lab. de Termotecnia i Energetica Universitat Politecnica de Catalunya (UPC) labtie@labtie. mmt. upc. es, www. cttc. upc. edu

Introduction

In the solar domestic water heating sector, the thermosyphon systems have achieved a good acceptance due to its relatively good efficiency, low installation and maintainance cost and due to the absence of mobile parts. As a result, these systems have become a really interesting device in order to exploit solar energy.

Thermosyphon systems typically consists of a collector and a storage tank which are mounted together outdoors. The outlet of the collector is connected to the top of the storage tank, and the inlet of the collector is connected to the bottom of the storage tank. When the water in the collector is heated up, density of water in the collector decreases with respect to the density of the water in the tank. When this difference of the density is high enough to overcome the resistance to flow forces (friction, expansions, contractions, other singular­ities…), circulation starts. During the night, water in the collector is cooled dawn and the density of the water in the collector increases. To avoid inverse circulation of hot water in the tank to the collector due to this fact, a check valve is normally installed in the pipe connecting the outlet of the collector to the tank.

Numerical studies on the designs of thermosyphon solar heaters are very often based on simple mathematical models as in [9, 14, 6]. In these models, thermal and fluid flow behaviour of the different parts of the thermosyphon cycle are evaluated by means of ex­perimental data (as the steady state efficiency curve of the collector) or unidimensional or zero-dimensional energy and momentum balances. All parts are linked together forming the cycle, and are solved by means of an iterative procedure.

According to the huge computational resources available nowadays, it is possible to go a step further in the modelling of thermosyphon solar heaters by making use of CFD codes. These codes are based on the numerical resolution of the multi-dimensional governing equa­tions (energy, momentum, mass, radiative heat transfer). A large number of control volumes would be required to discretize a whole thermosyphon system, therefore, CFD codes are not still suitable to do it. However, this kind of high level simulation is essential to get deeper insight into the physical phenomena of the flow and heat transfer in characteristic parts of the system and in order to calculate specific data that may be required by more simplified models such as heat transfer coefficients and friction factors. An example of CFD studies developed on solar collectors using CFD techniques can be found in [4], a flat plate trans­parently insulated cover is modelled solving the energy conservation low and the radiative
transfer equation. Another example of a detailed simulation of the components of a ther­mosyphon solar system can be found in [3]. There, the storage tank is studied by means of three-dimensional and transient CFD simulations of the temperature, pressure and velocity field (resolution of mass, momentum and energy equations). From this simulations conclu­sions on the influence of the inlet mass flow rate on the degree of thermal stratification are drawn.

The authors are currently using different simulation levels for the analysis of thermosyphon systems, they are divided in three groups, i) Simplified models; ii) Intermediate models; iii) CFD models. This papers describes the three levels of modelling focusing on the CFD mod­els. Some illustrative examples of results that can be achieved with these models are also