On the Determination of the Effective Thermal Capacity of Solar Collectors

W. Eisenmann, F. Pujiula, H. Koln
Institut fur Solarenergieforschung GmbH Hameln/Emmerthal (ISFH)

Am Ohrberg 1, 31860 Emmerthal

Tel.: ++49-5151 / 999-521; Fax: -500, E-Mail: w. eisenmann@isfh. de

1 Introduction

The determination of an effective thermal capacity of solar collectors is necessary for the description of time-dependent processes. The capacity is important for the theoretical simulation of the yield of solar thermal systems (mostly on a daily or yearly basis, e. g. for methods to guarantee or control solar yields). The procedures for the determination of the effective capacity described in EN 12975 give strongly varying results. Depending on the collector type this leads to more or less pronounced differences of the simulated yield, where higher capacities mean lower yields.

This paper investigates which of the procedures for determination of the capacity is most appropriate for the description of dynamic processes, with respect to daily or yearly collector gains.

2 Distributed Capacity and Models with Several Nodes

The basic difficulty of the determination of an effective capacity is that the thermal capacity is spread over the components of the collector, and that the thermal coupling with the temperature fluctuations of the fluid differs from one component to the other. For example, the absorber is generally in close thermal contact with the fluid, while the thermal coupling between the glass cover and the fluid is fairly weak.

Consequently, it would seem appropriate, at least from a scientific point of view, to develop collector models with more than one node (at least one node for the absorber and one for the fluid), see e. g. [2].

Nevertheless, for practical reasons one-node models should be preferred for the present revision process of EN 12975 and for collector tests. One reason is that there is no widespread experience with several-node models. Furthermore, the determination of the parameters of these models would most probably be quite difficult and inaccurate.

All the procedures for the determination of the collector capacity described in EN 12975 are based on one-node models.