The thermal performance of the collector described in Table 1 was measured in an outdoor test facility where the inlet temperature, the outlet temperature and the volume flow rate was measured. The temperatures were measured with copper — constantan thermocouples (Type TT) and the volume flow rate was measured with a HGQ1 flow meter. A 31% glycol/water mixture was used in the solar collector loop. Further, the global radiation and the diffuse radiation on horizontal were measured with two Kipp&Zonen CM5 pyranometers.

The collector performance was measured for

two different tilts: 45° and 90° (both facing south). A period of 11 days (17/5-28/5 2003) has been selected for validating the Trnsys model for the collector at 45° and a period of 7 days (12/8-19/8 2003) has been selected for validating the Trnsys model for the collector at 90°.

The necessary data for describing the collector are shown in Table 1. The heat loss coefficient, k0, was determined from efficiency measurements (Shah, L. J. & Furbo, S. (2004)) and split into two parts for the evacuated tubes and the manifold pipes respectively. F’ was calculated from theory (Duffie J. A. and Beckman W. A. (1991)), (Incropera F. P. and de Witt D. P. (1990)) and (ra)e and a were calculated with a simulation program for determining optical properties (Svendsen S. and Jensen F. F. (1994)).

In Fig. 6 the measured and calculated collector outlet temperatures are compared. It can be seen that there is a good degree of similarity between the measured and calculated temperatures. Further Fig. 5 shows the measured and calculated collector performance for the two periods. The difference between the measured and calculated performance lies