M. Cucumo — A. De Rosa — V. Ferraro — D. Kaliakatsos — V. Marinelli

Mechanical Engineering Department — University of Calabria 87030 Arcavacata di Rende (CS) — Italy Phone +390984494603 — Fax +390984494673 E-mail: m. cucumo@unical. it

More than 23,000 experimental hourly illuminance data on variously orientated horizontal and vertical surfaces were compared with some Perez prediction methods.

Whereas the illuminance data on the horizontal plane are predicted optimally, notable differences between the calculated and experimental data were observed for the vertical surfaces.

The authors propose a simplified calculation method which proves to be in better agreement with the experimental data.

1. INTRODUCTION

The quantitative knowledge of natural illuminance energy inside rooms is very useful when aiming to save energy. The starting datum for this calculation is obviously natural incident illuminance on the outside surfaces of windows and this paper deals with this topic.

For more than two years a network of instruments for measuring natural illuminance has been in operation at the Mechanical Engineering Department of the University of Calabria. This network is composed of five photometers to measure illuminance on the horizontal plane and on four vertical planes orientated to the north, south, east and west, of a photometer provided with a guard ring to measure diffuse illuminance on the horizontal plane, and a photometer mounted on a solar tracker to measure direct illuminance.

The photometers for measuring global light is type FET-GV, the photometer for measuring diffuse light are type FET-S, and the photometer for measuring direct light are type FET L03 0U DX, supplied by PRC Krochmann of Berlin, while the solar tracker, type 2AP (two axis positioner), was supplied by Kipp & Zonen of Delft (Holland).

In the paper the global illuminance values, recorded at Arcavacata di Rende (CS), on the five measuring planes, in the period 01/01/2001-5/11/2002, were compared with some calculation methods developed by Perez.