The role of spectral response of photosensors in daylight responsive systems

https://doi.org/10.1016/j.enbuild.2007.04.010Get rights and content

Abstract

Lighting control technologies using photosensors have a great potential for energy savings in areas with high levels of daylight. Although the proper application of these controls can exploit this potential, unfortunately, it has been accomplished in a small percentage of new projects. One reason is the difficulty in justification of energy savings, which in turn, is directly linked with the simulation of the behaviour of these lighting controls. The core of these systems is the photosensor, which adjusts the electric light output in proportion to the amount of the daylight that detects, using its spatial and spectral response. The aim of this study is to quantify the impact of photosensor spectral response on its illuminance values, by taking into account various daylight spectra as these are modified due to various types of coloured glazing. Five commercial photosensors were selected and their spectral response was measured. In addition, spectral transmittance of 16 commercial types of glazing was measured as well. Using these data, a set of simulations were performed using three colour channels in a typical office room and the relative differences in illuminance – and thus energy savings – among the photosensors are presented. The results show that differences are significant ranging from 36 to 118%, a fact that can affect the estimated payback period of a lighting control system.

Introduction

Despite the great potential for energy savings in areas with high levels of daylight and case studies documenting energy savings from dimming [1], [2], [3], [4], [5], [6], lighting control with photosensors have not been widely installed by building contractors. This is because of the added cost of the equipment and installation labour. In addition there is a perception that automatic dimming controls are unreliable, although, various studies have described their proper functioning [7], [8], [9], [10], [11], [12], [13], [14], [15], [16].

Accurate computation of daylight, accurate simulation of the performance of the photosensor and reliable simulation of the artificial lighting system output in relation to the control voltage are needed for the prediction of the performance and its effects on energy use [9], [10], [17], [18], [19], [20]. The basic operation of a photosensor is the production of a signal that is related to the amount and the distribution of lighting in a space in which it is placed. The performance of the photosensor can be complex because it depends from a lot of variables, such as:

  • the distribution of daylight and artificial lighting in the space in which it is placed;

  • the spectral composition of lighting;

  • the adjustment settings of the commissioning control;

  • the ambient light level;

  • the field of view.

The behaviour of the photosensor in response to variable lighting, blind and/or glazing conditions is significant in the design of a daylight responsive system. Fig. 1 shows a photosensor system diagram [21]. Its optical response is consisted of two main sub functions:

  • Spatial response: The spatial response describes the sensitivity of the photosensor to incident radiation from different directions, in other words, what the photosensor “sees.” Spatial response is analogous to a luminaire intensity distribution but describes sensitivity instead of light output.

  • Spectral response: The spectral response describes the sensitivity of the photosensor to optical radiation of different wavelengths.

Ehrlich et al. [22] have presented a method of simulating the photosensor behaviour based on the multiplication of two fisheye rendered images, one generated with the actual angular sensitivity of the photosensor while the other with an 180° field of view of the space as “seen” by the photosensor. Analyzing the final image, photosensor illuminance can be calculated accurately. For the estimation of energy consumption, functions of control voltage, light output ratio and consumed power of the lighting system are also needed. These can be estimated using experiments or using the manufacturer's data [23], [24], [25]. Other studies [9], [15], [16], [17], [24], [26] have also simulated the photosensor behaviour taking into account only the spatial response of the photosensor.

However, the photocells used in photosensors are sensitive to a wider range of wavelengths than what the human eye sees. In other words, photocells respond to portions of the ultraviolet (UV) and infrared (IR) spectrum as well as the visible spectrum. Filters that incorporated into the photocell element, limit the sensitivity to ultraviolet (UV) and infrared (IR) radiation. Bierman and Conway [21] examining the behaviour of the photosensor found that a high performance control algorithm could possibly, but not likely, compensate for both a non-ideal spatial response and a poor spectral correction. On the other hand, they concluded that the better the spatial response tracks change in working plane illuminance and the better the spectral correction of the photosensor, the finer will be the overall operation of the photosensor system. Their conclusions differ from the commonly held belief of lighting practitioners that photosensor system performance depends primarily on a product's spatial response.

Simulation programs (DAYSIM [27], SPOT [28]) which have been developed for the estimation of the electrical lighting savings, can take into account the spatial response for the photosensor but they use the CIE photopic luminous efficiency function or one colour channel. Daylight's spectral power distribution is ignored although there are some efforts for the creation of a comprehensive model which, using simple parameters, the complete characterization of the light coming from the sky vault (including its spectral distribution) can be estimated [29], [30].

Without taking into account the differences in components and factors that affect a daylight responsive dimming system someone can be led away from the initial expectations, miscalculating the energy savings and thus the subsequent payback period. Recent studies have showed that commercial photosensors with different spatial and spectral response can affect differently the accuracy of the performance of such a system [13], [17], [21]. The proper placement and control algorithms of the photosensor are very significant factors that must be considered during the initial design stage [7], [8], [9], [13], [21]. The optimum choice of ballast from a large variety of electronic dimmable ballasts with different characteristics is also required, because the differences in energy savings can be significant [23], [24], [25]. Furthermore, improper daylight conditions during calibration procedure [26] together with the presence of shading and/or daylighting systems can turn the system in poor performance [31], [32].

However, the effect of spectral response in a daylight dimming system has received little attention. The present paper compares the performance of a number of photosensors with various spectral responses, installed in a typical office test-room with different types of glazing, in terms of estimated illuminance.

Section snippets

Photosensors

Five commercial photosensors, from three different manufactures, were tested. These photosensors were considered as separate devices that can work with any electronic dimmable ballast models having a 0–10 V DC signal.

The photosensors have been codified using two digits. The first digit is the number of the photosensor (1–5) and the second digit corresponds to the manufacturer of the photosensor (i–iii for three manufacturers). For example, the 3ii code corresponds to the third photosensor

Simulations

Simulations were performed using Radiance software [36] which has been used in previous photosensor studies [9], [15], [16], [22]. The room (Fig. 8, Fig. 9) that was used for the simulation is a typical space in an office building and has been used in the past for IEA task 27 and SWIFT projects [37], [38]. The windows were located on the south facade of the building. The photosensor was placed in the geometrical center of the room mounted on the ceiling, oriented with its maximum response

Conclusions

Given nowadays extensive availability of glazing types with various optical properties and daylighting control systems, spectral simulation of illuminance is extremely important.

Lighting control systems are a complex technology that changes rapidly. A variety of controllers, software, sensors and devices are currently available, but there is lack of information concerning the actual performance of these systems and control strategies. In order to fully exploit their capabilities and implement

Acknowledgements

The project is co-funded by the European Social Fund (75%) and National Resources (25%)—Operational Program for Educational and Vocational Training II (EPEAEK II) and particularly the Program PYTHAGORAS II.

References (43)

  • D.G. Kaskaoutis et al.

    Checking the validity of the Angstrom's formula with spectral data of direct-beam irradiance obtained in Athens, Greece

    Atmospheric Research

    (2006)
  • D.G. Kaskaoutis et al.

    Comparison between experimental data and modeling estimates of aerosol optical depth over Athens, Greece

    Journal of Atmospheric and Solar-Terrestrial Physics

    (2006)
  • G.M. Giavis et al.

    Diurnal variation of the aerosol optical depth for two distinct cases in the Athens area, Greece

    Atmospheric Research

    (2005)
  • M. Szerman

    Superlink, a computer tool to evaluate the impact of daylight-controlled lighting system onto the overall energetic behaviour of buildings

  • R. Embrechts et al.

    Increased energy savings by individual light control

  • K. Opdal et al.

    Energy saving in lighting by utilisation of daylight

  • F. Rubinstein et al.

    Improving the performance of photo-electrically controlled lighting systems

    Journal of the Illuminating Engineering Society

    (1989)
  • R.G. Mistrick et al.

    Analysis of daylight photocell placement and view in a small office

    Journal of the Illuminating Engineering Society

    (1997)
  • A.S. Choi et al.

    Analysis of daylight responsive dimming system performance

    Building and Environment

    (1999)
  • NLPIP, Specifier Report: Photosensors, 6 (1)...
  • F. Rubinstein et al.

    On the calibration and commissioning of lighting controls

  • Cited by (46)

    • Operation guidelines for daylight dimming control systems in an office with lightshelf configurations

      2020, Building and Environment
      Citation Excerpt :

      The analysis results of this study are based on the limited configurations of the lightshelves and photosensors applied in a small office space. These analysis results generally consist of the results of other studies, although different window configurations, photosensor conditions, and shading devices are used [43–47]. In the other studies, several shielding conditions and moderate photosensor spatial patterns used for the photosensors functioned effectively to maintain the target illuminance in a given space.

    • Minimizing energy consumption for artificial lighting in a typical classroom of a Hellenic public school aiming for near Zero Energy Building using LED DC luminaires and daylight harvesting systems

      2019, Energy and Buildings
      Citation Excerpt :

      Daylighting affects both well-being and energy consumption [72–84], thus more lighting energy savings can be achieved when daylight harvesting systems are installed. Possible glare issues are solved through the use of shading - daylighting systems [14,15,72,74,85–90]. Today there is the ability to calculate the potential lighting energy savings at an early stage of design before the installation of any light control [91,92].

    View all citing articles on Scopus
    View full text