The hemispherical dome is 4m radius, this being essentially the largest dome we could fit into the room allocated. There is a one metre deep pit underneath the dome for access to the models and for instrumentation.
The structural form used was a geodesic dome constructed using a system produced by Tecna. This was considered to be the most cost effective structural system available to us, and provided a stable and strong frame with a lot of flexibility in lighting mounting methods.
The lighting of the dome is a compromise between the desire for maximal coverage of lit surfaces (so as to achieve a smooth luminance distribution), and the desire for minimal surface area (so as to reduce inter reflections and so make luminance control easier).
640 luminaires containing low energy compact fluorescent lamps (Philips CL 4500K) are mounted within each triangle of the dome frame. Due to the geodesic structure, this lead to an irregular spacing of the luminaires, but it was felt that this could be countered by the luminance control software. The individual luminaires are between 2 and 3 degrees apart. Philips electronic dimmable ballasts are used with the luminaires; these provided a large dimming range of between 3 - 100% brightness. To achieve a greater brightness range for those positions near the sun path, selected units are double lamped.
The luminaires were developed from standard category 1 fittings, with extra baffling to control side spill and inter reflections, They were produced for us by Litetronics. Overall the dome lighting can produce up to 7000 lux at the model stage under simulated overcast sky conditions.
The heliodon structure was made to be adaptable to a number of artificial sun types. The primary source at the moment is a stage lantern, a Selecon Pacific with either a 1kW tungsten or a 575W HMI lamp. The latter gives us more light and a higher colour temperature, but is less easy to control. The Selecon provides a good even illumination over the 2m stage and gives sharp shadows, but as a "point" source, there is some divergence. This hasn't proved to be a problem to date, but we are still looking for an appropriate mirror source to provide a technical parallel beam.
A big advantage of the Selecon for general use is its "cold mirror" design, which reduces the infrared heat load at the model. At the other end of the spectrum (literally) we also use a 4kW HMI source, adapted from a Desisti stage lantern. This provides over 80000 lux at the stage and ~600 W/m2 broadband irradiation overall. This has been used for testing solar panel systems; goggles and high factor sun block are required safety equipment when it is in use.
The dome lighting and the heliodon mechanism are computer controlled, using standard industrial PC I/O interface cards under Windows 95. The interfacing and software were developed in-house, using Visual Basic.
The dome lighting is controlled though a DMX512 system. This system is an 8 bit resolution 0-10V control communications protocol. A standard in the world of stage lighting, the DMX system provided the most cost effective solution for the number of control channels needed for the dome. Individually addressable receiver channels are grouped in modules about the dome, linked in a daisy-chain by a single twisted pair cable. The "universal interface" receiver modules were produced by Pulsar. The PC based DMX transmitter (supplied by fPf Ltd.) is memory mapped, so that rapid changes to the dome lighting can be achieved. The modelling of natural lighting dynamics is an area we are planning to explore.
The heliodon uses high resolution quadrature encoders to determine altitude and azimuth of the track and turntable. Position and speed is controlled in real-time. The heliodon positioning accuracy is better than 1/2 degree, and an accelerated day can be run in 2-3 minutes.
Instrumentation is being acquired as needed by projects and research. Lighting measurement/assessment systems we currently have include: