In the third of our series on lighting, we look at how the type of light source influences the design of the lamphead and the light produced.

Softly, Softly...

In the previous article I described the three most generally used types of light source: tungsten, metal halide and fluorescent. This choice inevitably imposes certain characteristics on the lighting instruments themselves. Fluorescent sources emit light from a coated surface, therefore a large area is required to produce a useful amount. They are inherently 'soft' sources because they create comparatively 'soft' shadows (see Hard and Soft Light below). Although some manufacturers have produced luminaires that, to an extent, concentrate the light to give emphasis to the subject, it is not possible to duplicate the effect that can be produced from say, a tungsten spotlight.

The Balcar Fluxlite a 375W Fluorescent Softlight. Courtesy of Pyser SGI Ltd.

Such tubes must have a large surface area, but do not require any great depth. Fittings can be designed to have large front areas but be only a few centimetres thick. These are very convenient for use against a wall or ceiling where for example, they blend into an office or domestic setting and appear in shot without attracting attention. Very little heat is produced - an advantage when close to the subject as they would ideally be, or in locations which do not have the level of air conditioning expected in a purpose-built studio.

A tungsten filament light source is obviously much more compact than a tube. Lamps can be produced in many configurations, which allows the luminaire manufacturer to select an appropriate type for the optical system of the head. However, the lamphead designer has to contend with the heat produced relative to the amount of light. Tungsten lamps produce in the region of 20 lumens of light per Watt of electricity compared to 80lm/W from a fluorescent or metal halide lamp, the additional power being wasted as heat.

Soft glass lamps of high power had to be physically large because they suffered from progressive blackening as tungsten vaporised from the filament and condensed onto the comparatively cool glass. A small envelope would have darkened more quickly. From the 1960's tungsten halogen lamps with quartz glass envelopes were introduced. These included a halogen such as iodine or bromine, which combine with the tungsten vapour at the envelope and re-deposit it onto the filament - the halogen cycle - to prevent blackening of the glass.

The capsule size could now be reduced by using a stronger, quartz glass. This allowed the internal working pressure of the lamp to be higher which reduced the evaporation from the filament and allowed the manufacturers more flexibility when balancing longer life against higher colour temperature and light output per Watt.

If running the lamp at less than the rated voltage, for example, dimmed down, even though less tungsten is evaporated from the filament, blackening may still occur because the lamp will not be hot enough for the halogen cycle to operate. In this situation, the lamp can be cleaned by running for a short time at normal voltage. Blackening can also be caused by over-running, which increases the evaporation of tungsten beyond the level that can be re-deposited by the quantity of halogen present.

ARRI Junior 5kW Tungsten Spotlight

Many halogen lamps operate at higher than atmospheric pressure. Although they are pressure tested by manufacturers, the risk of explosion should be borne in mind. Safety standards require that the luminaires are designed so that particles larger than 3mm cannot escape under gravity, and no part can escape directly in the event of an explosive outcome. Open-faced units should always be used with an appropriate safety glass or mesh.

In some cases, the user can make a trade-off between lamp life and colour temperature; a choice of lamps may be available within a range of 2900-3400K. For example the 500W K1 linear lamp has a life of 2000 hours and a colour temperature of 2900K, and can be used as a substitute for a 1000W P2/20 with a life of 300 hours and a colour temperature of 3200K.

Metal halide lamps emit light from a cloud of ionised gas (known as a plasma). The size of this cloud can be as small as 1.2mm in a 21W lamp. Such a comparatively small size allows for a more efficient optical design since the source is more nearly the ideal point source. The first designs of metal halide lamps were double-ended, linear lamps. These were long so that the seals were far enough from the heat of the plasma to be reliable. In order to ignite the lamp, a high voltage (up to 65kV depending on lamp size) had to be introduced across the electrodes. To accommodate these lamps and allow space around the contacts so that the high voltages could not arc to the housing, the lampheads had to be of large diameter. One positive effect was that because the housing had to be large, large diameter lenses were utilised with a long focal length, which made the source effectively smaller and gave a better light quality.

The introduction of single-ended metal halide lamps overcame many of the shortcomings of double-ended lamps and allowed a dramatic reduction in the size of the lamphead. The new technology solved many of the old difficulties and even gave some additional benefits that will be discussed in the next article.

Hard and Soft Light

If an 'ideal' point source of light is used to illuminate an opaque object, the result will be a sharp-edged black shadow surrounded by a brightly lit area. This is a hard shadow and a point source can be considered to be a hard source.

If the source is not a point, but has a significant size, then the black area on which no light falls will be smaller and it will be surrounded by a partly lit area (the penumbra) which receives light from some parts of the source but not all. The edge of the shadow will be more difficult to identify as there will be a graduation from dark to light. The larger the source, the less clearly defined would be the shadow. An 'ideal', infinitely large, soft source would produce no shadow, but would illuminate the subject and the wall behind completely evenly.

A hard source can easily be made softer by shining it through an opalescent material such as a frosted gel. The gel itself in effect becomes the light source, with light coming from every illuminated part. Moving the frost nearer the subject and further from the source would increase the softening effect. Bouncing the light off a white material such as expanded polystyrene can produce a similar effect.

A source can be made harder by moving it further from the subject or increasing the effective throw by reflecting from a mirror surface. However, to use a softlight as a hard light by using it far from the subject would be pointlessly inefficient because an unnecessarily large area would be illuminated.

Andy Barnett, ARRI GB 2002

next >>