In the second of this series on lighting, we look into some of the ways light can be produced.

The Source.

If you are faced with the prospect of putting together a lighting kit, investigating the huge variety of equipment available from many different manufacturers can be a painfully confusing process. Quoted performance specifications tell you about how much light is produced but do not say much about how controllable a source is or what kind of shadows will result. Manufacturers will say their product is wonderful but not really tell you what it is for. The aim of the next few articles is to describe the principles of the various types of luminaire and, hopefully, this will make it possible to categorise them and more easily compare what is on offer.

The most fundamental aspect of any luminaire is how the light is actually produced. The most familiar technology is incandescent or 'tungsten' lighting in which a metal filament is heated up by an electric current until it glows. This is basically the technology that makes a domestic light bulb work, although photographic lamps are rather more sophisticated than general service lamps. Most other sources fall into the category of 'discharge lighting' which rely on a mixture of chemicals to emit light when they are stimulated by a source of energy, generally an electric arc. For photographic purposes, the most useful types of discharge lights are: 'fluorescent', where the light is produced by a layer of chemicals coated onto the inside of a glass tube and 'metal halide', also referred to as HMI or MSR, where light is produced by a gas at high temperature. Other forms of discharge lighting exist, as do other methods of producing light, but those listed above are most widely used for general video and photographic lighting applications.

Intensity Distribution of Various Colour Temperature Light Sources

The light produced by discharge and incandescent lights is different in its nature. Incandescent light is made up of a spectrum of colours from infra-red upwards. Early experiments measured the light from cannon balls heated in a furnace: as the cannon ball is heated, it begins to emit infra-red. Heated further, it glows first red then orange, yellow and white as the bluer end of the spectrum is added to the red. Because the red colours are still present, the light becomes increasingly whiter, rather than blue. As a result, the temperature of a lamp filament can be used as a guide to the 'whiteness' of the lamp. The 'colour temperature' of a domestic lamp, giving a more reddish light, would operate at perhaps 2800K, but whiter photographic lamps generally operate at 3200K; the 'K' referring to the temperature scale named after Lord Kelvin.

In the design of filament lamps there is a trade-off between colour temperature and lamp life - the hottest, 3400K lamps only last a few hours and the filament is close to melting when in use. Running a filament lamp hotter still would produce something on the lines of a flash bulb where a very white light is given for a fraction of a second as the filament vaporises. Conversely, reducing the energy to the lamp by using a dimmer, for example, would cause a reduction in colour temperature.

Intensity Distribution of 4kW HMI Lamp. The white curve shows the radiant energy of the sun over Central Europe. Courtesy of Osram Ltd.

The physics of discharge lamps is rather different. Each of the chemicals producing the light emit only small numbers of very specific colours, an example being the characteristic orange of the sodium lamp. In order to produce light that is useful for filming, lamp manufacturers have to use a mixture of chemicals. The combination of different individual colours mix together to form something approaching the continuous spectrum produced by an incandescent light source.

The ability to mix a cocktail of colours has some interesting advantages. By choosing an appropriate combination of chemicals, discharge sources can be made which produce a much higher proportion of visible light and less infra-red (ie heat) than incandescent lamps. So for a given electrical power, a discharge source can produce perhaps four times as much visible light as an incandescent one, and proportionally less heat. Discharge lights can also be manufactured to give much higher values of colour temperature than would be possible from a filament lamp, in particular, giving the equivalent of 5600K light which matches the characteristics of daylight film stock.

In metal halide lamps, these chemicals are added into the mixture present inside the lamp capsule and are vaporised when the lamp is operating to form a cloud of glowing gas. In fluorescent sources, the same effect is achieved by varying the constituents of the coating on the tube. With different coatings, it is easily possible to manufacture tubes that produce light of different colour temperatures and even different colours. Simply by changing the tubes in the luminaire, it is possible to switch from the 3200K to 5600K light.

Although it is common to refer to the colour temperature of discharge sources, it is not scientifically accurate. More relevant to users is that most types of colour temperature meter can be confused by these types of sources. The meter tries to estimate based on its measurements of specific colours. If there happens to be a high or low level of those particular colours present, then the reading can be drastically inaccurate. True readings can only be obtained by averaging over the complete spectrum. A simpler method of spotting incorrect colour is to compare the suspect source with two others. It will be easy to see if two match and the third does not. In this situation, the eye can be a better judge than a meter.

The design of luminaires and the way they can be used is partly imposed by the nature of the light source itself. In future articles we will look at the ways light sources can be used to best advantage in different types of luminaire.

By Andy Barnett, ARRI GB.

next >>