TYPES OF LAMP

The widespread use of electric lighting began with the invention of the first practical incandescent lamp by Thomas Edison and Joseph Swan in the nineteenth century.  Since then there have been significant improvements in lamp efficiency as well as the different types of lamp.

As discussed in the overview, light sources used today in architectural lighting can be divided into two main categories: incandescent and luminescent gasseous discharge lamps. The gaseous discharge type of lamp is either low or high pressure. Low-pressure gaseous discharge sources are the fluorescent and low-pressure sodium lamps. Mercury vapor, metal halide and high-pressure sodium lamps are considered high-pressure gaseous discharge sources.

These are the most common light sources used in the field architectural lighting. Each light source will be described in terms of its three primary components: (1) light-producing element (lamp), (2) enclosure (luminaire), and (3) electrical connection.

INCANDESCENT LAMPS

In broad terms, incandescent lamps are cheap to install but expensive to run.  They can be justified if initial costs must be kept to a minimum and the annual hours of use are small or they are to be used intermittently with frequent switching.  In some cases, the effects required in display or prestige interiors may warrant the use of small incandescent sources due to the precise control possible, however they should not normally be used for the general lighting of interiors.

Features typical of an incandescent light globe.

Light is produced in an incandescent lamp by heating a thin metal wire to very high temperatures (around 2200°C), causing it to incandesce or glow. The wire is called a filament and the incandescence is a result of the filament's resistance to the flow of electrical current. Filaments are almost universally made from Tungsten as no other substance is as efficient in converting electrical energy into light on the basis of life and cost. Tungsten has four important characteristics in this regard: a high melting point, low evaporation, high strength yet reasonably ductile, and it has desirable radiation characteristics. The most common filament letter designations are straight (s), coiled (c), coiled coil (cc) and ribbon or flat (r). Coiled coil filaments are the most efficient and widely used filament type.

Different shapes and types of incandescent bulb.

The enclosure or glass envelope around the filament is called the bulb and serves two primary functions. First, the enclosure keeps air (more importantly, oxygen) away from the filament. When the filament is exposed to air, evaporation occurs very rapidly, to such an extent that the filament usually breaks within a few seconds. Secondly, the enclosure maintains a constant environment for the filament to retard the evaporation of Tungsten. The enclosure is usually filled with an inert gas such as argon and nitrogen, and comes in variety of shapes and sizes depending on its use and output requirements.

In many cases, the bulb serves a decorative function or forms an integral reflector, lens or filter. Bulb finishes are often applied in order to diffuse light from the very bright, concentrated filament when a softer source is required.  The two most common finishes are etched glass and applied silica powder.  Etched glass is known as 'inside-frosted' or simply 'frosted' and results in the appearance of a glowing ball of light within the globe.  Applied silica powder, called 'soft-white' by most major manufacturers, cuts output more than etched glass but makes the entire bulb glow more evenly.

There are a number of types of bulb colour coating in use:

Sprayed lacquers applied to the outside of the bulb are highly transparent (more efficient) but easily scratched or scuffed. 

Plastic coatings are slightly less transparent but have a high resistance to abrasion and weathering.

Transparent ceramic enamels are fused to the bulb by heat.  They are not as transparent as either the sprayed lacquers or plastic coating but are significantly more durable.

Dichroic filters are created by applying several thin coats of metallic film to the face of the lamp.  Because the film passes only wavelengths in small colour bands and reflects all others internally, the effect is slightly more efficient than passing the light through a conventional colour-absorbing material, and produces what some experts describe as a more brilliant light.

Bulb silvering can also be considered a lamp coating.  This involved coating a part of the lamp with aluminium to act as a reflector.  This can be done behind the lamp to increase its downward efficiency or in front for use in uplighting installations.

The base provides the electrical connection to the filament. Some bases are also used to position or align the filament in an optical system. There are eight types of bases: (1) screw, (2) screw with ring contacts (three-way), (3) skirted screw, (4) bipost, (5) prefocus, (6) disc, (7) bayonet, and (8) prong. The most common base is the screw base around the world, however in Australia the bayonet is the most common in domestic applications.

No commonly used light source emits equal amounts of each light frequency, including daylight. Incandescent lamps are known for their warm colour, resulting from the fact that they emit more lower frequency red and orange light than high frequency blue and violet. The graph below clearly shows this bias towards the lower end of the visible spectrum.

Spectral output of a typical incandescent bulb.

Tungsten Halogen

Some high intensity / long life globes are called tungsten halogen or quartz halogen.  These lamps are filled with a halogen gas, usually bromide or iodine.  The nature of this gas means that any tungsten atoms that evaporate from the surface of the filament combine chemically with surrounding iodine atoms.  In this state, they cannot form a black coating on the inside of the bulb, moving around until they impact with the hot filament.  When this happens, they split back into tungsten and iodine, depositing the tungsten atom back onto the filament and releasing the iodine atom to continue the cycle.  This allows much higher operating temperatures which require special bulbs, usually made from quartz or fused silica.

Tungsten-halogen lamps are dimmable. However, dimming will reduce the bulb temperature causing the tungsten-iodine cycle to stop, resulting in bulb wall blackening. Manufacturers claim that turning up the lamp to "full on" will clean the lamp. Extended dimming will increase lumen depreciation and reduce lamp life slightly.

Tungsten-halogen is an expensive incandescent lamp that has a very compact envelope which makes it an excellent lamp where optical control is important. It still has all of the negative aspects of the standard incandescent which are a relatively short life and a low efficacy which makes the tungsten-halogen expensive to operate and maintain. Color rendition, however, is excellent.

The normal voltage (120/240 V) lamp requires no auxiliary equipment (no ballast) which results in a slightly lower initial cost. The low voltage tungsten-halogen lamps require a step down transformer to reduce the line voltage from 120/240 V to 12 V. The transformer adds to the initial cost of the system and introduces a device that may require additional maintenance and has to be put somewhere.

The output spectrum of a tungsten halogen lamp is very similar to other incandescent lamps, shown above.

Electrical Discharge Lamps

When electric current is passed through a low pressure gas, the electrons flowing between the two electrodes collide with gas atoms, temporarily increasing their energy.  These atoms quickly decay to their stable state, releasing photons of ultraviolet radiation.  Phosphor coatings on the inside of the bulb absorb most of this energy and re-radiate it as visible light.

Flourescent Lamps

The most common application of this technology is in tubular fluorescent lamps.  A range of different phosphor coatings are used to modify the output spectrum.  The standard fluorescent tube has a diameter of 38mm and a length of 0.6, .9, 1.2, 1.5, 1.8 or 2.4 metres.  More recently, such lamps are available in both circular form as well as compact fluorescents utilising folded tubes of much smaller diameter.

The fluorescent lamp requires three elements or components to produce visible light:

• Electrodes (Cathodes)    
  Electrodes are the electron-emitting devices.  Two types of cathodes are in current use.  The hot cathode is a coiled coil or a triple-coiled tungsten filament coated with an alkaline earth oxide that emits electrons when heated.  The electrons are boiled off the cathode at about 900°C.  The cathode of a cold cathode lamp is a pure iron tube that also has an electron-emitting material applied inside the tube.  The cold cathodes are subjected to higher voltage, releasing electrons at about 150°C.  Cold Cathode lamps are used in special application such as neon signs and can be bent into different shapes.  The hot cathode lamp is the most common type of electrode used in fluorescent lamps for most applications.  Therefore, we shall not describe cold-cathode lamps.

• Gases
  A small quantity of mercury droplets are placed in the fluorescent tube. During the operation of the lamp, the mercury vaporizes at a very low pressure. At this low pressure, the current flowing through the vapor causes the vapor to radiate energy principally at a single wavelength in the ultraviolet region of the spectrum (253.7nm). The pressure of the mercury is regulated during operation by the temperature of the tube wall.
  
  The lamp also contains a small amount of a highly purified rare gas.  Argon and argon-neon are the most common, but krypton is sometimes used.  The gas ionises readily when a sufficient voltage is applied to the lamp.  The ionized gas decreases in resistance quickly, allowing current to flow and the mercury to vaporise.

• Phosphor
  This is the chemical coating on the inside wall of the tube or enclosure. When the phosphor is excited by ultraviolet radiation at 253.7nm, the phosphor produces visible light by fluorescence.  That is, visible light from a fluorescent lamp is produced by the action of ultraviolet energy on the phosphor coating on the inside surface of the tube or enclosure.  The phosphor mixture can be altered to change the color of the lamp or the lamp's spectral power distribution.

Effect of Temperature   
The most efficient lamp operation is achieved when the ambient temperature is between 20 and 30°C for a fluorescent lamp.  Lower temperatures cause a reduction in mercury pressure, which means that less ultraviolet energy is produced; therefore, less UV energy is available to act on the phosphor and less light is the result.  High temperatures cause a shift in the wavelength of UV produced so that it is nearer to the visual spectrum.  The longer wavelengths of UV have less effect on the phosphor, and therefore light output is also reduced.  The overall effect is that light output falls off both above and below the optimum ambient temperature range.

Effect of Humidity
Starting voltage requirements are affected by the electrostatic charge on the outside surface of a fluorescent lamp. Moist, humid air has unfavorable effects on the surface charge. This factor must be taken into account when the relative humidity exceeds 65%. A silicone coating on the outside surface of the lamp and the proper distance between the lamp and metal housing of the luminaire will usually solve starting problems under any conditions of humidity. However, dirt accumulation on the lamp will nullify the effects of the silicone coating and cause starting difficulties. Cleaning the lamp with an abrasive cleaner may also remove the silicone coating.

Burning Position
Fluorescent lamps should be operated in a horizontal position. Vertical operation causes a non-uniform distribution of gases in the lamp resulting in a reduction in light output and uniformity. In a vertical position, the mercury droplets are concentrated near the lower cathode increasing deterioration of the cathode and resulting in a reduction in lamp life.

Stroboscopic Effect
Stroboscopic is derived from the Greek meaning "to see motion."  The arc stream extinguishes during each reversal of the sine wave (100 times per second for a 50Hz current), however the phosphor coating continues to radiate light during this brief period.  Generally this is not noticeable, but it can make high-speed rotating machinery appear to stand still. The use of a series sequence ballast on rapid-start circuits will eliminate this problem.  Another solution is to use a lead-lag ballast, which puts one lamp out of phase with the other in a two-lamp unit.  This results in one lamp being at maximum light output while the other lamp is at zero output. The net effect is to eliminate the flicker.

Flicker is also most obvious at each end of a flourescent tube where the concentration of phosphor is less. It is therefore possible to reduce the perception of flicker by capping or obscuring from vision the two ends.

Spectral output of different classes of fluorescent lamp..

Disposal
Fluorescent tubes are actually categorised as hazardous waste and as such should not be disposed of through the normal waste stream in any real quantity. A single fluorescent tube contains enough mercury to pollute more than 30,000 litres of water. Most companies are unaware of this and subsequently more than 60 million tubes are sent for conventional unregulated landfill each year. Apart from the increasing difficulty in the disposal of fluorescent tubes there are health hazards associated in their handling. For example an employee attempting to dispose of a fluorescent tube in a skip may be exposed to flying glass and small amounts of toxic materials including lead and mercury released as dust or vapour which could be inadvertently inhaled.

Low Pressure Sodium Vapour

Another commonly used discharge lamp is based on Sodium Vapour.  When this type of lamp is first switched on, a small current passes through the gas giving off a faint red discharge.  After several minutes the sodium inside evaporates.  The resulting sodium vapour produces the almost totally monochromatic emission characteristic to this lamp (589-589.6, yellow).  This makes colour perception very difficult which means that it is almost solely used for street lighting.

The Low Pressure Sodium lamp has the highest lamp efficacy of all sources, but it is monochromatic (single wavelength) yellow.  It is variously referred to as LPS (low pressure sodium) or SOX (sodium oxide). The light-producing element is an arc tube.  The arc tube is U shaped and constructed of borate glass. The tube is dimpled to maintain a uniform distribution of sodium throughout the arc tube.  The arc tube contains a small amount of argon and neon to aid in starting the lamp.  The pressure inside the arc tube is approximately 1e^-3mm mercury; the enclosing space between the arc tube and the outer enclosure is under a vacuum.  Visible light is produced by the action of the electrons in the arc stream on the sodium. The excited sodium emits photons at one of two wavelengths,  resulting in essentially monochromatic yellow light (95% at 589nm and 5% at 586nm) as shown below. 

Spectral output of a low pressure sodium lamp.

The rated life for all wattages is around 18,000 hours based on a burning cycle of 5 hours per start.  Burning position is critical to lamp life since lamp failure is due to the migration of the sodium toward the electrodes.  This migration causes an increase in the watts consumed by the lamp over its life, which results in electrode failure. The lumen output of these lamps actually increases slightly over the life of the lamp. Lumen output is said to be constant over the operating temperature range of -10°C to +40°C. The effect on lumen output when the lamp is operated outside this temperature range has not been published.

High pressure sodium lamps are also used. These are not as efficient, but radiate energy across the visible spectrum. They are typically golden-white in colour. Note the sudden dip in its response at 589nm. This is due to self-absorption at those frequencies by the gas itself.

Spectral output of a high pressure sodium lamp.

High Pressure Mercury Vapour

Mercury vapour lamps have resonant emissions at 185nm and 254nm, both in the UV range.  At high pressure, the gas itself absorbs some of this radiation and re-emits it as visible light.  This emission is concentrated in 5 narrow bands, giving a violet-blue-green appearance.  As this spectrum is red deficient, the perception of many colours is distorted.

The light producing element is an arc tube which contains two operating electrodes and a starting electrode.  The arc tube is constructed of quartz to allow ultraviolet radiation to be transmitted.  The arc tube contains mercury and small quantities of argon, neon, and krypton.  When the lamp is energized, an electric arc is struck between the main and starting electrode.  As the mercury ionises, resistance inside the arc tube decreases.  When resistance inside the arc tube is less than external resistance, the arc jumps between the main electrodes.  The mercury continues to ionise, increasing the light output.  The light produced is in the typical mercury lines (404.7 nm, 435.8 nm, 546.1 nm, and 577.9 nm), plus ultra-violet (UV) energy.  The arc tube is operated at from 1 to 10 atmospheres of pressure.

A clear Mercury Vapor lamp produces a blue-green visible light. To improve this color, a phosphor coating is placed on the inside surface of the outer enclosure. The ultra-violet energy produced by the arc tube excites the phosphor coating producing additional visible light which improves the color rendition of the mercury vapor lamp. The main colors added by the phosphor are reds and oranges. 

Spectral output of a high pressure mercury vapour lamps.

Phosphor coated or white mercury vapor lamps are recommended for all applications where color is important.  There are three standard modified mercury vapor lamps:

• Color Improved: very poor on reds, marginal color, not recommended.
• Deluxe White, DX: increased red, good color, recommended.
• Warm White Deluxe, WWX: excellent reds, excellent color, highly recommended, decreased lumens.

Metal Halides are also often added to mercury vapour lamps to improve their colour quality.  These provide further emissions, making the spectrum even more continuous.  Metals such as thallium, indium or sodium iodide are the most common additives.

Spectral output of mercury lamps with metal halide additives.

Life testing of all high pressure gasseous discharge lamps is based on a burning cycle of 10 hrs/start.  The life of a mercury vapor lamp can be described in terms of its usable life or its rated life.  The rated life of these lamps is the number of hours they will continue burning. As lumen output degrades substantially in an old lamp, the useful life is the number of hours it puts out the required light levels. For most light sources, individual lamps tend to fail near the rated life.  But for mercury vapor, it is not uncommon for a lamp to continue burning for several times its rated life.  This poses a maintenance problem because the maintenance personnel, noting that the lamp is still burning, fail to replace a lamp that is only putting out a fraction of its rated lumens.

The lumen depreciation curve for a mercury vapor lamp is very steep, indicating that lumen output falls off rapidly with life.  Lumen depreciation is a function of the ballast and wattage.  Light output is also a function of the supply and regulation of the voltage to the lamp.  The steepness of the curve suggests why mercury vapor is the only lamp that is listed with a rated and a usable life:  long before we reach rated life (when half the lamps have failed) the remaining lamps are putting out such a small fraction of their rated lumens as to render them unusable.