Unremitting research has been responsible for striking improvements in the electric illumination of houses, streets, docks and airports. The tungsten filament lamp remains the standard for domestic lighting, but for other purposes great advances have been made with electric-discharge lamps of various types
MODERN STREET LIGHTING, seen from the vehicle driver’s point of view. The most notable feature is the absence of pools darkness on the road surface. The lamps are mounted at such a height that they do not cause undue dazzle to the road user. Electric-discharge tubes are used: mercury-vapour tubes giving a bluish-white illumination and sodium tubes giving a yellow colour.
THE tendency to take the wonders of science and engineering for granted is nowhere more apparent than in the sphere of lighting. A switch is touched, the current flows and darkness is turned into light. The process is simplicity itself.
Few of the countless users of electric light can have any conception of the enormous amount of research that is carried out on this highly specialized and important subject. The efficiency of modern lighting is being increased from day to day, preconceived ideas are going by the board and an entirely new technique is developed as required to deal with lighting of special kinds.
The revolution that is taking place in the lighting of the roads of Great Britain is a good example of a complete change of technique. Users of the roads are conscious that many roads are more brilliantly illuminated than they were before, and that the lamps are of a yellow or a blue colour; but road users may not realize precisely what these changes imply.
In the huge laboratories of the General Electric Company, Limited, at Wembley, Middlesex, a complete section is continually dealing with problems of illumination, and much of the work of this section is devoted to finding the best means of meeting the demand for safe and adequate illumination of the country's highways. Other subjects receiving close attention are the lighting of airports and docks, the floodlighting of buildings, every form of interior lighting, and the development of new types of illuminant.
A laboratory of this kind must work in close co-operation with the factory in which the lamps are made. For this reason a large room is given up entirely to life-test work, that is, to find out how long commercial types of lamp will last. It is necessary for the research staff and the factory to learn everything possible about the commercial product, and this can be done only by a close study of its behaviour in working conditions.
Samples of lamps of all types are therefore collected from the Osram lamp works, and they are tested for the correctness of their rating (by photometric methods) and for their life. Lamps are tested also for their general robustness, and several ingenious pieces of apparatus are installed for this purpose. One particularly brutal test involves the rolling of fairly heavy ball bearings down a series of inclined chutes, in such a way that the metal balls come into violent contact with the glass bulbs of a row of lamps mounted at the foot of the chutes.
A STANDARD AND LANTERN of a type now extensively used for street lighting. A mercury-vapour lamp is mounted inside the lantern, which is so designed that it gives the best possible distribution of light over the road surface, from the point of view of a driver or of a pedestrian. The glass of which the lantern is made comprises an elaborate system of minute prisms for directing the light in the desired
The number of impacts is counted automatically and the life of the lamps under test is recorded. Another piece of apparatus subjects lamps to the kind of vibration that is normally experienced by units of motor-car equipment, for motor-car bulbs are an important item in the output of a modern lamp factory.
Another subject of research is the tungsten filament of the ordinary gas-filled bulb for domestic use. Although gas-discharge lamps are coming into their own for industrial and exterior lighting, the tungsten filament is still regarded as the standard source of light for ordinary homes.
The filament in a gas-filled lamp has to operate at a temperature of about 2,400° centigrade - roughly twice the temperature of molten steel. It has to withstand this temperature for 1,000 hours without undergoing a measurable change in thickness, if it is to pass the tests of the laboratories. The thickness of such a filament is about 0·015 mm. (roughly half the thickness of a human hair), and to study the structure of the tungsten a special microscopic equipment has been installed.
Whenever it is necessary to study and experiment with a complicated manufacturing process, it is impossible to carry out the work in the factory in which the process is used. Thus a certain amount of duplication is tolerated, and the research laboratories are equipped with miniature factories in which the methods used are similar to those used for large-scale production. Tungsten wire and completed lamps are dealt with in this way at the Wembley laboratories.
The life-test room for lamps is a most impressive sight, but is somewhat hard on the eyes of a visitor. Many thousands of lamps are mounted in racks and run at their normal brilliance either for a specified number of hours or until they fail. The heat radiated from this test is considerable.
Much work is being carried out on electric-discharge lamps of various types. The “Osira” high-pressure mercury-vapour lamp is becoming familiar because of its extensive use for road lighting, and the long tubes filled with neon and other gases have for many years been one of the main features of advertising at night in London and elsewhere.
A newer type of lamp, however, on which experimental work is being carried out, consists of a tube filled with neon or some other gas and coated on the inside with a fluorescent powder.
High-pressure mercury-vapour tubes, which normally give a bluish-white light similar to that now associated with road lighting, can be made to give a green glow of considerably greater brilliance by this method of using fluorescent powder. Many other brilliant colours may be produced by this means.
ONE OF THE SWITCHBOARDS controlling the distribution of electricity to the various sections of the G.E.C. research laboratory, in which the total amount of power required is tremendous. In the life-test room alone several thousands of lamps of every size are constantly consuming current. On the left of the switchboard is a long corridor in which tests are carried out on the parabolic mirrors used for the floodlighting of airports.
The fluorescent materials used are tested in a room which can be illuminated with ultra-violet light. Although there is no visible light in this room - a white porcelain block appears to be dull brown in colour - screens coated with various fluorescent substances will emit vivid glows of almost every conceivable colour.
Several uncanny experiments can be carried out with ultra-violet light. Human flesh assumes a weird powdery appearance, suggesting the last stages of decomposition, but finger-nails and teeth fluoresce brilliantly. Silver and copper coins cannot be distinguished from one another; many white objects appear dull brown or black. Some fluorescent screens exhibit a strong “afterglow” - that is, when removed from the ultra-violet light they “die out” slowly, instead of immediately ceasing to fluoresce.
The main purpose of this section of the laboratory, however, is the investigation of the possible future of gas-discharge tubes coated with fluorescent powder.
The “Osira” lamp, as used for street lighting, is in quite a different category. Whereas the long tube lamps are run at high voltages which necessitate the use of step-up transformers between the 230-volts mains and the lamps, the “Osira” operates at the mains voltage. It is fairly compact - it is not a long tube - and smaller sizes are being developed already.
The tungsten filament lamp may be said to be an imitation of the sun. It produces light because a body that is sufficiently hot will radiate light.
Ever-Increasing Temperatures
Producing light by heating bodies is always wasteful, because other radiations that are of no use for seeing are produced also. The history of the development of the filament lamp has been one long story of ever-increasing temperatures. The tungsten filaments in gas-filled lamps are made far hotter than the metal filaments of the older vacuum lamps; and those filaments, in turn, operate at far higher temperatures than the carbon filaments of the earliest electric lamps.
If, however, it were possible to make a filament as hot as the sun, its efficiency for lighting purposes would be only one-sixth of what should be possible were all the energy converted into light.
The engineers have long realized that hot bodies are not the only possible sources of light. Light is caused by a complicated movement of electrons, and heat is only one method of making electrons move. Another method is to discharge an electric current through a gas. This is believed to be the oldest electrical method of producing light. It was in use more than one hundred years before the filament lamp was produced.
A discharge tube, by virtue of the narrowness of the band of frequencies that it radiates, is highly efficient, but it may suffer from certain disadvantages. One of these is its property of emitting light of one well-defined colour.
By making a discharge tube radiate a “chord” instead of a “single note” it is possible to overcome this disadvantage. A “single note”, representing true monochromatic light, may give high efficiency but a poor colour rendering. The mercury-vapour lamps used for road lighting are not monochromatic, but they are deficient in red, and this accounts for their effect on human complexions. Under the glare of the yellow-hued sodium lamps, brightly coloured cars appear black or brown. Most coloured objects assume a photographic effect of sharp contrasts.
ELECTRIC-DISCHARGE TUBES ON LIFE-TEST. These tubes are of the type used for advertising and for the exterior illumination of cinemas, and operate at high voltage. The tubes being tested are enclosed within a wire cage which cannot be entered by an operator until special precautionary measures against electric shock have been taken. Hundreds of tubes of all sizes and all colours are tested over long periods, sometimes for years.
A light deficient in red is not necessarily a disadvantage for road lighting, and the “Osira” lamp, by virtue of its high efficiency, has proved almost ideal for that purpose. Smaller lamps are being developed, however, which may ultimately give a good enough colour rendering to make them suitable for use in the home. Already an electric-discharge lamp, containing carbon dioxide, has been made to give a close imitation of daylight and is suitable for such specialized work as colour matching.
Mercury vapour has been used experimentally for discharge tubes for many years, and early in its history it was discovered that mercury gave light of an unpleasantly blue colour. Its use was restricted, until it was discovered that by increasing the pressure of mercury vapour in the tube, the “chord” radiated underwent great changes, the light became more of the colour that was required and the efficiency was greatly increased.
This increase in pressure brought several problems in its train, and the production of a practical lamp, as distinct from a laboratory toy, involved many years of research. A new glass had to be developed to stand the high temperature. New electrodes were required to stand the impact of the powerful discharge. In 1931 an experimental lamp was ready. Its efficiency was more than twice that of a filament lamp, and its light, although markedly deficient in red, gave a colour rendering which was true enough for street lighting. In June 1932 the first street lighting installation was set up, and early in 1933 a complete installation was put into operation in service conditions, a stretch of Watford Road, Wembley, being chosen for the purpose. The new lamps behaved perfectly and demonstrated that they were no longer to be regarded as experimental. Transport trials were next carried out, batches of the new lamps being sent all over the country by parcel post, passenger train and goods train. There were breakages. Mechanical construction was altered and the breakages were reduced in number. Finally they were eliminated almost completely. By the beginning of 1934 the “Osira” lamp was a standardized product, and since then high-pressure mercury lamps have become familiar in most of the large cities and towns in Great Britain.
Work on road lighting, however, did not end with the perfection of the new type of lamp. Special lanterns to give the best possible distribution of light had to be produced, and before this work could be even started, it was essential that the best possible distribution should be known and agreed upon. The laboratories therefore began a comprehensive study of the special conditions of road lighting, particularly from the point of view of light distribution. The old theory, held in many quarters, was that it was necessary to illuminate objects on the road to make the road safe for traffic at night.
Model Street Experiments
This theory was soon found to be fallacious, and the newer technique is based on the theory that the road surface should be illuminated so that objects on the road stand out by contrast. By day the motorist sees, normally, by the contrast of dark objects against a lighter background. He is now being enabled to see in the same way by night. Even if the road surface appears black - as many modern road surfaces do - the motorist views it from a glancing angle.
Working on the assumption that of all the factors which affected the visibility, the brightness of the road surface was one of the most important, the laboratories embarked upon a comprehensive series of tests. Lanterns were designed, and were tested indoors for their candle-power output in different directions: in other words, their “polar diagrams” were plotted. These lanterns were then fitted on posts in a street, and the road surface was viewed, criticized and photographed from every angle.
AIRPORT FLOODLIGHTING UNIT in which tubular lamps with long filaments are mounted in front of specially designed parabolic mirrors. The unit may be rotated or its elevation may be changed by hand. A 9-kilowatts floodlight of this type gives a maximum candle-power along the beam of more than one and a half million, and will illuminate an area of about 7,700,000 square feet.
The main requirement was not to give a distribution which would make the road appear evenly lighted when viewed from an aeroplane. By making use of the reflecting properties of the road surface at glancing angles it was proposed to give to this surface a uniform brightness from the only angle which mattered - that from which it would be viewed by a motorist, cyclist or pedestrian.
Much of this work, in the early stages, was carried out on the roads. Later, as the work became more and more a matter of routine - as the conditions were better understood - it was taken more into the laboratories. In one of the test rooms there is a model street, stretching, apparently, as far as the eye can see. Different types of lighting can be switched on in an instant - central lighting, lighting on both sides, diagonal lighting and the like. The “street” can be twisted in either direction, so that the effect of lighting at a gradual bend may be studied. Different sets of scenery may be provided by the turn of a knob, and the whole road surface may be changed, to give a matt or a glossy effect.
Prevention of Glare
The effect of direct glare may be studied, and also the clearness with which different objects on the road surface stand out. Other models have been built specially for studying such separate problems more closely. In one instance a photograph of a brilliantly illuminated road is mounted in a dark cabinet and lights may be switched on which give to each lantern in the photograph a brilliant central point of light, to simulate the effect of this glare on visibility.
New types of lanterns are tested and their polar diagrams plotted. The glassware emanating from the factory is checked for accuracy of moulding; for every lantern comprises an elaborate system of minute prisms for directing the light in the desired direction and for cutting off the light that is not required. Many people regard these lanterns as being made merely of diffusing glass - but they are far more scientifically made than that. When a new type is received from the factory, every single prism is optically tested in the laboratories.
Outside tests, on the roads, involved much photography, and care was taken that the photographs were correctly reproduced to give an exact idea of the lighting conditions on the road.
Two model cats were taken out on to many roads and photographed in different conditions of lighting. They appear in many of the standard photographs used at the laboratories, and have also figured in many impromptu episodes on the roads of Great Britain between 3 a.m. and 4 a.m. Motorists have a habit of expecting cats to move out of their way. These particular cats were not so made.
Finality in safe road lighting has not yet been reached, and research continues. Such subjects as the best distribution of lamps at a gradual bend are being studied. Rises and falls in the road also bring their own particular problems, and as each is solved another is brought up. Outdoor lighting has lagged behind indoor lighting chiefly because of expense: there is so much more space to be lit. The efficiency of discharge lamps, however, is so much higher than that of filament lamps that the adequate lighting of the country’s roads is proceeding at a far more rapid pace than it was a few years ago.
AN AIRPORT BEACON, which flashes a characteristic code signal and enables a pilot approaching an airport to identify it from a great distance. A large group of neon tubes, operating at a high voltage, is used, and the unit is carefully screened and protected by safety devices. The signal code is controlled by a motor-driven flashing mechanism, and a safety governor ensures that the light is kept continuously on, should the motor fail for any reason.
The relatively new small mercury-vapour lamps are claimed to be suitable for the lighting of side roads, the high-power lamps being used for main and secondary roads. The small lamps, moreover, are suitable for industrial lighting, and they give a light output three times as great as that obtained from a tungsten filament lamp of equivalent wattage.
By virtue of their colour characteristics, however, they are not suitable for domestic use. Research into this matter is now being carried out, and the combination of electric-discharge tubes and fluorescent powders may become responsible for the evolution of a new system of illumination for houses.
A new lamp which is strictly in the experimental stage is of the high-pressure water-cooled type. In size it is minute. It is merely a glass tube about 2 in. long, filled with mercury vapour at much higher pressure than is used for the commercial products. Light of tremendous intensity can be produced from these tubes, but they become so hot that water cooling is essential.
White light is a mixture of all the colours of the spectrum. The “Osira” lamp gives most of its radiations in the yellow-green-blue range; sodium light is yellow; neon is red. The water-cooled high-pressure mercury lamps are undoubtedly a step on the road to the production of white light with much greater efficiency than that available with tungsten filament lamps.
Airport lighting is a specialized subject which is becoming increasingly important as night flying develops, and it is not surprising to find that the G.E.C. laboratories carry out much research on this subject. It is important for the commercial success of aviation that good lighting should be provided at every great airport - and now the point of view of the approaching pilot must be considered.
The pilot must be able to locate the airport by means of a characteristic beacon. As he approaches, the landing area must be clearly mapped out for him by boundary lights. An illuminated wind-direction indicator conveys vital information to him before he starts his glide. A wind-speed indicator is a refinement which he will certainly appreciate. Obstruction lights warn him to keep clear of buildings, masts and trees. Finally, the ground on which he lands will be floodlighted.
Airport Floodlighting
A typical beacon consists of a large group of neon tubes, identifiable by a characteristic signal - possibly a letter in the Morse code. The flashing mechanism is motor-driven, and a safety governor ensures that the light remains switched on if the motor should stop because of a failure.
Boundary lamps are evenly spaced round the edge of the landing-ground, and consist of small filament lamps mounted within orange globes. These lamps are mounted on stems 3 feet high. An interesting feature is a replaceable weak section that will break readily, minimizing the danger of a collision if a plane should run into a light standard.
A typical wind-direction indicator consists of a flat metal T mounted on a pivot. Its upper surface is painted white, and has several small lamps spaced about one foot apart. It is placed in a position adjacent to the landing area, but where it will not create an obstruction.
The floodlighting is perhaps the most important part of airport lighting. The space illuminated must be sufficient to ensure that planes do not overrun the lighted area, and the floodlights must be so placed that undulating ground does not cast deep shadows, which, from the air, appear as holes. The light from the floodlight must be kept below the horizontal so that the pilot is not dazzled by it while he is circling round the aerodrome. In the laboratories the specially designed mirrors for these and other floodlights are tested by a photographic process which reveals flaws that would not be visible to the unaided eye.
Several lamps are used for safety, so that the accidental failure of one or more of them does not plunge the airport into darkness. Lamps with long horizontal filaments are used for airport floodlighting, and they are arranged at the foci of parabolic reflectors of optically worked silvered glass. A 9-kilowatts floodlight will illuminate an area of about 7,700,000 square feet, and the maximum candle-power along the beam is more than one and a half million.
EXPERIMENTS WITH NEW LAMPS form an important part of the work of a lighting laboratory. In the lamp-development section, which forms a link between research and production, many factory processes are carried out on a small scale. New types of lamps are made in small quantities and are checked for their life and for their efficiency.
