Читать книгу Lighthouses - Peter Williams - Страница 15

How did lighthouse illumination progress from the earliest times to the introduction of dioptric lenses in the 1820s? The story begins with early attempts to magnify the light from a small spot into a powerful beam that could be seen by mariners as they sailed over the horizon and saw land. The first beacons were open fires on windswept headlands, more smoke than flame, so practically useless at night when most needed. The early beacon fires were usually in a cresset or fire basket at the top of a tower, and the poor light they emitted was not reflected and directed seaward. The development of reflectors from flat plate mirrors to spherical cones produced some oddities, before the invention of the ultimate parabolic reflector that can still be found in use in the 21st century.

One of the first known lighthouses to use mirrors was built on the flat, featureless Nile delta at Alexandria on Egypt’s Mediterranean coast. Engineered by Sostratus for the Pharaoh Ptolemy in 285 BC, it was a monolithic stone tower 400ft (122m) high and called the Pharos, from which pharology, the science of lighthouses, gets its name. The design of the Pharos has been replicated all over the world and has attained a mythical status. Scant information remains about this tower, but it has been suggested by a number of lighthouse historians that the light source was a wood-fuelled furnace, not at the top of the tower as would be expected, but in a more readily accessible chamber nearer ground level. The smoke was taken away by a chimney system, separated from the light beam that was transmitted to the top of the tower by a series of flat plate mirrors, an early example of a reflective or catoptric system.

In Greco-Roman times, highly polished bronze, tin and even silver were used to improve the reflectivity of the metal surface of mirrors. This is the same technology used in the first known application of mirrors in a lighthouse to enhance light from the open wood or coal beacon by arranging them on the landward side of the fire. They were used at the Gollenberg lighthouse in Pomerania (in what is now Poland on the Baltic Sea) in 1532 and the Swedish Landsort lighthouse in 1669. Harwich High light on the east coast of England had a brass plate behind its enclosed fire, but was reported as ‘not very effective’. But perhaps the most interesting application was at Corduoan lighthouse at the entrance to the French west-coast estuary of the River Gironde. At this lighthouse, in 1727, the engineer M. Bitri was tasked by the French king to rebuild and improve the light. Bitri constructed an inverted cone with its apex about 3ft (1m) above the flames. To this he attached a series of overlapping, flat metal plates to reflect the fire around the horizon. While the theory was good, in practice the smoke quickly deposited a layer of soot on the reflector so degrading the exhibited light. Cleaning these reflector plates while the fire was burning cannot have been a very popular job. Neither coal fires nor flat reflectors, scattering the light rather than concentrating it, achieved their makers’ performance aspirations.

Tallow candles up to 2lb (1kg) in weight were used in many early lighthouses with varying degrees of success. North Shields lighthouse at the entrance to the River Tyne used them from when it was commissioned in 1540, but it was not until 1736 that copper reflectors were added in an attempt to give a better light to mariners entering the river. John Smeaton, England’s first celebrated lighthouse engineer, used a chandelier holding 60 candles at Eddystone lighthouse off the Plymouth coast in 1759. He had planned to use 24 oil lamps but decided that they would give off too much smoke and soot, so instead he fitted candles without any enhancement. One disadvantage of using candles and reflectors was the changing of the focal point of the flame as the candle burnt down. This caused the light beam to become less intense, and was therefore a major problem for mariners as they sailed past. To some extent this was overcome by using a device to keep the flame at a consistent level. Oil lamps and reflectors were fitted at Eddystone in place of the candles in 1810. The output of these oil lamps was said to be fives times that of a candle or five candle-power (or candela, a measurement of lighthouse output that lighthouse engineers still use).

By the time Eddystone was refitted, bowl-shaped, spherical reflectors were in use, made from a variety of materials. This type of reflector still had shortcomings, as the shape was not a correct optical design. It caused the rays collected to bounce back to the source so that emerging light was diffused and also diverged when it left the reflector rather than being concentrated into a beam. In France, Isaac-Ami Bordier-Marcet tried a different approach. His system, which he called the fanal sidéral lamp, used a single light source with two metal cones, whose apexes faced each other with the light sitting between them. The resulting enhanced light appeared 360 degrees round the horizon. Though never used in major lighthouses, this design was popular for many years for French harbour lights.

The greatest advance in reflector design was introduced in 1757 by the Swede Jonas Norberg, who had succeeded in his attempt to formulate the conic parabolic reflector for the light he was building at Korsö. He used a fundamental law of optics that the angle of incidence of the light ray is equal to the angle of its reflection. The curve of a parabola causes the light rays to reflect the oil flame, set at its focus, in a ‘cylinder of light’ with parallel rays. During a test Norberg used an 8-in (20-cm) diameter parabolic reflector with a candle as his light source. He reported seeing the light from 8 miles (12.9km) away, but without the reflector the candle was visible from only a very short distance. Others were experimenting with the same idea. William Hutchinson, a Liverpool dock master, needed to develop approach lights to Liverpool in order to guide vessels safely through the sandbanks in the entrance channels to the port. At Bidston lighthouse, built in 1772, he used a single large oil lamp with a wick 14in (35.5cm) across, set at the focal point of a 13ft 6in (4.1m) parabolic reflector. This monster was made by moulding a concave depression in a plaster-of-Paris block set in a wooden tray. The reflective material was a myriad of glass facets each about 1in (3cm) by ¹/₃in (1cm) set into the concave face. For the smaller lightshe used soldered tin pieces.

In Scotland, Thomas Smith, an Edinburgh lamp maker, started with the same method as Hutchinson, then, using his metalworking skills, evolved the design into shaped metal reflectors. His stepson Robert Stevenson, whom he brought into his business, soon recognized the superiority of the metal reflectors made by Robinson of London for Trinity House, the lighthouse authority of England and Wales. Smith’s first silver-plated reflectors cost 15 guineas (today’s value approximately £1,300) each and Stevenson fitted them at Inchkeith lighthouse in the Firth of Forth in 1804. Trinity House standardized on the use of 21-in (53-cm) diameter reflectors made of copper with a fusion of silver on the reflective side. The focal length was 4in (10cm) and the total reflective area was 518.6 square inches (3,345 square centimetres). The light output was 350 times the candlepower of the illuminant. They were manufactured by Robinson and his assistant Wilkins in their London workshop at a cost of £31 10s each (roughly £2,200 today). Robinson and Wilkins had a monopoly with Trinity House until 1845, and also supplied many other lighthouse operators both at home and overseas.

Fig 1

A rotating catoptric (mirror) system with metal reflectors. The reflectors were about 20in (50cm) in diameter. The clockwork motor was driven by a weight that descended the height of the tower.

Fig 2

Fountain-type oil lamp with catoptric reflector. The oil lamp is lowered for cleaning and polishing the reflector and raised to the correct position when lit to ensure that the flame is at the focus of the reflector and is working to maximum efficiency.

Not all the attention of lighthouse engineers at that time was on magnifying the light. The Swiss engineer Ami Argand developed a smokeless oil lamp in 1784 that revolutionized lighting in the home and workplace as well as lighthouses. By passing a controlled stream of air over a series of concentric wicks enclosed in a glass chimney he gave the world a bright, smokeless light. Combining his lamp with a parabolic reflector was a major leap forward in lighthouse technology. It was immediately adopted by Trinity House, which by 1799 had 11 lighthouses in operation with this equipment. Robert Stevenson reported on a typical installation when he visited St Ann’s Head lighthouse in Wales during his tour of British lights in 1801. He noted,‘The light is from Argand burners with parabolic silvered reflectors each 20.5in diameter. In the one lantern there are sixteen reflectors and in the other eleven’. He was referring to the front and rear towers that were 300 yards (274m) apart. He further commented on the ‘complete state of cleanliness and good order’ and that they were ‘better than the Smalls [18 miles (29km) away offshore] and a brilliant light’.

These were fixed lights but already in 1781 at Carlsten in Sweden the first revolving parabolic reflector array to make a flashing light had been put into use, again by Norberg. Using a clockwork mechanism, Norberg’s six reflectors and lamps completed a revolution every five minutes and produced six strong and three weaker flashes. The light produced a strong flash every 50 seconds, while at least one of the weaker lamps was visible at all times to the mariner. The revolving system was the most important innovation in light projection to date, as for the first time mariners could distinguish one light from another by the frequency and power of the flashes it emitted. It was the forerunner to all the great advances that occurred a hundred years later during what came to be known as the ‘Age of Magnificence’ in lighthouse illumination.

All these lights were set up on headlands and on the approaches to ports, but sometimes no suitable piece of ground could be found, so instead the light was mounted on a floating rig. Floating lights were first used in Roman times, when galleys showing a beacon flame were anchored at the entrances to estuaries and harbours during the night hours to assist homeward-bound ships. They were not a permanent fixture and were used only as needed. The first recorded use of permanently stationed light vessels was in 1731, when a Kings Lynn merchant Robert Hamblin and his London partner David Avery placed a converted Dutch galliot to mark the Nore Channel in the approaches to London. The galliot was fitted with a simple light raised each night from the deck. It was not very effective, but very profitable, and paved the way for other light vessels to mark the channels and extensive sand banks on the busy shipping routes into London. Sand banks were difficult to light from the earliest times, while reefs and isolated rocks gave the early lighthouse builders a tough time. It was impossible to gain a foothold on shifting sands, even to survey let alone construct a lighthouse, but a ship could be anchored on the sand and exhibit a light to warn other ships away from the danger. Occasionally light vessels were, and still are, used while a lighthouse was being built. Robert Stevenson used a light vessel in 1807 when he was planning the work on Bell Rock. His tender Pharos had large copper lanterns fitted with 10 lamps and small, silver-plated reflectors.

Fig 3

Section through an Argand Lamp showing how the air for combustion was drawn in at the base of the lamp and through the centre of the wick into the flame. Air was also drawn up the outside of the wick to assist with smoke clearance, preventing soot forming on the inside of the glass chimney.