The first cable of 1858 was laid by the U.S. frigate Niagara and H.M.S. Agamemnon, after having been manufactured with all the precautions suggested by Professor Thomson's researches. It is hard to realise how difficult such an enterprise was at the time. The manufacture of a huge cable, the stowage of it in cable tanks on board the vessels, the invention of laying and controlling and picking-up machinery had to be faced with but little experience to guide the engineers. Here again Thomson, by his knowledge of dynamics and true engineering instinct, was of great assistance. In 1865 he read a very valuable paper on the forces concerned in the laying and lifting of deep-sea cables, showing how the strains could be minimised in various practical cases of importance—for example, in the lifting of a cable for repairs.
A first Atlantic cable had been partly laid in 1857 by the Niagara, when it broke in 2000 fathoms of water, about 330 miles from Valentia, where the laying had begun. An additional length of 900 miles was made, and the enterprise was resumed. This time it was decided that the two vessels, each with half of the cable on board, should meet and splice the cable in mid-ocean, and then steam in opposite directions, the Agamemnon towards Valentia, the Niagara towards Newfoundland. Professor Thomson was engineer in charge of the electrical testing on board of the Agamemnon. After various mishaps the cable was at last safely laid on August 6, 1858, and congratulations were shortly after exchanged between Great Britain and the United States. On September 6 it was announced that signals had ceased to pass, and an investigation of the cause of the stoppage was undertaken by Professor Thomson and the other engineers. The report stated that the cable had been too hastily made, that, in fact, it was not good enough, and that the strains in laying it had been too great and unequal. It was found impossible to repair it, so that there was no option but to abandon it.
This cable probably suffered seriously from the violent means which seem to have been employed to force signals through it. Now only a very moderate difference of potential is applied to a cable at the sending end, and speed of signalling is obtained by the use of instruments, the moving parts of which have little inertia, and readily respond to only an exceedingly feeble current.
A second cable was made and laid in 1865 by the Great Eastern, which could take on board the whole at once and steam from shore to shore. It was also well adapted for cable work through having both screw and paddles. As Thomson points out, "steerage way" could be got on the vessel by driving the screw ahead, so as to send a stream of water astern towards the rudder, while the paddles were driven astern to prevent the ship from going ahead. This was of great advantage in manœuvring on many occasions.
This cable also broke, but a third was laid successfully in 1866 by the same vessel, and the second was recovered and repaired, so that two good cables were secured for commercial working. On both expeditions Professor Thomson acted as electrical engineer, and received the honour of knighthood and the thanks of the Anglo-American Telegraph Company on his return home, when he was also presented with the freedom of the city of Glasgow.
He afterwards acted as engineer for the French Atlantic Cable, for the Brazilian and River Plate Company, and for the Commercial Company, whose two new Atlantic cables were laid in 1882-4.
Mirror Galvanometer and Siphon Recorder
Since whatever the potential applied at the sending end of the cable might be (and, of course, as has been stated, this potential had to be kept to as low a value as possible) the current at the receiving end only rose gradually, it was necessary to have as delicate a receiving instrument as possible, so that it would quickly respond to the growing and still feeble current. For unless the cable could be worked at a rate which would permit of charges per word transmitted which were within the reach of commercial people, it was obvious that the enterprise would fail of its object. And as a cable could not cost less than half a million sterling, the revenue to be aimed at was very considerable. This problem Thomson also solved by the invention of his mirror galvanometer. The suspended magnet was made of small pieces of watch-spring cemented to a small mirror, so that the whole moving part weighed only a grain or two. Its inertia, or resistance to being set into motion, was thus very small, and it was hung by a single fibre of silk within a closed chamber at the centre of the galvanometer coil. A ray of light from a lamp was reflected to a white paper scale in front of the mirror, which as it turned caused a spot of illumination to move along the paper. A motion of this long massless index to the left was regarded as a dot, a motion to the right as a dash, and the Morse alphabet could therefore be employed. This instrument was used in the 1858 cable expedition, and a special form of suspension was invented for it by Thomson, to enable it to be used on board ship. The suspension thread, instead of being held at one end only, was stretched from top to bottom of the chamber in which the needle hung, and kept tight by being secured at both ends. Thus the minimum of disturbance was caused to the mirror by the rolling or pitching of the ship.
The galvanometer was also enclosed in a thick iron case to guard it against the magnetic field due to the iron of the ship. The "iron-clad galvanometer" first used in submarine telegraphy (on the 1858 expedition in the U.S. frigate Niagara) is in the collection of historical apparatus in the Natural Philosophy Department of the University of Glasgow.
The mirror galvanometer then invented has become one of the most useful instruments of the laboratory. Mirror deflection is now used also for the indicators of many kinds of instruments.