a dome, a knob, a point, or a flame. In case a, the point and the flame produce a silent discharge, and protect the other bodies; in case b, they are all struck indiscriminately, and by sparks of the There seems little doubt same kind; it seems immaterial also whether the object is placed directly on the lower plate or is insulated. Do these two cases occur in nature? that they do, especially if we consider the clouds as made up of isolated masses. In the experiments described, the discharge is oscillatory. Are the conditions which permit oscillation (§ 329) realised in the case of a thunder-cloud? This is more doubtful; but it may be observed that the effects which we have in view are not so much connected with the oscillatory character of the discharge as with the extreme rapidity of the variations of potential. T FIG. 381. 474. Lightning Conductors.-Shortly after the discovery of the properties of points, Franklin conceived the idea of utilising them for protecting buildings against the effects of lightning discharges. Franklin's lightning conductor consists of a long metal rod pointed at the top, placed at the highest part of the building to be protected, and connected with the earth by continuous good conductors. As now made, lightning conductors have a fine point of platinum or a stouter one of copper (Figs. 380 and 381). Franklin ascribes to the lightning conductor a twofold effect. Under the influence of the cloud the point allows the electricity of the opposite kind to escape, and this, carried by the particles of air, silently neutralises the electricity of the clouds. This is the pre FIG. 380. ventive effect. But instead of supposing that the electricity which issues by the point neutralises that of the cloud, we may also attribute the preventive action to the electrification of the air by the point; in this way there may be formed an electrified cloud, which tends to produce at all points below it a potential of opposite sign to that of the thundercloud, and therefore neutralises the influence of the latter. If such a cloud of electrified air remained floating above the conductor, equilibrium would be rapidly attained, and the point itself would cease to act. If, notwithstanding the point, a discharge takes place, it strikes the rod in preference to the other parts of the building which are within the radius of its protection, and the conductor leads the electricity to earth without any damage. This is the preservative effect. The conductor is usually an iron rod, 1.5 centimetres in diameter, experience having shown that lightning has never damaged a rod having this section; its lower end is connected with water or earth that is always damp. All large pieces of metal belonging to the building, whether inside or outside, should be connected with it. It is sometimes assumed, but quite arbitrarily, that a conductor protects a circular area, the radius of which is equal to twice the height of the conductor. In the erection of lightning conductors, and in instructions drawn up for this purpose, the case a has alone been taken into consideration, and even within these restricted conditions it does not appear that sufficient allowance has ever been made for the effects of self-induction. The duration of the discharge is so short that the effects of self-induction greatly preponderate over those of resistance. A flat band of metal is better than a solid circular rod. The relative advantages of copper and iron have frequently been discussed. Iron appears preferable if magnetic permeability does not come into play. The experience of more than a century shows that Franklin's lightning conductor is an efficient protection in the great majority of cases, but it would be an exaggeration to say that it has never failed except by some defect in the construction. 475. Theoretical Conditions. If we inquire what conditions theory prescribes for the protection of a structure and all that it contains against damage by lightning discharge, we arrive at the conclusion that the structure should be a closed conductor, such as a room, the walls, flooring, and ceiling of which are of iron. Whatever forces may be exerted from outside, the potential in the state of equilibrium will be constant for all points of the surface and the bodies which it encloses; there would be no trace of electricity on any of them. When equilibrium is disturbed, as by a lightning discharge, it is to be supposed that the sides would form a protecting screen for all bodies in the interior, and that sparks like those of Hertz would not be obtained. In any case, these sparks would not be dangerous, and, except perhaps in the case of a gunpowder magazine, there would be no need to consider them. For an external body connected with the ground no sparks are possible in the case of equilibrium, but in the case of a sudden discharge, there might be dangerous sparks. We have already seen that, in order to produce a conducting surface at constant potential, it is not necessary that the metal should be continuous; the conditions are realised by a network of even larger meshes. The experiment in § 335 shows that even when equilibrium is disturbed a network has the same effect as a continuous surface. The essential condition is that no conductor shall project into the interior, which might have a potential different from that of the envelope, and might therefore form a kind of electrode capable of giving sparks. Such, for instance, would be gas or water pipes; conductors of this kind should be connected with the conducting enclosure at the point where they enter. 476. Practical Conclusion.—The general result at which we ultimately arrive is that the most certain way to protect a building from the effects of lightning is to encase it in a metal network in perfect conducting communication with the earth. The network may be made of galvanised telegraph wire. The conductors should follow the ridge, the corners, the angles, the chimneys, &c., and loops or sharp angles should be avoided. Gas and water-pipes should be connected with the network on their entrance, and inside the building the pipes should be in metallic connection wherever they are near each other. The external network should be connected with the ground by the greatest possible number of points. All masses of metal on the outside-roofing, eaves, rain-troughs, &c. -should form part of the network. Connection should be made with the underground water-bearing stratum by means of large plates dipping in a pit dug in this stratum, and so deep that even when the water is at its lowest the plate is partially immersed. Large masses of metal inside the building should be connected with each other, but instead of joining them to the outer network, it is better that they should have an independent connection with the earth. The building thus protected might be struck by lightning, but there would be nothing to fear from the effects. A few strokes of lightning, probably harmless in any case, might perhaps be avoided by providing the conductors with points. From this point of view, it should be attempted to surround the building with an electrified atmosphere, and this would be attained more easily the more numerous are the points and the more widely they are distributed. We are thus led to the plan of lightning conductors devised by Melsens. The most important point, and that most frequently neglected, especially in the older installations, is that of connection with the earth. A system of lightning conductors badly connected with the earth is not only useless but dangerous. |