Non-arcing metals.

The construction is clearly shown. It consists of a number of short cylinders made of non-arcing metal which are held close to each other in a porcelain casing. The lightning discharge strikes across from cylinder to cylinder until it reaches the ground.

One of the choke coils of an arrester, used in connection with alternating current circuits of very high e. m. f., is shown in Fig. 149, by connecting a large number in series, and combining these with a corresponding number of reactive coils. These arresters and coils are assembled in a frame with the arresters arranged at the top of the frame, and the coils piled side by side under them.

Fig. 150 shows a type of arrester for direct current circuits, especially electric railways.

A number of theories have been advanced to explain the action of so-called non-arcing metals. We know from experience that with some metals the length of arc that can be sustained with an e. m. f. is greater than with other, and a question naturally arises as to what causes this difference. If we investigate the action from the instant when the arc begins to form we will be able to discover what occurs, and thus be able to form some idea of the cause of the difference in action with different metals.

Suppose we take two rods of metal pressing against each other end to end. If we separate them the current will jump across the intervening air space. At the instant the arc is formed the resistance between the points is very high, being due to the length of the column of air between the points. The current passing over this resistance generates a large amount of heat, and this vaporizes the metal, thus filling the space between the ends of the

Non-arcing metals.

rods with metallic vapor, which has a much lower resistance than the air. If the vapor is of high resistance, the amount of heat developed will be greater than if it is of low resistance and as a consequence more metal will be vaporized. If the resistance of the vapor is very low the amount of heat developed will be correspondingly low, and therefore the amount of metal vaporized will be comparatively small.

With a metal which gives a high resistance vapor, the heating is prevented from becoming very great from the very fact that the high resistance cuts down the strength of the current, and with the low resistant vapor the heat is prevented from becoming very small from the fact that he low resistance causes the current to increase in strength. Thus we see that while the high resistance vapor acts to confine a large amount of heat within a small space, it also acts to keep down the total amount of heat developed by reason of not permitting the current to rise to a considerable strength. On the other hand, the low resistance vapor enables a large current to flow, and thus a large amount of heat to be developed, but this is distributed over a much larger volume of vapor.

From these facts it follows that if the vapor is of high resistance, the arc will be of small crossection, which will help to increase its resistance, while if the vapor is of low resistance the arc will be of large crosssection which will help to reduce its resistance. The conclusion we are naturally led to, therefore, is that the higher the resistance of the vapor of a metal, the smaller the arc sustained between the ends of rod made of that metal. If two metals whose vapor has the same specific resistance are of different fusibility, then the two which melt at the highest temperature will be vaporized, the

Non-arcing metals.

least with a given amount of heat, therefore, the resistance of the arc produced by it will be the greatest. We therefore see that a non-arcing metal is one which does not vaporize readily, and whose vapor has a very high resistance. Experimentally it is found that zinc and antimony are about the best non-arcing metals we have. Now both of these fuse at low temperature, so that the only conclusion we can come to is that the resistance of their vapor is sufficiently high to overbalance the low fusibility.