The iron melts like wax; but the action seems too powerful, the molten metal hissing and evaporating distinctly. In such a case one of the 3 parallel groups is rut out. Should the action then be too sluggish, one or more parallel groups is added. Sometimes the arc proves too small or extinguishes frequently; in such cases the number of cells in each group has to be increased.
The carbon holder (Fig. 147) resembles a pair of scissors, and consists of two copper bars having a round hole near the end, in which the pencil is held firmly, either by the friction of the parts or by means of a little wedge as shown in the figure. The flexible cable passes through the wooden handle. During working, the holder becomes hot, and may have to be cooled by plunging it into cold water. The operator wears strong leather gloves, and his hand is further protected by a metal screen fixed on the holder. He looks at his work through a dark glass as in Fig. 148, which protects both his eyes and face from the radiated light and heat better than ordinary dark spectacles would do. The lungs also may need protection from the vapours of copper, lead, and other metals or alloys. When possible, means should be provided to carry off such vapours with a blast of air. The construction of the holder permits of a quick replacement of the carbon pencil. The diameters of these carbons vary greatly. For more delicate work, where a few cells would suffice, fine pencils of only 1/16 in. are required; while boiler plates, such as mentioned above, are welded together by means of thick carbon rods of up to 2 1/2 in. diameter. The carbon is pointed before using it.
Ordinary light-carbons do not answer well, as they are generally too soft; the inventor prefers Carre carbons.
One of the most important applications of the new process is for welding plates of all thicknesses. For the very finest sheets of 1 mm. and less, the Electro-Hephaest Company prefer a modification of the Elihu Thomson process, although their own process is sometimes equally good (compare Figs. 168-170). But all stronger plates up to several centimetres thickness are subjected to the arc.
To effect this with ordinary plates, the edges are feathered as in Fig. 149, and pressed together. The furrows are filled with little pieces of the same material, and the arc is then applied while fresh pieces are added until the furrow is completely filled with the molten mass. The plates are immediately afterward finished under the hammer. In making iron welds, the small pieces for filling are always of wrought iron. With iron, a flux of clay sand is recommended; with copper, borax, or sal-ammoniac. .When the plates are joined on their lower surface, Benardos suggests a powerful electromagnet, placed as indicated in Fig. 150, to prevent the liquid metal (provided the material be para-magnetic) from flowing off; whether this suggestion will prove practical is doubtful. The apparatus shown in Fig. 151 looks more practical; it is intended to be employed when making vertical seams. The pincers p carry two pieces of graphite or coke g forming a sort of chamber at the spot where the fusion is to be carried on. As soon as the mass has hardened sufficiently, the carbon pieces are pushed farther up. Carbon pieces are frequently employed to prevent the flowing off of the fused material.
Figs. 152-161 exemplify other ways of joining plates in cases where a perfectly straight surface is not insisted upon. For thinner plates, the method Fig. 161 seems to offer particular advantages; for two 1/5 in. plates a seam 1 yd. long can be made in 7 minutes. When plates are to be joined at an angle, the process is of course exceedingly simple.
If two iron bars are to be joined end to end, the one bar is roughly centered in a lathe, and the other pressed against it; the body of the lathe is connected with the negative pole. A few momentary touches with the carbon will make the two bars stick together sufficiently so that they move as one piece with the lathe. While the lathe is turned slowly, the welding is effected by the addition of material in small quantities at a time. To join two telegraph wires, the ends are bent (Fig. 162), a little iron ring is pushed over the hooks, and the whole is fused into a sort of button; the resulting joint leaves nothing to be desired as to conductivity and breaking strength, and the whole operation can be accomplished with a few cells, and in 2 minutes for 4 mm. wires.
So far we have only spoken of joining materials of the same kind. But the intense heat of the arc supplies alloys which are hardly known under other circumstances, so that iron and copper, tin, zinc, lead, steel, cast iron and steel, wrought iron and steel, aluminium and platinum, etc., can be united. This promises important progress in the working of metals. Prof. Ruhlmann has exhibited specimens of iron plate welded to red copper, iron plated with tin, and iron plated with lead. In such cases there is probably at the junction of two metals a layer of alloy. Chemical manufacturers would be thankful for cheap copper retorts coated inside with platinum, or iron vessels coated with lead. Prof. Ruhlmann saw at St. Petersburg a number of copper tubes soldered into a cast-iron plate, and this iron plate coated with copper several mm. thick.
If the metals can be joined by the electric arc, they can also be separated by the same means. For instance, holed can be made if the metal is permitted to flow off. To pierce a hole 1 in. diameter through two plates 1/2 in. thick takes about 4 minutes. The next step is to rivet the plates in this way; this is shown in Fig. 159, where they are 1/2 in. plates, the rivet 3/4 in. thick, and the operation took 8 minutes. It seems, however, more advisable to punch or drill the holes.