The arc method is essentially a fusing process, though with due care it is used for heating to plasticity the edges of iron sheets to be welded by pressure and hammering. It has been found applicable in special cases to the filling of defective spots in iron castings, by fusing into blow-holes or other spaces small masses of similar metal, added gradually, and melted into union with the body of the piece by the heat of the arc. Similarly, a more or less complete union between separate pit<:es of iron plate J to J in. in thickness has been effected by fusing additional metal between them. The range of operations to which the arc process is applicable is naturally somewhat limited, and depends to a large extent upon the skill acquired by the operator, who necessarily works with his eyes well screened from the glare of the large arc. Unless the space in which the work is carried on is large, the irritating vapours which rise from the arc stream add to the difficulty. Strong draughts of air which would disturb the arc must also be avoided. These factors, added to the relative slowness of the work and the uncertainty as to its result, have tended to restrict the application of arc welding in practice. Moreover, much heat-energy is dissipated in the arc flame and passes into the air, while, owing to the time required for the work, the metal undergoing treatment loses much heat by radiation. Yet the method requires little special machinery. The current may be taken from existing electric lighting and power circuits of moderate potential without transformation, and may be utilized with simple appliances, consisting chiefly of heavy wire leads, a carbon or metal electrode with a suitable handle for its manipulation, a choking or steadying resistance, and screen of dark glass for the operator's eyes.
In 1874 Werdermann proposed to use, as a sort of electric blowpipe, the flame gases of an electric arc blown or deflected by an air jet or the like—a suggestion subsequently revived by Zerener for arc welding. The arc m this instance is deflected from the space between the usual carbon electrodes by a magnetic field. The metal to be heated takes no part in the conduction of current, the heat is communicated by the gases of the arc, and, to a small extent, by the radiation from the hot carbon electrodes between which the arc is formed. The process is scarcely to be called electric in any true sense. Another curious operation, resembling in some respects the arc methods, has been proposed for the healing of metal pieces before they are brought under the hammer for forging or welding. The end of a metal bar is plunged into an electrolytic bath while connected with the negative pole of a lighting or other electric circuit having a potential of 100 to 150 volts. The positive pole is connected with a metal plate as an anode immersed in the electrolyte, or forming the side of the containing vat or tank. A solution of sodium or potassium carbonate isii suitable electrolyte. That part of the bar which is immersed acts as a cathode of limited surface, and is at once seen to be surrounded by a luminous glow, with gas bubbles arising from it. The immersed end of the bar rapidly heats, and may even melt under the liquid of the bath. It is probable that an arc forms between the surface of the metal and the adjacent liquid layer, the intense heat of which is in part communicated to the metal and in part lost in the solution, causing thereby a rapid heating of the bath. This singular action appears to have been first made known by Holio and Lagrange. It is distinctly a form of electric heating, having no necessary relation to such subsequent operations as welding, and is, moreover, wasteful of energy, as the heat is largely carried off in the liquid bath.
The process of Elihu Thomson first brought to public notice in 1886, has since that time been applied commercially on a large scale to various metal-welding operations.Thomson process. The metal pieces to be united are held in massive clamps and pressed together in firm contact; and a current is made to traverse the proposed joint, bringing it to the welding temperature. The union is effected by forcing the pieces together mechanically. The characteristic feature of the process is the fact that the heat is given out in the body of the metal.
The voltage does not usually exceed two or three, though it may reach four or five volts; but as the resistance of the metal pieces to be joined is low, the currents are of very large values, sometimes reaching between 50,000 and 100,000 amperes. Even for the joining of small wires the current is rarely less than 100 amperes. Such currents cannot, of course, be carried more than a few feet without excessive loss, unless the conductors are given very large section. With alternating currents, also, the effectiveness of the work speedily diminishes, on account of the inductive drop in the leads, if they are of any considerable length. The carrying of the welding currents over a distance of several feet may, in fact, lead to serious losses. These difficulties are overcome in the Thomson welding transformer, which resembles the step-down transformers used in electric lighting distribution by alternating currents, with the exception that the secondary coil or conductor, which forms part of the welding circuit, usually consists of only one turn of great section, S S (fig. 1). This is often made in the form of a copper casing, which surrounds or encloses the primary coil P P in whole or in part. The primary coil is of copper wire of many turns. The secondary casing, with the primary enclosed, is provided with the usual laminated iron-transformer core, I, constituting a closed iron magnetic circuit threading both primary and secondary electric circuits. The terminals of the single-turn secondary serve as connexions and supports for the welding clamps C D, which hold the work. The clamps are variously modified to suit the size, shape and character of the metal pieces, MN, to be welded, and the proportions of the transformer itself arc made proper for the conditions existing in each case. The potential of the primary circuit may be selected at any convenient value, provided the winding of the coil P P is adapted thereto, but usually 300 volts is employed, and the periodicity is about 60 cycles. Inasmuch as only the proposed joint and a small amount of metal on each side of it are concerned in the operation, the delivery of energy is closely localized. The chief electrical resistance in the welding circuit is in the projections between the clamps, where the electric energy is delivered and appears as heat. A portion of the energy is, as usual, lost in the transformation and in the resistance of the circuits elsewhere, but, by proper proportioning, the loss may be kept down to a moderate percentage of the total, as in other electric work.
Fig. 1.—Thomson Welding Transformer.
The pieces are set firmly in the welding clamps, with the ends to be joined in abutment and in electric contact. The projection of each piece from the clamp varies with the section of the pieces, their form and the nature of the metal, and the time in which a joint is to be made; but it rarely exceeds the thickness or diameter of the pieces, except with metals of high heat conductivity such as copper. When the pieces are in place the current is turned into the primary coil of the transformer, sometimes suddenly and in full force, but more often gradually. Switches and regulating devices in the primary circuit permit complete and delicate control. At least one of the clamps, D (fig. 1), is movable through a limited range towards and from the other, and is thus the means of exerting pressure for forcing the softened metal into complete union. In large work the motion is given by a hydraulic cylinder and piston, under suitable control by valves. At about the time the current is cut off, it is usual to apply increased pressure. The softened metal is upset or pressed outwards at the joint and forms a characteristic burr, which may be removed by filing or grinding, or be hammered down while the metal is still hot. Sometimes the burr is not objectionable, and is allowed to remain. Lap welds may be made, but butt welds are found to be satisfactory for most purposes.
Fig. 2.
The appearance of round bars in abutment before welding is shown in fig. 2 at A; and at B they are represented as having been joined by an electric butt weld, with the slight upset or burr at the joint. Before the introduction of the Thomson process a few only of the metals, such as platinum, gold and iron, were regarded as weldable; now nearly all metals and alloys may be readily joined. Such combinations as tin and lead, copper and brass, brass and iron, iron and nickel, brass and German silver, silver and copper, copper and platinum, iron and German silver, tin and zinc, zinc and cadmium, &c., are easily made; even brittle crystalline metals like bismuth and antimony may be welded, as well as different metals and alloys whose fusing or softening temperatures do not differ too widely.
If the meeting ends conduct sufficiently to start the heating, it is not necessary that they should fit closely together, nor is it necessary that they should be quite clean, the effect of the incipient heating being to confer conductivity upon the scale and oxide at the joint.
