If, by imprudence, the switch should be moved to the right while no stock is inserted and the clamps in contact with each other, the switch cannot be locked, and the fuse in primary will be blown, without, however, causing any damage.

The insertion of the distance gauge between clamps locks the switch, so that primary can only be closed after withdrawing the gauge.

The operation is as follows: -

1. The wires to be welded are gauged.

2. The distance gauge and cut-off are set to correspond with number on gauge.

3. The movable clamp is moved to the right and gauge inserted.

4. Both wires are securely clamped, care being taken to abut the ends squarely against the disc.

5. The gauge is withdrawn.

6. The switch is moved to the right.

7. The reactive coil moved toward a position of minimum reaction, and restored to maximum after weld has been completed.

8. The clamps are opened and the weld removed, after which the operation can be repeated.

If a good many welds of the same size material are to be made, the reactive coil may once for all be set in a given position, and the switch alone be operated.

Welds made with the automatic machine have attained a uniformity not obtainable with very skilled operators working without it. In fact, small, easily-fasible wires can scarcely be welded with certainty with ordinary apparatus. For aluminium especially the automatic apparatus is needed.

The reactive device used in connection with welders is the type recently described by Prof. Thomson, in which a cast is made to more or less cover the primary, the self-induction of which is to be altered. To obtain still a larger range, the windings of primary can be coupled in series, or multiple series, or multiple, by a switch in base of coil.

As mentioned before, the welding of easily fusible metals may sometimes cause a rupture of the secondary circuits, which, owing to its violence and volume of energy, may cause a burnout of the primary if not guarded against. This danger is not very great in the automatic machine, since the end pressure does not depend upon the attentions of the operator. A special device is, however,, used as an extra precaution against all emergencies.

Breaking a high-tension circuit rapidly is not easily done. An arc generally follows the break, and this lengthens the time of the rupture. If the voltage of the circuit to be broken is so low that over the slightest distance an arc cannot follow, the break will be instantaneous. This is the ease with the secondary circuit of welding transformer. The voltage, being ordinarily only one volt, would, even if increased tenfold, not be able to maintain an arc. However, multiplied in the primary it will cause E.M.F. sufficient to pierce through the insulation as ordinarily used. This action is similar to a Rhumkorff coil, in which the interrupter is caused to break under oil or water. The discharge is taken care of by a special apparatus.

While the machine before us is only one type embodying the principal features contained in all, others have been manufactured, or are in process of construction, in which the automatic character has been carried out even further, as in the welding of rings and chains. The present model in its first form is able to produce 250 ft. of chain in a day without any attendance to speak of, the plain wire being fed into the machine at one end and the complete chain coming out of the other.

The automatic principle is, however, not confined to small conductors. We have welders with 40,000 watts output, capable of welding 1 in. copper or 2 in. iron, constructed and working daily on that plan. The projections are determined by adjustable stops; the pressure, the most important of all, is hydraulic, and regulated by an automatic reducing valve, the exhaust being used for cooling clamps at the same time. The primary is controlled by adjustable reactive coils from a constant potential circuit, and is interrupted at the proper time by the clamps. While not all machines to-day are operated on this plan, I am convinced that the evolution of the welding process tends in that direction, and that the welding in future will be, in the full sense of the word, a purely mechanical operation. (H. Lemp.)

A method of welding by electricity, devised by Nicolas Von Benardos, of St. Petersburg, has recently; attracted considerable attention on the Continent, and possesses several features which render it quite distinct from the various processes of fusing and reducing by means of the electric current or the electric arc, proposed from time to time by Siemens, Clowes, Thomson, Wallner, and others. It seems, moreover, to be thoroughly practical. An account of the process, together with a description of the advantages which are claimed as resulting from it, has been given by Prof. Ruhlmann, of Chemnitz, after a careful study of the Benardos process at St. Petersburg; and the following account follows closely his able and interesting articles published recently.

Benardos works directly with the electric arc produced between a carbon pencil as one terminal and the metal to be treated as the other terminal. This has been suggested and tried before. But the carbon was made the negative pole, as it was feared that otherwise the consumption would be embarrassingly rapid. Hence the metal became the positive pole, that is to say, it became exposed to energetic oxidation; and a great deal of the trouble experienced by other experimenters arose from this circumstance. In the Benardos process the carbon forms the positive terminal; it is, of course, quickly consumed, but can easily be replaced; on the other hand, there is a favourable reducing action going on in the fused metal. The great importance of this modification can easily be tested by changing the poles, when the work soon becomes enveloped in a dense cloud of oxidised products. The intense heat of the arc melts even the most refractory metals almost instantaneously; but the action is purely local, like that of the blowpipe, and only those parts upon which the arc plays directly are attacked, the adjoining portions undergoing little change; and the fused mass solidifies and cools very quickly. In the Benardos process the material requires little or no preparation.