I now append a table showing the dimensions of the most important parts of coils up to those capable of giving a 3 in. spark. In the case of the condenser the reader will understand that the figures imply the approximate total superficial area of foil. I leave it to the reader to determine upon the size and number of sheets that he will utilize to obtain the total area required, as this will vary to suit individual requirements.

We must now direct our attention to the larger type of coil - viz., those ranging in size between 4 and 12 in. Before considering the definite dimensions of various large coils, there are certain points dealing with the general design which I venture to suggest, among which some that I am not aware of being given elsewhere may prove of value to intending makers.

Firstly, I need scarcely say that it is essential that the secondary of any coil capable of giving upwards of a 4 in. spark should be wound in insulated sections of never more than 1/8 in. thickness, and that the method of winding in double, yet insulated sections, according to Hare, is not to be beaten, both in point of efficiency and convenience. Perhaps a simple disc, composed of two superimposed thicknesses of paraffined blotting-paper, suffices to separate the sections in the case of coils ranging in size between 4 and 8 in.

Beyond this, let Hare's method be adopted. With 4, 5 and 6 in. coils, I do not consider that the annulus of cotton wound into the section-former preparatory to feeding in the wire is necessary. It is better in this case, to allow the diameter of the central aperture of the section to slightly exceed that of the main insulating tube, so as to admit of a small interstitial space wherein wax may be poured when mounting each section on the tube after winding. In the case of 3 to 12 in. coils, the potential difference between the two extreme ends becomes so great that it is necessary to gradually increase the thickness of the insulating medium separating the primary winding from the secondary at the extremities of the coil are approached. This is effected by winding into the section-former, preparatory to feeding in the wire, thoroughly dry paraffined cotton until an annulus of the desired depth been formed. In the case of 8 in. coils the depth of this may be 0 at the center, gradually increasing to 3/8 in. at either extremity, while with 12 in. coils, 1/4 in. at the center, tapering to 1/2 in. at the extremities.

When mounting the sections of the secondary the insulating properties of the coil may be materially increased by interposing at about every sixth section, a disc of thin sheet ebonite of a diameter slightly exceeding that of its adjacent section and having a central aperture of just sufficient size to permit of its being slipped over the main insulating tube. The thickness may well be 1-16 in., so that in a coil having forty-eight double sections, this would only result in an increase in the lateral measurements of half an inch.

In the secondary of a large coil it is advantageous that the contour of the winding, when finished, should conform with the direction of the magnetic lines of force of the core. The sections may therefore taper from a smaller diameter at the extremities to one somewhat larger at the center.

Before leaving the subject of the secondary let me add one more suggestion dealing with the winding of the sections, and that is, instead of merely feeding the wire through a small trough of wax, let the whole reel be immersed, whilst winding, in a paraffine bath. Then the wire will be heated throughout, and the wax which is absorbed by the insulation thus prevented from solidifying on its way to the section-former - an important point, as the mechanical stability of the sections is dependent upon this.

The iron core, though in every way the very heart or essential part of a coil does not, in my opinion, receive sufficient attention by those who essay to construct their own coils. It is comparatively simple to make the cores of small coils such as those figured above. But the core of a large coil is a more formidable thing. The amateur, or even anyone who is not in possession of a special machine for cutting and straightening the individual wires, will find it impossible to satisfactorily make a large core. Even though he may take the greatest care in endeavoring to straighten each wire separately, he will find that, at most, he can only get his core to weigh two-thirds that of the purchased article, size for size. The reason is that all the wires of those that are machine-cut lie side by side, and parallel, throughout their entire length, with the obvious result that no space is wasted and the core consequently contains the maximum quantity of metal. The reader will find this factor figured in the table given, and if he is successful in his attempt at construction, then by all means let him profit by it; otherwise I strongly urge the purchase of a machine-cut core. In either event it will be necessary to anneal the core, which during the operation should be enclosed in a metal tube.

As regards the primary winding, I do not consider that on the whole any advantage is to be derived from the use of more than two layers of wire. Let each layer, however, be wound, as previously mentioned, in right-handed helices, - that is, the second lying in the grooves of the first. A double covering of cotton to this wire will afford sufficient insulation, provided the whole core, with the primary winding when complete, be first of all heated in an oven and thence transferred direct to a paraffine bath constructed for the purpose, and left therein for about 24 hours, or in other words, until air bubbles cease to be emitted. In all cases where a coil is designed for experimental work, it is advantageous to arrange, as suggested by Hare, that the layers of the primary may be connected either in series or parallel as desired.