CHAPTER IV.

CONDENSERS.

A condenser is an apparatus whereby a charge of electrical energy may be temporarily stored, the amount of energy it will hold determining its "capacity." The capacity of a condenser is measured in micro-farads, the commercial unit representing one millionth of a farad. A farad equals the capacity of a body raised to the potential of one volt by a charge of one ampere for one second at one volt—_i.e._ = one coulomb.

The measurement of the capacity of a condenser is accomplished by the use of a ballistic galvanometer. The latter instrument has a bell-shaped magnet suspended in a coil of fine wire. When a momentary current is passed through this coil the magnet hardly commences to rotate until the current has practically ceased. A beam of light is reflected from a mirror fixed to the magnet on to a scale. The degree of deflection is compared with that obtained by the discharge of a condenser of known capacity, and the capacity of the condenser being measured is deduced by a simple rule. The farad, which is the unit of capacity requiring a condenser of an immense size, is replaced by a commercial unit, the micro-farad—that is, one millionth of a farad.

The original form of the condenser was the Leyden jar, which owes its name from the town of Leyden in Europe.

The Leyden jar is made as follows (Fig. 29): A clean uncracked glass jar with a wide mouth is coated on the inside and outside with tinfoil; sometimes loose tinfoil is filled inside, the tinfoil, however, not reaching more than two thirds of the jar's length from the bottom. A cork is fitted, and through the middle of it a wire is passed touching the inside coating of tinfoil and terminating in a metal sphere outside. A simple Leyden jar can be made in a few moments by half filling a glass bottle with water and wetting the lower half of the outside; a wire run through the cork into the water finishes the job. But this is at least only a makeshift, although a fair amount of current has been collected from a leather engine belt in motion in one thus made.

A condenser can be easily made as follows (Fig. 30):

Procure a clear glass plate, _G_, free from flaws, 11 inches square by 3∕32 inch thick. Give this a good coating of shellac varnish all over, sides and edges. Cut out of smooth tinfoil two sheets, _T_, 8 inches square, and round off the corners with a pair of shears. There must be no sharp corners, projections, or angles to induce leakage. Lay the glass plate on a sheet of paper, and mark its outline thereon with a pencil; then remove it and substitute a sheet of the tinfoil, and mark that. This will enable you to centre the foil. Give one side of the glass plate another coat of varnish, and so lay it on the paper that its outline coincides with the pencil outline. When the varnish has partly dried take a sheet of the trimmed foil, and by observing the pencilled marks you can lay it on the varnished plate exactly in the centre. Lay down the top edge first along this line, and carefully deposit the remainder of the foil in place. Next, with a flat brush full of varnish go over the plate, pressing out any air bubbles, and ensuring both a flat and a well-varnished surface. When this is dry, turn over the plate and repeat the operation on the other side.

If desired, a metal hemisphere of at least an inch in diameter may be attached with varnish, first scraping the foil to make a contact. The whole plate can be swung in a cradle of two silk threads, laid on a glass tumbler, or mounted on end in a shellacked block of wood.

A strip of tinfoil, _S_, attached at the corner can be used as a connector. The plates must be joined in the following manner when two or more are used in conjunction, and a quantity of current is desired. They should be placed so the connecting strips project alternately from each side (Fig. 31), and all on each side joined so as to leave two terminals, one to the 1, 3, 5 plates, the other to the 2, 4, 6 plates, and so on, which, when joined, will have the same effect as would result from the use of two large plates of the same total area. The nearer the plates are together the greater capacity they will have, always supposing the insulation is good, the insulation being known as the dielectric. Another good method, when a high quality of glass can be procured, is to lay the tinfoil on the plates without varnish, piling one on top of the other, tinfoil and glass alternately, and clamping the whole securely, laying a piece of cloth top and bottom to avoid cracking the glass from the pressure. This must be kept from moisture; a strip of paraffined paper stuck along the edges and extra paraffin run on will answer very well.

In constructing these glass condensers, they must be designed to correspond with the coil with which they are to be charged. In the foregoing description we have allowed a margin of 1½ inches of glass around the foil coatings. This will make 3 inches as the maximum distance between the coatings. Although a 2-inch spark from the coil would not jump this interval, a certain discharge will take place, and the less this occurs, the more serviceable the condenser will be. Therefore a greater margin should be allowed for a longer spark than 2 inches.

In the commercial condenser for telephone and telegraph use, paraffin and paper are substituted for glass, as will be described later. Heavy paraffin oil gives excellent results, but its fluidity is disadvantageous.

There is no valid reason why paraffin could not be used on the glass plate condensers, care being observed that it is free from dirt and metallic chips. In fact, the space between the glass plates of the multiplate condenser may be filled in with paraffin, and thereby exclude the air. Only a condenser so built up is not convenient to take apart for experimental purposes.

The foregoing description of a glass insulated condenser was written with the assumption that a good quality of glass be used. But the ordinary window glass is generally useless, and paraffined paper is preferable. The quality of glass known as "hard flint glass" is best, the superior qualities being imported from Europe. This latter is used in the manufacture of the standard Leyden jar for lecture purposes.

Were it not for its cost, the finest dielectric we could use would be sheet mica. Unfortunately sheet mica over 3 inches square is expensive, and becomes rapidly more so as it becomes larger.

Standard condensers for testing are made with mica carefully selected, and retain the charge for the maximum length of time. The built-up mica condenser is immersed in molten paraffin until the same has permeated the sheets, and then the complete mass is put under a pressure until the paraffin is well set.

PAPER CONDENSER.

The paper used in the manufacture of the commercial form is a special thin, tough linen paper carefully selected, sheet by sheet, to avoid pin-holes or flaws, and kept in an oven until used to ensure absolute dryness.

When this cannot be procured, use thin unsized writing paper of a good quality, well dried, and absolutely clean. As an example of the necessity of cleanliness, a light lead-pencil mark would serve to conduct the current entirely from a charged sheet to wherever it terminated, and if suitably located, utterly destroy the usefulness of the apparatus. Ink, which most generally contains iron, will cause trouble, and although some cheap foreign condensers are built up of old ledger pages, yet their efficiency is very uncertain.

The paper used in commercial condensers is from four to seven thousandths of an inch in thickness.

SERIES.

The smaller the amount of surface the less will be the capacity, but the quicker the discharge. The apparatus heretofore mentioned has had the alternate plates connected together in two series, presenting a large surface and rendering a large amount of current. A condenser so made will have a low voltage or potential, but is not so liable to leakage as one made to render a high potential. The multiple condenser of a large capacity will hardly discharge and spark over an air gap requiring a contact of the two electrodes. But a smaller one, consisting only of a single pair of small plates, will spark across quite a considerable air gap.

A number of charged condensers may be put in series, and the resultant potential thereby increased. Cut a number of pieces of paper of the desired size, say 6 inches square, and a number of sheets of foil 3 inches square. Round off the corners of the foil and build up first a sheet of paper, then a sheet of foil in its centre, then another paper and another foil sheet, and so on. There is to be no connection from sheet to sheet, only the inductive action of one on its neighbor. The foil must be considerably smaller than the paper in this construction, owing to the greater tendency to discharge round the edges of the sheets, owing to the greater potential of the current.

When the requisite number of sheets have been built up, leave a sheet of foil top and bottom for connection, tie between two pieces of stout card or board, and immerse in the molten paraffin. When thoroughly soaked, remove and put under pressure until cold. It will be found undesirable to make these with more than a dozen pairs of sheets, but to make a number of blocks of that number for ready service.

Fig. 32 shows the arrangement of the apparatus to charge a Leyden jar, the plate form being connected in a similar manner. The jar is stood upon an insulating support—a dry tumbler will answer—with the ball _B_ connected to one pole of the coil. From the outside tinfoil coating _T_ a wire runs to the discharger _D D_, which is in circuit with the secondary coil, _S_. The discharger balls _D D_ are carefully approximated until the spark just passes, this latter point being of great importance. Were the discharger balls too near the spark would probably pierce the dielectric of the condenser, therefore the balls should be carefully _brought near_ to each other until the exact distance is found. Even if the insulation of the condenser were not pierced, yet a path would probably be opened through which some succeeding discharge would pass, and ruin the instrument.

Another method of charging is to leave an air gap at _B_; then there is not much liability of the condenser discharging back through the coil—an undesirable event, as it would most likely perforate the insulation of the coil.

In designing or using any apparatus intended to hold a charge of high potential, it must be kept in mind how readily points or sharp edges serve to allow the current to pass off—we might almost say evaporate. Given two bodies, one a globe and the other a rectangular block, each well insulated from the earth or any other large body, and the globe would be found to hold its charge long after the block had dissipated all trace of the charge given to it. Therefore round off every edge and angle, projection or point.

In making handles, supports, or any work requiring an intervening high insulation, hard rubber is preferable to glass where there is liability to moisture. When the apparatus is as shown in Fig. 33, the condenser is alternately charged and discharged with a loud noise, the vivid sparks passing across the discharger balls _D D_ possessing great deflagratory powers.

In experimenting with a Ruhmkorff coil it is not advisable to leave the instrument working while the secondary terminals are beyond sparking distance, as there is a great strain on the secondary insulation. Nor is it wise to use only one electrode in an experiment, unless the other is connected to some apparatus of an approximate capacity to that at the other, for the foregoing reason.

ROLLED-UP CONDENSERS.

Now that the condenser has become so important a factor in telephone work, many schemes for cheapening and facilitating their manufacture have been devised. One in particular merits description, the "rolled-up" condenser having come largely into use. The tin-foil is supplied in rolls containing many yards of foil of the requisite width for the condenser to be made. Likewise rolls of paper are provided, exceeding in width, however, those of tin-foil. These rolls are arranged upon horizontal spindles in front of an empty spindle, or mandrel, upon which the condenser is to be formed. A few turns of the paper ribbon are made around the mandrel, then the foil is brought forward and a few turns made, then follows a turn of paper ribbon and another of foil, and finally a paper layer; and the mandrel being rotated, the alternate layers of foil and paper are laid on and rolled around each other on the mandrel until the requisite quantity is obtained. It then becomes an easy matter to cut the paper ends so no contact is possible between the layers of foil. The whole thing is slipped off the mandrel, secured by a rubber band or two, placed in a hot paraffin bath, and left to become saturated while still warm and before the paraffin has time to harden; the cylinder is put under a press and squeezed flat, driving out excess paraffin, and leaving the condenser in a convenient shape to handle. Connections are then made to the foil leaves, and a case of wood or metal completes the work.

There is no reason why aluminum foil or lead foil, or, in fact, any thin sheet metal should not be used in condensers. In telephone work, paper covered with gilt paint was tried, and worked fairly well, but was ultimately rejected in favor of tin-foil. In some cases, when it is desired to construct a condenser for high potential work, the oil-tank apparatus can be used. This is readily made of any desired dimensions, as follows: Procure a square glass jar, such as is made for storage batteries, a few pieces of sheet metal cut to fit loosely in the jar, some glass rods and sufficient clean "transformer oil" or heavy paraffin oil to nearly fill the jar. The sheets of metal can then be hung from the glass rods into the jar, being separated one-half inch, and the oil poured in. Two plates, about 8 inches by 6 inches, will hang nicely into a type D^3 Chloride Battery jar, which is 7⅞ inches long by 9½ inches high by 3¼ inches wide. Altering the relative distances between the plates will give considerable adjustment to this simple condenser, or, if desired, more plates may be inserted and connected up, as in the tin-foil condensers. This type can be made portable, but it is not to be recommended unless no objection is had to emptying and refilling the jar with oil.

ADJUSTABLE CONDENSERS.

In operating large coils, it is convenient to be able to vary the capacity of the condenser on the primary circuit. To make an adjustable condenser presents no more difficulty than a non-adjustable one, simply more labor. For example, the large condenser used with the 6-inch spark coil might be divided into four sections, containing 2000 square inches, 500 square inches, 300 square inches, and 200 square inches of surface (see Fig. 34). Wires leading from the ends of the foil sheets _C C_ are to be brought to the brass plates _G G_. The brass rods _B B_ are connected by binding posts to the coil, each strip being well insulated from its neighbor. Any combination is possible by the insertion of brass plugs in holes drilled between the strips. The plugs must be fully large enough to make good contact on each of the two strips between which they are inserted, and should be turned taper. With the largest coils the condenser and contact breaker are generally mounted separately, and are fully adjustable.

SPECIFIC INDUCTIVE CAPACITY.

Dry air 1.000 Sulphur 2.590 Hard rubber 2.290 Paraffin 1.996 Shellac 2.750 Kerosene 2.225 Paraffin oil 2.710 Castor oil 4.962 Olive oil 3.575

Condensers made with dielectric of high inductive capacity (insulation being equal) will retain greater charge than those made with dielectrics of low inductive capacity. Thus, one made with shellac would be nearly half as great again as with paraffin.

Capacity of a condenser increases with area of foil surface, with diminished distance between foil plates and with increase of insulation.