A condenser consists of two conducting surfaces separated by an
insulator or dielectric. Fig. 53 shows a diagram of a simple condenser
in which _A_ and _B_ are two tinfoil sheets separated by a sheet of
glass, _C_.
[Illustration: Fig. 53. Simple Condenser.]
If _A_ is connected by means of a wire to a static machine a positive
charge will collect on the glass at _A_ and induce a negative charge at
_B_, so that if _A_ and _B_ are connected to a small spark gap the
charge will leap the gap in the form of a spark.
When a condenser discharges through a coil of wire, the discharge
consists of a large number of exceedingly rapid oscillations or
surgings. The first passage of current more than empties the condenser
and it becomes charged in the opposite direction, that is, the
conducting coatings change their polarity. A reverse discharge then
occurs which also oversteps itself and the oscillations thus go on but
become rapidly weaker until they die completely. The time consumed in
the discharge may have been only a fraction of a second, but during that
short period the current perhaps oscillated several thousand times.
If a condenser is discharged through a conductor of high resistance the
discharge passes out slowly, and dies away gradually in one direction
without oscillating. One of the fundamental equations of wireless
telegraphy is therefore that there will be oscillations in a circuit if
the resistance in ohms is not greater than the square root of four times
the inductance in henries divided by the capacity of the condenser in
microfarads.
The capacity or the ability of a condenser to store electricity depends
upon the area and form of the conducting surfaces, the thickness of the
dielectric between them, and a factor known as the specific inductive
capacity of the dielectric. The unit of capacity is called the farad and
is defined as the condenser which would be raised to a potential of one
volt by a charge of one ampere flowing for one second. A condenser of
such a capacity is, because of its enormous size, impractical to
construct, and the unit ordinarily used is therefore the microfarad, or
one millionth of a farad.
Capacity may be calculated from the following formula:
Capacity equals K(A/D),
where _K_ equals a constant depending upon the specific inductive
capacity of the dielectric, _A_ the total area of tinfoil and _D_ the
thickness of the dielectric.
*Leyden Jars.*—Transmitting condensers in a wireless telegraph station
usually take the convenient form of a jar, coated inside and out with
tinfoil and known as a Leyden jar.
The jars should be of good Bohemian or Jena hard glass and coated with
tinfoil only for about three-quarters of their height, as otherwise the
discharge is liable to pass over the top. The tinfoil must be thick to
avoid blistering, and is stuck to the glass with shellac varnish. The
blistering of Leyden jars is a serious fault, for when this condition
exists, the capacity is thereby altered to such an extent that the
period of the closed circuit may be sufficiently altered to throw the
system out of tune and decrease the radiation of energy.
Considerable expense may be saved if the glass jars are purchased and
coated by the amateur. The best jars are those imported from Germany,
which have wide mouths so that they may be easily coated inside with
tinfoil.
[Illustration: Fig. 54. Leyden Jar.]
The jars must be _thoroughly_ cleaned and dried before they are coated.
Give the inside a thorough brushing over with shellac varnish, and
before it is dry, carefully insert the tinfoil and press it smoothly
against the glass. The outside of the jar is treated and coated in the
same manner. The inside and outside of the bottom are also coated by
cutting the tinfoil in circular pieces and shellacking them on.
[Illustration: Fig. 55. "Aerial Switch."]
The whole upper part of the jar is given one or two coats of shellac in
order to prevent the collection of moisture and brush discharging. A
wooden plug fitted in the top of the jar supports a brass rod,
terminating at the lower end in a chain or spiral spring which connects
with the inner coating. When trouble is experienced because of an
imperfect contact between the coating and the chain or rod, a layer of
brass filings an inch or two deep placed in the bottom of the jar will
remedy the difficulty. The upper end of the rod usually terminates in a
small brass ball or a binding post.
The wooden plug or cover is dried in an oven to expel all moisture and
then boiled in paraffin.
Small Leyden jars may be very conveniently made from six-inch test tubes
and mounted in a rack so that the capacity of the condenser will be
adjustable by removing one or more of the tubes. An ordinary test tube
rack such as is used in chemical laboratories serves very well for this
purpose. The tubes should be connected in parallel, that is, all the
outside coatings together and all the inside coatings together.
Figs. 55 and 56 illustrate condensers of this type which are on the
market. The tubes are all separately removable so that the capacity may
be adjusted.
[Illustration: Fig. 56. Amco Oscillation Condenser.]
*Glass Plate Condensers.*—Glass plate condensers offer several
advantages over Leyden jars and are coming into wide use. They are not
so bulky or expensive and, above all, do not blister.
Plate condensers are often placed in a rack and made adjustable by means
of movable contacts. Much the better plan is to place the plates in oil,
as this eliminates all corona or brush discharges and much sharper
tuning is rendered possible. The container is usually a tight wooden box
filled with oil or paraffin after the plates are in place.
It is impossible to state the size of condenser suitable for induction
coils of a given power or spark length, because many factors such as
inductance, length of aerial, etc., which differ in various stations,
influence the capacity. A condenser of convenient size suitable for
coils or small transformers consuming from 250 to 300 watts is that
described below. It is about the proper size for the small open core and
1/4-K.W. closed core transformers, described in the last chapter.
[Illustration: Fig. 57. Clapp-Eastham Oscillation Condenser.]
The glass plates may be secured by removing the emulsion from old 8 x 10
inch photograph plates. Hot water will soften the gelatin on the plates
so that it may be very easily scraped off. Twenty-four plates of this
size are required. The tinfoil is cut 8 x 8 inches, so that an inch
margin is left on all sides. The alternate sheets are connected together
by heavy tinfoil or thin copper foil strips. The condenser should be
placed in a convenient sized wooden box and poured full of paraffin.
The plate condenser shown in Fig. 57 is of .02 microfarad capacity. The
condenser is mounted in a plain wooden box with several binding posts
brought out, so that the capacity may be varied by connecting in various
sections. The condenser is manufactured and designed by the
Clapp-Eastham Company for use with the transformer illustrated in Fig.
51.
It is very necessary to have the transmitting condenser adjustable so
that its capacity may be varied, for the proper value depends upon the
wave length, spark frequency, power and persistency of the wave train.
When the condenser capacity is too small the spark will be somewhat
flaming like an arc, and the potential to which the aerial is charged
will be low. If too much capacity is used the spark will be very
irregular and intermittent.
[Illustration: Fig. 58. Methods of Varying Capacity.]
Fig. 58 shows condensers connected in series and in parallel and a
combination of the two. Two condensers of equal capacity connected in
parallel have twice the capacity of one, while in series they will have
only one-half the capacity of either. This may be otherwise stated as
the capacity in series is equal to the reciprocal of the sum of the
reciprocals of their capacities separately. By this means of connecting
either in series or in parallel almost any desired adjustment of
capacity may be brought about.
Oftentimes a high voltage may be divided between two condensers by
placing them in series and thus using them safely on a voltage which
would rupture either one alone. For example, two condensers built for
20,000 volts and to be used on 30,000 volts could be made to perform
this duty safely and only undergo a potential of 15,000 volts, which is
a large margin of safety.
It is obvious that if the capacity of the circuit were to remain
constant, four condensers connected up in series-parallel would be
necessary.
In case of several Ley den jars or condensers connected up in a
transmitting circuit, the leads or conducting wires connecting the
various units should all be of as nearly the same length as it is
possible to have them.
The resistance of metallic conductors to high frequency currents is
several times their normal resistance to constant currents. The larger
the diameter of the wire the greater is this ratio. This increase of
resistance is due to the fact that the high frequency currents permeate
wires only a very short distance. In the case of copper, the depth is
only about one three-hundredth of an inch and with other metals much
less. Therefore it is advisable to use as leads and conductors of large
condensers, stranded wires or flat ribbons of sheet copper in order to
present more surface and offer less resistance than solid conductors of
an equal cross sectional area.
Iron must never be used, as its resistance to these currents is over
fifty times that of copper.
After connections are once established between the jars or the condenser
units, they cannot be altered nor the capacity changed without re tuning
the circuits afterwards.