CONTENTS

Certificate

Declaration

Acknowledgement

Preface

Abstract

CHAPTER PAGE

1 General Survey 1

2 The Transformer 2

3 The Condenser 8

4 The Oscillation Transformer 10

5 The Interrupter 13

6 The Construction of the Boxes 19

7 Assembling 21

8 Theory of the Coil 23

9 Dimensions of 7" Standard Coil 26

Chapter 1: GENERAL SURVEY

By far the largest and most interesting branch of science is electricity, for Maxwell has
proven mathematically, and Hertz verified experimentally, that light is an ejectromagnetic
disturbance in the ether, and thus added that subject to the realm of electricity. Amongst the
various phenomena of electricity, those of the high-tension current are the most interesting
and instructive. With such a current all the wonders of the Geissler and Crookes tubes may be
seen.With it waves for wireless messages may be sent out into space, and a great number of
other experiments carried out. It is the purpose of this book to show how a satisfactory
apparatus for producing these currents may be constructed, and also to describe a few of the
uses for such a coil. The apparatus, as described in this book, is most commonly known as the
Tesla High-Frequency Coil, and consists, in general, of four parts :

i. The Step-Up Transformer;

2. The Interrupter;
3. The Condenser;
4. The Oscillation Transformer.

Each of these will be fully considered in subsequentchapters. Before entering upon the
description of the Tesla highfrequency apparatus, however, it would be well to make a few
general remarks which are of the greatest importance. Throughout the whole work of
construction the most exacting care must be given to the matter of insulation. All thewire
used must be carefully tested, and each layer of wire in the transformer must be thoroughly
shellacked, and then insulatedfrom the next layer, by two turns of carefully oiled paper. In the
condenser, which is really the vital part ofthe apparatus, the glass should be of the best grade
obtainable. It must also be free from all air bubbles. It is inthe high-frequency apparatus,
however, that the greatest care as regards both construction and insulation must be taken. The
secondary consists of one layer only of No. 32 B. & S. gauge, double cotton-covered wire,
wound on an octagonal frame, formed of strips of vulcanized fibre fastened to two end pieces
of wood. When winding the wire, care must be taken that no two adjacent wires touch, for
that would cause a short circuit. When the wire is completely wound, it is given about five
coats of shellac, not only to act as an insulator, but also to prevent any slipping of the wires,
The primary consists of a thin band of copper, making two and a half turns around a circular
frame surrounding the secondary. The frames on which the primary and secondary are wound
must be \'ery firm and sulDstantial, so that ai occasional jar will not displace any of the wires
on the secondary. All connections must be soldered, and the connecting wires run through
glass tubes. When the apparatus is finished, two carefully made box6 must be constructed.
These must be oil tight. This iii General Survey 3 accomplished by mortising all joints, and
then giving the boxes, especially the joints, about four or five coats of shellac. Into one box
the transformer fits, and into the other the condenser and oscillation transformer. Then the
boxes are filled with pure paraffine oil, which is the only efficient insulator for these high-
tension currents. Some who intend to build this coil will think that all these precautions
regarding insulation are extreme, but it will be found that, in dealing with high-frequency,
high potential currents, too much care cannot be taken, for "Good insulation is the key to
success in high tension work”

Chapter 2: THE TRANSFORMER

The transformer — sometimes called a converter—ismerely an induction coil that is

connected directly to the alternating-current mains, without the use of an interrupter, and is
used to raise or lower the voltage. In a transformer the number of watts in the primary equals
approximately the number of watts in the secondary. In the case of any step-up transformer,
the ratio of the number of volts in the primary to those set up in the secondary is nearly the
same as the number of turns of wire in the primary to the number in the secondary; but the
amperes decrease in the inverse ratio. The transformer used in the coil described in this book
is of the common induction-coil type, oil-immersed, step-up transformer. It takes the
alternating current from the mains at no volts or 55 volts, and steps it up to about 10,000
volts. The efficient working of a transformer depends largely upon the design of the core. The
iron used must be of high permeability and should have little retentivity. A straight core is
always best to use; for, on the fall of the current from its maximum value to zero, the
magnetic flux falls from its maximum value, not to zero, but to a value which depends The
Transformer 5 on the residual magnetism. The residual magnetism in an open circuit is much
less than in a closed magnetic circuit, so that when the current suddenly becomes zero, the
magnetic flux drops lower in an open circuit than in a closed one. As the electromotive force
in the secondary is proportional to the fall in the magnetic field, it is greater with a straight
core than with a closed circuit of iron. The coil designer is obliged to determine the length of
the iron core from the experience of others, as the mathematics for calculating it is too
complex, although simple and useful in the case of closed circuit transformers. If the core
is made too long the primary magnetizing current will be too large, while if made too short
the secondary coils would have to be made of too large a diameter to be efficient. There
is, therefore, a certain length which will give the best results. In the case of this transformer
the length of the core was determined after having gained all possible information from
certain eminent men who had made a life study of these matters; in fact, all the dimensions of
the transformer for this special use were determined in this way. The iron core is made up of
pieces of No. 20 or 22 B. & S. gauge iron wire 18" long. The wire is first cut nearly to
size with a pair of pliers, and,' when assembled, the ends of the bundle are sawed off square
with a hack saw. An ordinary piece of iron pipe, a little less than 18" long, and having
an internal diameter slightly less than 2", is tightly fille with these wires. When putting the
wires in, stand the pipe on end on a smooth surface, and force in each wire untilit hits this
surface. When the bundle is finished, the upper end is sawed off «-ith a hack saw to exactly
i8." The tube containing the iron wires is now placed in a coke or coal fire and left there until
the fire burns itself out, thus insuring slow cooling. This heating and subsequent slow cooling
so softens the iron wires that their retentivity is reduced to a minimum. When cool, the wires
are taken out and sandpapered to ^emo^'e any superfluous oxide. They are then, one by one,
dipped into boiling water, wiped dry, and while still warm are coated with thin shellac
varnish. When the shellac is dry they are again packed, as tightly as possible, in the pipe, to
hold them in the desired shape. Then, while still packed closely together, they are forced
slowly out of the pipe; starting at the end thus released, they are tightly bound with a narrow
cotton bandage, which can be obtained from any surgical supply house. The bandage should
be between one and two inches wide, but no more. When the entire core is wrapped with this
cloth, the cloth should be heavily shellacked. The ends of the core are now filed flat and
smooth; after this it is put in a warm place to dry thoroughly, when it will be ready for the
primary winding. The use of the insulating varnish on the iron wires is to arrest eddy currents
as much as possible, thus preventing the iron wire from becoming heated and energy wasted,
which would lower the transformers efficiently. The primary is wound in two sections of two
layers each, one above the other. No. 12 B. &l S. gauge, double cottoncovered copper wire is
used. About 2 pounds will be The Transformer 7 required. The primary may be wound by
hand, by erecting two wooden supports 17" apart, and having a 2" hole bored in each, to
receive the iron core. Then, by turning the core by hand the wire may be wound fairly well.
But as it is rather diffcult to wind the wire tightly in this way, it would be more satisfactory to
wind it in a lathe, if the amateur has access to one. To mount it, cut a half-inch piece from the
end of the pipe in which the core was formed, and slip it over the extreme end of the core.
Make the ring fit as tightly as possible by placing between it and the core a few strips of tin or
other thin sheet-metal. Now clamp it firmly in the chuck. The other end of the core should
also be fitted with a half-inch piece of pipe and supported at this place in the steady rest. The
one piece of pipe is used to prevent any of the wires from being forced in unequally at the
points where the chuck clamps it, and the other to afford a smooth bearing surface for the
steady rest. If there is any tendency for the core to slip out of the chuck, the tail stock, with
the centre removed, may be pressed up against it. About I ft. from the end of the copper wire
take a couple of turns of tape around it. At this point bind the wire to the iron core, about i"
from its end, by taking several turns of tape around it. Proceed now to wind the wire tightly
and closely to within i" of the other end. Here the winding of the primary is stopped for a
short time in order to give the wire a good coat of shellac. After the shellac has dried, another
coating is given it, and then the second layer is wound on while the wire is still wet. When the
windinghas come to within about six turns of the starting point, a piece of tape doubled back
on itself is laid on the first layer, with its ends projecting beyond the unwound portion of the
second layer. The looped end of the tape must be on the outer side of the winding. See Fig. i.

The winding of the second layer is finished over the piece of tape, the last turn being brought
through the loop in the tape. The loop is drawn tight by pulling on the other projecting ends.
In this way the last turn is kept from slipping off. By using this method or fastening it is
unnecessary to use any bobbin heads for the primary; this is a decided advantage, as, with a
removable primary, bobbins are always getting loose. The wire is cut off about 2' from this
ending The Transformer g in order to allow plenty of wire for making the various
connections, which will be described in a later chapter. When this layer is thoroughly
shellacked, the first section of the primary is complete. The second section is wound directly
on top of the first, starting at the same end, and being sure to wind in the same direction. Each
layer when wound is thoroughly shellacked, and the last turn is fastened in the same manner
as before. If the wire has been put on carefully 164 turns can be wound in the 16" and the
total diameter will be 2|".
The secondary is wound in four sections. It will first be necessary to procure two micanite
tubes, the one fitting tightly within the other. The inner diameter of the smaller tube is a trifle
greater than 2!", the external diameter of the larger one being 35". The length of the tubes is
18" and their thickness -J". Now turn out a wooden rod so that the larger tube will fit around
it tightly. Mount the rod in the lathe with the tube on it, clamping one end of the wood in the
chuck, and supporting the other end on a centre. With a thin parting tool, cut off seven rings,
three i" wide, and four, 3^" wide. If the amateur has no lathe the rings may be cut off in a
mitre box. Out of some quarter-inch sheet-fibre, cut eight circular pieces, 6" in diameter and
having a 3" hole in the centre. Slip one of the 1" rings on the smaller tube, and with Le Page's
glue fasten it to the extreme end of the tube. Next slip on one of the circular discs of fibre,
and then one of the 3J" rings, fastening them with glue. Two more discs are put on, and then
another 3^" ring. After this comes another disc and a i" ring, followed by a disc and a 3,{"
ring. Then put on two more discs and the remaining 31" ring. This is followed by the
remaining disc and i''ring. Be sure that each ring is carefully glued in place.Before putting on
the discs, small holes should be drilled in them, through which to carry the wires. The
romijlctedbobbin for the secondary is seen in Fig. 2. The discs numbered2, 3, 6, 7 ha\'e the
holes fcjr the connecting wires drilled on their inner edge, while the others have them
drilledabout \" from their outer edge. Obtain a wooden rod uponwhich the secondary bobbin
will lit tightly. It should be i8i" long.
If the coil builder is skilled in winding wire in the lathe, the winding may be done there much
more rapidly than by hand; but for an amateur, who has had but little experience with lathe
windirig, or for one who does not possess a lathe, the following method is given. In winding
in the lathe, great care must be taken that the wire is not snapped off when the end of the
layer is reached, and while the jjaper is being wrapped on before the next layer is wound.
For the hand winder, the wooden rod, on which the secondary bobbin fits tightly, is drilled in
at both ends for about 4" with a litde less than a {" hole. Pieces of |" iron are then driven into
these holes, to serve as an axh,-. They should fit tightly, so as to turn with the cylinder. About
6" shouldproject at one end, which is bent into a handle. 1 !/' at theother end is sufficient for a
bearing.
The standards are made of '(" oak, fastened jg|" apart  to a baseboard 2' long. A piece of oak f"
square and 2"long is fastened with two screws to the top of each standard,to serve as a cap. A
I" hole is then bored with its centreon the joint. This allows the cylinder to be taken out of its
bearings when necessary. Two iron washers are slipped over the shaft at the short end to act
as a thrust bearing, and two washers, with an open, steel-wire spring between them, are put
on the other end. This will give the friction required to enable the amateur to stop the winding
at any time, and still be sure that the cylinder will not rotate and soloosen the turns of wire.
The dimensions of the winder are seen in Fig. 3.

As the wire must be wound under some tension, and as it is tiresome to give the required
tension by letting the wire run through the hand, the holder shown in Fig. 4 was devised. It
consists of an axle which fits the spool tightly, and which is 4" longer than the spool. There is
a thread cut on one end of this axle for about 2" It is then mounted in twowooden standards
fastened to a baseboard. Iron washers are put between the spool and the standards for the
spool to bear on. An open spring made of piano wire is slipped up on the threaded end of the
shaft, outside of the standards. A washer and a nut are now put on to give the required tension
to the spring. A lock nut is put on to keep this nut from turning. Care must be taken to detect
any breaks that may occur in the wire. When winding the wire it quite frequently happens that
a little kink will cause a break; but because it is covered by the cotton insulation, it will be
wound on the bobbin, unknown to the coil builder. To detect these breaks immediately, the
authors used the following method. A ring is cut out of a piece of sheet brass or copper. It
is }" wide and 3" in diameter. This is fastened by several flat-headed brass screws to one side
of the spool on which the wire is bought. If the wire has been bought from a reliable dealer,
the inner end will be found projecting outside of the reel. This wire is soldered to the ring on
the outside of the spool. A strip of sheet copper or brass, which is of such a length that it will
bear on the ring, is fastened to the upper end of the standard, on the side on which the ring is.
From here a wire is led to one pole of a dry cell.

On the winder a strip of sheet metal is fastened to one of the standards. It is best to fasten it to
the one farthest up from the handle. It is bent so that it presses firmly on the projecting axle,
which has been polished to make good electrical contact. A wire is then led from the brush on
the standard to a binding post on the baseboard. A telephone receiver is now connected in
series with the binding-post and the other pole of the cell. A watch-case receiver, with a head
attachment, is the best to use. If the amateur has only the Bell receiver, an attachment to hold
it to his head can easily be arranged.
If the amateur prefers he may use a sensitive galvanometer. Everything is now ready for the
winding of the secondary. To begin, pass about i' of the wire through the hole in the bobbin
heads numbered 2, 3, from the side on which bobbin head 2 is. The insulation should be
scraped off of the end of the wire for about 2", and then this bare part should be tightly
wrapped on the axle between the washer and the wooden cylinder. It will now be seen that
there is a complete circuit through all the wire on the spool. The diaphragm in the telephone
receiver is drawn down or if a galvanometer is used, the needle will be deflected. If the wire
should break, the diaphragm will return to its normal position and a click will be heard, or in
the case of the galvanometer the needle will return to the zero position. When this happens
the break should be located and the wire soldered. Acid should not be used in soldering, as a
little left on the wire will corrode it and spoil the electrical connection. Rosin is the best thing
to use as a flux.

Chapter 3: THE CONDENSER

A CONDENSER is an apparatus for accumulating a large quantity of electricity on a small

surface. The form may vary considerably, but in all cases it consists essentially of two
conductors separated by a non-conductor or dielectric, and its action depends entirely upon
induction. The thinner the dielectric and the greater its specific inductive capacity, the greater
is the capacity of the condenser. A thin dielectric, however, cannot withstand a high potential.
Besides the thickness, the dielectric strength depends on the character of the material.

The condenser used in this apparatus is especially designed for long continued use on high
voltages. The dielectric used is glass and the plates are made of sheet brass. When finished
the condenser is immersed in pure parafi&ne oil. For the dielectric 95 sheets of glass, yV
thick and io"xi2" in size, should be obtained. These sheets may be had cut to size for about
nine or ten cents apiece. In purchasing them each sheet should be examined carefully to see if
there are any air bulAles. If any are found in a sheet of glass it should be rejected.
The brass used is number 32 or 34. Forty-six sheets8"xio" are required. In one of the corners
of the shorter side is a tongue 2" wide and i j" long. A j" hp is bent across the top of this
tongue. If the brass can be had in rolls 8" wide, a little more should be obtained and the
tongues cut out of it. They should be cut 3"xi|". The extra i" is for soldering them to the
plates. Rosin, not acid, should be used as a flux. These tongues should always be soldered in
the position shown in the figure.
As a rule 12" is the only width that can be obtained in most places. When this width is used
the tongues are cut right on the sheets. The condenser occupies the part that has been
constructed for it in the oscillation transformer box. The frame in which the condenser is built
up is made out of well dried pine. The base is made of a piece of ]" x 11" pine, iif" long. The
sides are cutout of y'xi2" material lof" long. The ends are also cut from j"xi2" wood and are
11" long. The sides and ends should be planed up smooth on both sides, so as to make them a
little less than i" thick. The completed frame is seen in Fig. 7.
Place the condenser frame on a table or some other flat surface, with one of the ends down.
Before putting any of the glass sheets in the frame, they should be carefully wiped clean so as
to remove any dust or moisture. Commence by putting two glass sheets in the frame so that
they reach the bottom of the frame. Place a brass sheet on top of these glass plates so that
there is a i" margin of glass all around the sheet, except were the tongue comes out. If the lip
on thetongue has been bent carefully it will just fit up against the sheets of glass. Without
displacing the brass lay two sheets of glass on top of it. A brass sheet is next put in, but in this
case the tongue comes out on the reverse side. There should be a i" margin around the brass
as in the previous case. After the brass come two more sheets of glass. This process is kept up
until the 46 sheets of brass have been put in place. Three sheets of glass are placed on top of
the last brass plate. If the glass and the brass used are the size called for in this book the last
sheet will just go m tightly. In forcing in the last few sheets it is a good idea to lay a cloth
between the last two, to take up any excess pressure, which would otherwise crack the glass.

Set the condenser upright, when finished, and solder a piece of No. 16 bare copper wire about
3' long to each of the lips in turn, down the one side, and the same is done on the other side.
Two leather straps should be fastened to the sides of the frame to lift it by when lowering into
the oscillation transformer box. The object for building the condenser in a separate frame,
instead of in the division in the oscillation transformer box, is to facilitate moving should the
condenser ever require rebuilding, due to the rupturing of the glass sheet . when the
condenser is placed in the box, the end which has the three glass sheets should be placed
against the partition, that is, nearest the oscillation transformer. This is to present the spark
from the oscillation transformer breaking through into the condenser, instead of following its
air path.
Chapter 4: THE OSCILLATION TRANSFORMER

It is this part of the apparatus, so simple in construction, in which the most care as regards
insulation must be taken. The success of the whole apparatus depends on the care with which
this part is constructed. The least fault, such as two wires touching, or many other small
similar mistakes, may cause a short circuit and require its reconstruction. It is not wise to
hurry the work, as it will be necessary to reconstruct it if careless.
The end supports are made out of any suitable piece of wood. The two supports for the
secondary are 8" in diameter, and from f" to i" in thickness. Eight equidistant points are
marked off on the periphery and slots J" deep and \" wide cut at these points. See Fig. 8.

These slots are for the fibre strips, on which the secondary is wound, to fit into. These strips
are 17" long and \" square and are cut from the best vulcanized fibre obtainable, eight being
required. In each end of the strips a hole is drilled and countersunk to receive a small brass
screw, which is to fasten them to the end pieces. A wooden rod about i" in diameter is now
obtained and a shoulder turned on each end.
A good way to straighten them is to lay them between two boards, placing some heavy
weight on the top board, and lea\c them thus over night. About 4 or 5 ounces of Xo. 28 B. &:
S. gauge double cotton covered copper wire is required for the secondary. There are several
equally good \\'a\-s in which the wire can be wound on the frame. Two of these methods will
be described, as the authors have found them both satisfactory. The first method is intended
for those that have a lathe at their disposal. A cylinder of wood 4" in diameter and 18" long is
first turned out, and on it are screwed the eight strips of fibre, so that their ends are in line,
and that they strike one another. If the amateur has a couple of clamps made of strips of sheet
iron with a bolt through the ends, it will greatly help matters by clamping them around the
strips, about 6" apart, and moving them as needed. A light cut is first taken off the strips, and
then they are polished with a file and sandpaper. A No. 18 thread is now cut, starting about i"
from the end of the strips to within an inch of the other end. It should be cut just as deep as
possible. In order to make a clean cut the tool must be cry sharp and several light cuts should
be taken instead of one. When the strips are mounted on the frame again it will be seen that
there is a continuous groove in which the wires will lie without touching one another.
The secondary frame is now supported between centresin the lathe so that it just turns easily.
Around one endof one of the strips wrap about i' of Xo.2S wire. This isto be used for
connections. Starting at this point wind thewire tightly on the frame, always keeping the wire
in the groove cut for it. About i' extra should be left at the end for making the connections.
When finished, a hea\y coat of shellac is gi\'en to the wires where they rest on the fibre
strips. When this is dry all the wire is heavily shellacked. A soft brush should be used so as
not to displace any of the wires. The secondary should now be placed in a warm
place to dry.
The following method can be used in case a lathe is not available. A spool of silk thread and
some silk are required. There must be enough silk to make two turns around the secondary
frame. It is wrapped on tightl}', and smoothl}'. and shellacked in place. It might be mentioned
here that shellac dissoh'ed in wood alcohol should not be used. The wire is then wound on,
starting about i" from the end up to within an inch of the other end. A silk thread is wound on
at the same time between the turns, to keep them apart, ^^'hen it is all wound the wire
is heavily shellacked and the frame put in a warm place to thoroughly dry.
Get two pieces of J"xi2" pine 12" long and find the centre of each by the intersection of their
diagonals. With these points as centres describe two circles, with radii 4I'' and 5^' on each
board. Di\"ide the circles ha\-ing the radius 5}" into 36 equal parts. These points can be
located with a protractor or mariner's compass, there being one e\ery 10°.At each of these
points drill a \" hole. In each corner of these boards cut out a piece i"x i". See Fig. ii.
At a planing mill buy eighteen -]" dowels, 36" long. These may be had for one cent apiece.
Each one should be cutin two, which will gwe you thirty-six \" rods, 18" long.Fasten one of
the 12" end pieces to each of the ends of thecompleted secondary frame, by four flat-headed
brass screws.See that the circular end pieces of the secondary just fit on the smaller circles
drawn on the boards. The ^" rods are now driven into place, thus forming a circular cage on
which the primary is wound. For the primary get an 8' piece of No. 36 copper ribbon i" wide.
AAYap the end of the ribbon once around the end of the dowel shown in Fig. 12 and solder it
in place. A piece of copper wire should also be soldered on here. Starting from here take two
and a half turns around

the secondary frame. This will bring you to the end of the dowel diametrically opposite the
dowel from which you started. Wrap the ribbon once around this dowel and solder it in place.
A piece of wire should also be Fig. 13. — Completed Secondaryor Oscillation Transformer.
soldered on. The turns of the ribbon should be equally spaced. Get four pieces of i"xi" pine
183" long and fasten them as braces from the one end of the frame to the other, the pieces
fitting in the corners that were cut out for them. Turn two bushings out of a piece of f" brass
or copper rod. They are shown in Fig. 14. The length over all is V' and the shoulder is I"
thick. A §" hole is drilled down the centre, and the shoulder is drilled and slightly
countersunk at two opposite points to recei\"e two small brass screws. Two pieces of
vulcanized fibre or ebonite J"xi'', 2" long, with a 5" hole drilled h" from the end, are fastened
to the middle of the top side of the primary frame. The bushings just made are fitted into the
holes and screwed in place. The wires from the secondary are soldered onto these bushings.
when this is done the oscillation transformer is finished and all that needs to be done is to
connect it up properly. The completed oscillation transformer is seen in plate III.
Chapter 5: THE INTERRUPTER

Let us now consider that important part of the apparatus, the primary spark-gap. The function
of this, as previously stated, is to provide a path of high resistance until the condenser is
charged to its full capacity. Then it suddenly breaks down, and allows the current to surge
back and forth across it, until the current is damped out by resistance and other factors in the
circuit. After the oscillations have ceased, the ideal spark-gap should return to its maximum
disruptive strength before the condenser can be charged by the next cycle from the secondary
of the transformer. In practice this is far from being the case. The air between the discharging
balls becomes heated, and offers a comparatively low-resistance path for the current. This
results in an arc being formed, which prevents the condenser from performing its function.
The mechanical problem which confronts the designer is to find some way to get rid of this
heated air and thus prevent the arc being formed. The object of this chapter is to show several
ways of partially accomplishing this result, but in no case perfectly. The simplest form of
spark-gap for the primary of the oscillation transformer, one which has given fair results with
the writers, consists of merely two adjustable brass balls. No provision is made for blowing
out the arc that forms, so that considerable of the energy of the transformer is wasted.
Nevertheless sufficient oscillations are set up to bring the coil to within i" or 2" of its
maximum length of spark discharge.

THE SIMPLE AIR-GAP

Two standards of brass J" in diameter and 4" long are mounted on a piece of hard rubber.
They should be about 5" apart. A -^^" hole is drilled J" from the upper end of each rod and is
taped with a standard \" machine screw tap. These two holes must be in line. Next two pieces
of \" brass rod 3!" in length are threaded for the whole of their length to fit the holes tapped in
the standards. Two balls Y in diameter are turned out of brass or are procured from a dealer in
physical supplies. These balls are drilled and threaded to fit in the brass rods. Two vulcanized
fiber rods are turned out of f" rod, 3" in length, and a \" shield 2" in diameter is screwed on
one end of each handle. The handle is drilled and tapped fori" to fit the brass rod on which it
is screwed. The shield is to safeguard the operator's hand from sliding off the fibre handle and
coming into contact with the transformer current, which would probably be fatal. Connection
can be made to the brass standard near the bottom by drilling

.
a small hole through them, and then drilling and tappinganother hole at right angles to the
first, for a thumb screw to bind the wire. A set screw at the top of the standards to clamp the
rods in place after they have been adjusted will be a convenience to the operator. See Fig. i6.

THE AIR-BLAST INTERRUPTER

In order to make the air-gap more efficient, getting rid of much of the heated gases between
the spark terminals, a mechanical means can be used of forcing in cold air, thus driving out
the heated gases and keeping the resistance much higher. To efficiently accomplish this a
piece of f" hard glass tubing is drawn out into a nozzle having an opening ^\" in diameter.
This is mounted on a brass standard and is connected by means of a rubber tube to a foot
bellows such as is used in the laboratory to operate a blast lamp or any other suitable supply
of compressed air. A good blast of air will effectually wipe out any arc that tends to form,
thereby increasing the disruptive length of the bright oscillation transformer discharge. The
operator will find, however, that it is a tedious task to pump a foot bellows, occupying so
much of his time as to handicap him in performing experiments with the high-frequency
discharge, and he will soon decide that the best policy is to construct either a magnetic wipe-
out or a motor-driven interrupter.

THE MAGNETIC INTERRUPTER

The magnetic blow-out is well suited for those who have a source of direct current at their
disposal; either the no-volt lighting circuit or a suitable battery current. For those who have
only the alternating current and who wish to use the magnetic wipe-out the writers have
added to this chapter a simple home-made electrolytic rectifier of their own designwhich will
give a current suitable for magnetizing the magnet of this interrupter.

Two standards of the same form as those used in the simple interrupter, but 6" in length and
having spark balls I" instead of J" in diameter, are mounted on a hard rubber base 10" x 7".
The fibre handles and shields are also necessary for this interrupter. Two electromagnet
bobbins 5^" long and having an iron core |" in diameter with fibre heads 3" in diameter are
procured. An iron yoke made from i"xj" iron 7" long has 2 holes drilled in it in the middle i^"
from both ends, and the bobbins are fastened to it by a screw in the core. Two polar pieces of
f" square iron, filed into an egg-shaped point are screwed to the upper ends of the core. The
bobbins are wound full of No. 14 B. & S. gauge cotton-covered magnet wire, if they are to be
operated on a battery current of from eight to ten volts such as a plunge battery. If a direct
current of 1 10 volts is available No. 24 should be used. If the rectifier described at the end of
this chapter is used No. 22 wire should be used as the voltage of the rectified current on no
volts alternating is about 90 volts. The magnets must be thoroughly saturated in order to give
the best results.
THE MOTOR-DRIVEN INTERRUPTER

This interrupter is the one the writers used in their earlier experiments with the 7" standard
coil described in the latter part of this book. It consists essentially of a fan motor run on the
alternating current at no volts, driving a brass disc having a number of projections bolted
around its face, and a brass oscillator so mounted that the distance separating it from the
projections on the disc can be varied at will. The motor may be of any suitable design that the
builder may possess. A small battery motor rimning on direct current can be pressed into
service if the amateur doesnot care to go to the expense of purchasing a fan motor or has not
the facihties for building one. The directions for building a suitable induction motor are given
at the end of this chapter.

To make the disc for this interrupter turn out of \" sheet brass a circular piece lo" in diameter.
If no lathe is available it may be procured at a model maker's quite reasonably. Lay off on its
face two concentric circles, 8" and 9}" in diameter respectively. Di\ide the inner of these into
thirty equal divisions and draw radial lines from the centre of the disc through each of the
points marked o£F, thus dividing the outer circle into the same number of equal divisions.
Drill a hole through each of the points laid off on both circles and tap them to jQt a standard
4-36 machine screw. A number of brass angle pieces made by bending j'g" brass into the form
shown in the figure are procured. Two holes are drilled and tapped in each one to fit a
standard 4-36 flat'-headed machine screw. These pieces are screwed to
the brass disc with j" screws. A \" hole is drilled in the centre of the disc and three ig" holes
are drilled on a circle having a radius of |". Nextturn out a circular block of hard rubber 2" in
diameter and the same shape as in Fig. 20. The brass disc is screwed to it with three brass
wood screws {" long and the whole is fastened to the shaft of the motor so as to be well
insulated from it. To mal<:e electrical contact with the brass plate a brush to bear on the back
near the centre is cut out of a piece of y^" sheet spring brass. This piece should be
10" long and Y wide. It is mounted on a piece of hard rubber with a suitable binding-post, so
as to press against the back of the disc.
A SMALL SELF-STARTING SINGLE-PHASE INDUCTION MOTOR

To build a motor, to run on the single-phase alternating current of no volts, suitable for
running a fan interrupter, is perhaps ver}' difficult. The builder will require more tools and a
much greater knowledge of machine shop practice to construct an efficient motor than to
build all of the parts of Tesla apparatus combined. For those who have had but litde
experience in motor construction, the writers suggest that the amateur purchase an
alternating-current fan motor or a suitable direct-current batter}- motor. The following
description of the building of this motor is given in order that this book may be complete in
itself and so that the coil builder will have all the necessary data to build the complete
apparatus without reference to other works. The first step in the construction is to make the
necessary patterns for the base and yokes. There are two castings required. The base
supporting the punchinss for the stator is cast directly on the standard which supports the
motor.
It is assumed that the amateur pattern-maker is aware that an iron casting is smaller than the
pattern from which the mould was made, therefore shrinkage must be allowed for in the
pattern in order to be sure that the casting will be large enough. One eighth of an inch to the
foot is about the proper amount. Turn out a circular piece 5 " in diameter and i" in thickncss.
Then a rod i" in diameter, and 6" long, swelling into a graceful enlargement at the lower enrl,
is turned out and fastened with glue and nails to the centre of the < in iilar disc.

A piece of wood is cut out of ;" stock to the form shown in the figure. The radius of curvature
of the arr must be 2\" so that the stator will fit it accurately. This jjiere is glued and nailed to
the top of the upright. The whole pattern isgi\cn two coals of l)csl shellac \'arnisii, containing
suOicicnt lampblack to make it jet black. 'I'his tompktt's llic pattern for the base. In order to
pro\i(le a su|)porl for (he bearing of the rotor shaft we must make a pattern lor a yoke. To
make this pattern we turn out of \\" sloik a circular disc 5I" in diameter and of the same form
as in the drawing. Two castings are made from this palleni, one to lit each end
of the stator. They not only ser\e to furnish bearings for the rotor, but also to enclose the
entire motor and thus keep out moisture and dust. After the patterns arc shellacked they
should be sent to a foundry where the castings can be obtained quite reasonably. It requires
one casting of the base and two of the yoke or journal. When the castings arc obtained they
should be chipped and all the roughness filed off.

AN ELECTROLYTIC CURRENT RECTIFIER

It was only at the last moment that the authors decided to make public the results of their
experiments on an electrolytic current rectifier, which has proven highly satisfactory. Its
advantages are that it is easily and cheaply built, it requires only slight attention, its efficiency
is very high, and the current which a small set will rectify is very large. To a great many it
may seem out of place in putting in this description of a rectifier, which is entirely foreign to
the Tesla apparatus. The reasons for so doing, however appeared, to the authors at least,
sufficiently great, for if the amateur constructs the magnetic wipe-out he will need a source of
direct current at about 80 or go volts pressure, since this current can hardly be obtained from
the lighting circuits which are generally alternating or from batteries.
Then having this source of direct current he will be able to substitute a D. C. motor for the
induction motor described in this chapter.
The greater number of rectifiers now on the market use the method of choking out the one
half of the alternatingcurrent wave and it is to this fact that their low efficiency is due. The
high efficiency of the apparatus devised by the authors depends on what might be called the
alternate path connection or method; that is, there are two paths for the current to traverse,
one of enormous resistance and one of very low resistance. The idea of this form of rectifier
came to the authors in the following way. They were experimenting on some cathode tubes of
peculiar construction, using a 12" induction coil. The current from the secondary of this coil
is oscillatory in character, of course. It was observed that the discharge through the tube was
^-ery unsteady, especially when the interruptions were not very rapid. A line of experiments
was carried out to determine the cause of this unusual' effect, with the result that the
resistance was found to be enormously greater for currents in the one direction through the
tube than in the opposite direction, due entirely to the difference in the forms of the two
electrodes.
After discovering this fact they wondered if some electrolytic cell might not be made which
would possess the same properties and could be used to rectify the ordinary alternating
currents. From a previous study of the effects of various electrodes on the electrolysis of
certain solutions we arrived at se\'eral cells which exhibited these properties to
a marked extent. It was found that an aluminium electrode was the essential thing in every
cell, together with some acid salt capable of forming an oxide with aluminium. The other
electrode might be any conductor unaffected by the solution. Some of the conductors suitable
for the other electrode are iron, carbon, and lead, and the following solutions all gave more or
less satisfactory results. Acid sodium carbonate, acid sodium phosphate, acid potassium
tartrate, potassium alum, and in fact most of the ionizable, slightly acid sulphates, carbonates,
tartrates, and phosphates. By merely putting one of these cells in the circuit, the one
half of the alternating current wave may be choked out. But this method gives an efficiency
of less than 50%. Thus the authors were led to devise the alternate-path method. Before
describing this method, however, we will take up in detail the properties of a single cell. After
a current is passed for a few minutes through one of the cells, a coating of oxide is formed on
the aluminium electrode which is practically a non-conductor. While this does not prevent the
difference of potential from being maintained across the cell, it does prevent the ions from
giving up their charge and in this way it acts like a polarized copper plate in a single galvanic
cell. This condition of enormous resistance exists when the aluminium is the anode. When on
the reversal of the current the aluminium becomes the cathode there is
merely the resistance of the electrolyte encountered. Any cell possessing this property is
called asymmetric.

As stated before, a single cell by being merely put in the circuit would choke out the one half
of the alternating wave, but as this gives an intermittent current, the following is the method
devised by the authors. Three cells are needed in all. Two of these consist of one electrode of
aluminium and one of iron, with a solution of sodium acid carbonate.
The third has two aluminium plates and one iron plate between them. The same solution is
used. On looking at Fig. 30 it will be seen that when E is positive the current can flow from
either plate 2 or 3 across the electrolyte to plates I or 4. The path from 3 to 4 is of enormous
resistance, as the aluminium is the anode, but the path from I to 2 is of low resistance and
hence the current takes this path. AA'hen H becomes positive the current can flow from 6 to 7
or from s to 4. It takes the path from 6 to 7 as this is of low resistance. In this way both waves
of the alternating current are used and the only loss is due to the resistance of the electrolyte.
Chapter 6: THE CONSTRUCTION OF THE BOXES

The next thing to consider in the building of a Tcsla coil is the boxes which contain the
transformer and high-tension coil. One box for transformer, condenser, and high-tension coil
might be used, but for a coil of this size the weight would be objectionable. Two separate
boxes give the ideal result. They ha\'e the advantage of not being too bulky to handle, and the
transformer in this form can be used separately, if so desired. A single box, however, has the
advantage of taking up less room and of having all the high-potential connections inside
where they are safe, except those which lead to the interrupter which is placed on top of the
box. Oak makes the most substantial box, but it is harder to make tight, owing to the fact that
the shellac \'arnish which is used for filling up the pores in the wood does not sink into oak
with the same readiness as it does in a softer wood. Pine is the best material to use as the
joints will require considerable filling up to make them imper\-ious to paraffine oil, which
will soak through almost anything in time. Sugar pine may be readily stained and looks very
neat when \-arnished. For the sides and ends of the transformer box it requires a piece of
straight-grained pine free from knots, ij" x lo", 6" long. The bottom should be made of a piece
of i J" x 14", 26" long, and the top of a piece of i"x 11", 2' long. Cut the pieces to the size
shown in Fig. 31, and plane the edges true. The end pieces must be mortised into the sides i"
from the end. These tongues and grooves may be cut with a saw and chisel if a rabbeting
plane is not at hand. After the sides

and ends are finished the tongues and grooves are given a heavy coat of shellac, which has
been dissolved in grain alcohol, and while still wet are put together. Six long brass screws
with round heads are to be used in each board to hold the sides. A brass washer Y l^i diameter
should be placed on the screw to prevent the head from sinking into the wood. Next the edges
should be gone over with a plane if necessary so that the bottom board will fit flush in all
places. The bottom board is to be i" wider than the width of the box, so that it laps over V' on
each side. \\'\\tn the bottom board is cut to size, the edges are rounded off with a plane to gi\-e
a finish, and then it is fastened to the box with long ilatheaded brass screws, placed every
four inches along the sides and ends. A coating of shellac is given to the edges of the box just
before putting the bottom on, to help make it tight. The screws must be forced in until they
are flush with the wood. Next the inside should be given five or six coats of shellac,
paying especial care to get it into the joints, and allowing each coat to dry before applying the
next. A small brass cock in the end near the bottom is a convenience in emptying the box of
its oil, but the labor of putting it in so that the box will not leak is such that a siphon is
quick enough for an occasional emptying of the oil. The box for the high-frequency coil and
condenser must have the same care taken in its construction as in the case of the transformer
box. The dimensions are given in the working drawings in Fig. 32. A partition is put in
between the condenser and oscillation transformer, but several holes should be bored in it
near the bottom to allow of the free circulation of the oil. This box must also have several
coats of shellac, as the insulating oil used will leak through in spite of all the precaution
taken.
After the boxes are finished they should be stained or varnished to suit the taste of the builder.
Walnut stain looks well, and as it is dark it covers up a multitude of faults The Construction
of the Boxes 63 in the wood working. If the boxes are well made a good ciling followed by
several coats of shellac makes a very good finish.
Chapter 7: ASSEMBLING

It is not wise to hurry when assembling the apparatus, for if the high-tension wires are not
properly insulated, brush-discharge effects will be noticed on operating. In nine cases out of
ten poor insulation will result in puncturing his condenser and probably burning out his
transformer. Care should be taken to follow these directions. First mount the transformer in
its box. After lowering the transformer into its box bring its four primary leads to four heavy
binding-posts on the end of the box. The two inner terminals of the two sections are brought
to two adjacent binding-posts, and the two outer ones to the other two, in such a manner that
connecting electrically the two middle binding-posts puts the sections in series and
shortcircuiting the two outer pairs throws the sections in parallel. See the diagram. This is
accomplished by means of a piece of fiat brass with slots filed in it so that it just fits across
two binding-posts, or by a short piece of brass rod which fits in the holes of the binding-
posts. The secondary terminals are soldered directly to two brass rods f" in diameter and 3"
long, which extend through the opposite end of the box for ij". These rods are insulated by
two heavy hard rubber or fibre bushings made as follows: A 2" piece of hard rubber or fibre
rod 2" in diameter is turned down to iV', except for a i^" flange on one end the full diameter
of the rod. A §" hole is drilled down the centre of the bushing. The bushinsrs should be a
driving fit both in the end of the box and over the brass rod. The rods are 6" apart and 8" up
from the bottom of the box. This brings them well above the level of the oil, thus assuring no

leakage at this point. The leads from the sections should be led to the brass rods through
glass tubing bent in the desired shape. A hole is drilled and tapped in the end of each rod to fit
a standardbrass set screw. Another hole is drilled at right angles to the first about \" from the
end of the rod to meet the first hole. This makes an efficient binding-post to hold the
conductor. This finishes the connections on the transformer, which can now be placed in the
position which it is to occupy. The box is filled with enough pure paraffine oil to cover the
transformer. This oil should be of the best quality obtainable, free from moisture and
impurities, such as is used for insulating purposes. It should be allowed to soak into all the
sections for 24 hours before using. The primary terminals are connected to a source of
alternating current by means of a suitable switch and fuse capable of carrying 30 amperes.
The next step is putting in the connections and terminals in the high-tension box.

Three brass rods similar to those used in the transformer box, with the same form of
bushings, are driven through holes in the end of the high-tension box next to one end of the
oscillation transformer. These holes should be about 6" from the bottom and 3" apart. The
brass rods project I" williin the box. The bushings and rods can be made oil tight by giving
them a good coating of Le Page's glue before driving them into place.
A 2" strip of wood is glued on in the lower inside end of the box below these rods, to prevent
the end of the oscillation transformer from coming in contact with them.

.
The condenser is first lowered into place in the box and a wire is run from one terminal to the
nearest outer brass rod, to which it is soldered. The wire should be enclosed in a glass tube
suitably bent and should follow the lower edge of the box. A wire is run from the other
terminal of the condenser in a similar manner to the other outer brass rod. It follows the other
lower edge of the box. A tap wire is led from this conductor at a point opposite to where one
terminal of the primary band of the oscillation transformer is to be. A wire is also soldered to
the middle brass rod. The oscillation transformer is now lowered into place. Tlie tap wire
from the condenser lead is soldered to one end of the primary band and the wire from the
middle brass rod to the other end of the band. These leads should be run in glass tubes an as
directly as possible. They should also be kept under the oil. The connections are shown in fig.

After all connections have been securely soldered the box is filled with oil, so that the entire
apparatus is completely immersed. The oscillators and standards can now be constructed.
Two fibre or hard rubber bushings 2]" in diameter and i.\" in length, and having a flange ]"
thick and 3" in diameter turned on one end, are set in two holes cut in the co\'er directly
above the holes in the brass bushings on the oscillation transformer. A J" hole is drilled
through the centre of each bushing. Two J" brass rods 10" long are enclosed in fibre tubes f"
in outside diameter and gY' lo^ig- The tubes should fit the rods tightly. The ends of the brass
rods project from the fibre and can be slightly tapered to fit the bushings on the oscillation
transformer.
The oscillators consist of two brass balls i" in diameter screwed on the end of two ^f/' brass
rods 12" long, which are to slide easily in two holes drilled \" from the top of the standards
through both the fibre and the rod. A set screw at the top of each standard will be of
convenience in clamping the rods in any desired position. In order that the discharge gap may
be adjusted while the coil is in operation two vulcanite handles f" in diameter are screwed on
the ends of the rods, carrying the oscillators, for about li". Slide the standards through the
bushings in the cover until the rods make good contact with the bushings on the oscillation
transformer. This completes the connections in the second box. It should now be placed in its
final position at the high-tension end of the transformer, leaving a space of about i^' for the
interrupter between the two boxes.
The particular form of interrupter which has already been built is connected to the binding-
posts of the high-tension coil as in the wiring diagram. These leads and those from the
transformer to the high-tension box should be of No. 12 B. & S. gauge hard copper wire and
enclosed in the heaviest glass tubing obtainable. They should be as straight andas short as is
consistent with safety in operating the primary spark-gap. A suitable panel switchboard, with
the necessary fuses, transformer and interrupter sv/itches, makes a desirable acquisition to the
apparatus, but it is not essential. This matter is left to the taste of the individual worker.
When the connections have all been made and the oil has dri\en out all the air that it can,
open the interrupter gap about i" and cautiously close the transformer switch. If no excessi\'e
load is taken by the transformer as manifested by the 30 ampere fuses or an alternating-
current ammeter in series, if one is obtainable, the spark-gap can be slowly closed until the
condenser discharges across it. Then, if the directions have been carefully followed in
building the apparatus, a heavy, bluish white, snapping discharge of over 12" in length will
pass between the oscillators, upon the further adjustment of the interrupter gap.
CHAPTER 8: THE THEORY OF THE TESLA COIL

Although, in the introduction, the authors stated that they would not attempt to give a
mathematical explanation of the coil, still they feel that a few facts regarding the theory
would not be out of place here, in that it may suggest certain improvements to the reader. It
will also be of assistance if the amateur wishes to construct a coil of his own design. The first
thing to consider is the transformer. Its action, as is well known to almost everybody, depends
on electromagnetic induction. The alternating current flowing in the primary sets up an
alternating magnetic field, which being linked with the secondary induces an electromotive
force in it. When the secondary is open there is theoretically no current passing through the
primary, due to its high selfinduction, except that necessary to magnetize the core. As a load
is thrown on to the secondary, the current through the primary automatically adjusts itself as
the self-induction is decreased by the opposing ampere turns of the secondary, that is, if the
transformer is self-regulating for varying loads. The normal current through the primary of
the transformer used in the 12" coil is from 22 to 25 amperes, the secondary voltage being
about seventy-five hundred, and thus the amperage in the secondary is about -3 oTTo get the
required voltage of seventy-five hundred in the secondary on fifty-five volts in the primary it
is necessary to connect the two sections of the primary in parallel, as this has the effect of
cutting the .primary turns in two. On one hundred and ten volts the sections are connected in
series. The use of the transformer in the Tesla apparatus is merely to charge a condenser, and
thus it is seen that an ordinary induction coil or even a static machine of the proper
dimensions could be used, but they are not nearly as handy. Another important matter in
connection with the construction of a transformer to be used for creating electric oscillations
is to secure a sufficiently small resistance in the secondary. The reason for this is that the
transformer is used to charge a condenser.
When an electromotive force is applied to the terminals of a condenser, the full difference of
potential is not created between the terminals of the condenser immediately, but rises
gradually. The time required to charge the condenser depends on its capacity (C) and the
resistance {R) of the charging circuit. The product CR is called the time constant of the
condenser, and practically the condenser is charged in a time equal to ten times the time
constant. The time constant is to be reckoned as the product of the capacity (Q in microfarads
and the resistance {R) of the charging circuit in megohms.- The time is gi\en in fractions of a
second.
The condenser, more than anything else, constitutes theessential part of the Tesla coil. It plays
the same part as the mechanical interrupter in the ordinary induction coil. Its action, however,
is purely electrical and its great advantage lies in setting up the currents of enormous
frequency. When any condenser is discharged, the discharge may take one of several forms,
depending only on the three electrical constants of the discharging circuit — inductance,
capacity, and resistance. The discharge may be either oscillatory or entirely unidirectional,
consisting only of a gradual equalization of the potentials on the two plates. This may be
made clear by the following mechanical illustration. Suppose a glass U-tube to be partly fdled
with mercury, and the mercury to be displaced so that the level in one side of the tube is
higher than in the other. There is then a force due to the difference of level, tending to cause
the liquid to return to an equal height in both limbs. If the mercury is now allowed to return,
but is constrained, so that it is released slowly, it goes back to its original
position without oscillations. If, however, the constraint is suddenly removed, then owing to
the inertia of the mercury it overshoots the position of equilibrium and oscillations are
created. If the tube is rough in the interior, or the liquid viscous, these oscillations will
quickly subside, being damped out by friction. What we call inertia in material substances
corresponds with the inductance of an electric circuit and the frictional resistance experienced
by a liquid moving in the tube, with the electrical resistance of a circuit. If we suppose the U-
tube to include air above tiie mercury and to be closed up at its ends, the compressibility of
the enclosed air would correspond to the electrical capacity in a circuit. The necessary
conditions for the creation of mechanical oscillations in a material system or substance are
that there must be a self-recovering displaceability of some kind, and the matter displaced
must possess inertia; in other words, the thing moved must tend to go back to its original
position when the restraining force is removed, and must overshoot the position of
equilibrium in so doing. Frictional resistance causes decay in the amplitude of the oscillations
by dissipating their energy as heat.
In the same way the conditions for establishing electrical oscillations in a circuit is that it
must connect two bodies having electrical capacity with respect to each other, such as the
plates of a condenser, and the circuit itself must possess inductance and low resistance. Under
these conditions, the sudden release of the electric strain results in the production of an
oscillatory electric current in the circuit, provided the resistance of the circuit is less than a
certain critical value. We have these conditions present when the two coatings of a Leyden jar
are connected by a heavy copper wire.
The main thing in constructing condensers to be used on high charging voltages is the solid
dielectric. There are in all only a few dielectrics suitable for high-tension work, and this
number is reduced when cost and internal energy loss in the dielectric are considered. Glass
of certain compositions, ebonite, mica, and micanite are practically all that are suitable, and
of these flint glass is the best, as its dielectric constant is high, being from 5 to 10, and its
dielectric strength is also great. Glass is brittle, however, and liable to have
flaws which sooner or later give way. The capacity of a condenser depends on the area of the
plates, their distance apart and the constant of the dielectric used, and is expressed by the
following formula in microfarads, where K is the dielectric constant, 5 the total area of the
plates expressed in square centimeters, and D the distance apart in centimeters,
The oscillation transformer is nothing but a modified transformer with an air core. The only
important facts about its construction are that it should be built to withstand great voltage
differences between the turns, and that the primary should have as small an inductance as is
practicable, in order to make the frequency as great as possible. No advantage is gained by
having many close turns in the primary, because the increase of inductive effect on the
secondary, due to an increase in the number of primary turns, is about exactly
annulled by the decreased current through the primary due to its own greater inductance.
The function of the interrupter is to destroy any arc that may be formed across the terminals
of the primary sparkgap, for if this arc is not wiped out there will be no true oscillatory
discharge in the condenser circuit or only a feeble one. The reason for this is that as long as
the arc discharge continues, the secondary terminals of the transformer are reduced to nearly
the same potential, or at most differ only by a few hundred volts.
The function of the primary spark-gap is to regulate the voltage to which to charge the
condenser. Since the potential difference between the spark balls is almost equal to the
potential difference across the condenser, the condenser will discharge at a voltage
determined by the length of the airgap. Now there is a certain length of spark-gap which is
best suited for each coil and it can easily be determined by trial. As a rule it is best to start
with a rather short spark-gap, gradually lengthening it out until a point is ahnost reached,
when opening it out any further would cause it to cease passing. This spark length almost
always gives the best results. In the earlier part of this chapter it was stated that the
highfrequency resistance and inductance should be substituted for the ordinary resistance and
inductance, when dealing with circuits which are subject to the action of electric oscillations.
The processes and means used for the measurement of low-frequency alternating currents and
potentials are not always applicable or correct either when applied to high-frequency
measurements. The main reason for the difference between the two cases is to be found in the
fact that a high-frequency current does not penetrate into the interior of a thick solid
conductor of good conductivity, but is merely a surface or skin effect.
When tra^'ersed by an alternating current, there are five qualities of a circuit to be considered.

1. The resistance of the conductor, which is always greater for high-frequency currents than
for the ordinary currents; that is, direct currents and alternating currents up to about a
frequency of loo per second.
2. The inductance of the conductor depends on its geometrical form, material, and the nature
of the surrounding insulator. The greater the frequency, the smaller the inductance becomes.
3. The capacity of the conductor, depending on its position with regard to the return circuit
and other circuits and on the dielectric constant of the surrounding insulator.
4. The dielectric conductance of the insulator surrounding the conductor.
5. The energy dissipating power, due to other causes than conductance, such as dielectric
hysteresis, which exist in the dielectric. Under this heading comes the loss of energy from the
brush discharges through the air between the conductors.

If the constants of a circuit for low-frequency currents are known, the values of the constants
for high frequencies can be calculated fairly correct. The high-frequency constants can,
however, be measured directly, but the apparatus is rather delicate and inconvenient for use
and besides not always satisfactory. If the coil builder cares to measure the constants of a
circuit for himself, he will find the description of the necessary instruments in other books as
it is beyond the scope of this work.
In these experiments an assistant should adjust the sparkgap so that no more current passes
than is necessary. This is to prevent the spark from burning the operator. To convince the
audience of the tremendous voltage passing through the operator's body he has merely to
bring one hand up to a lighted incandescent globe, while he grasps one terminal of the coil
with the other hand. On the near approach of the hand the filament will violently vibrate and
then shatter, blackening the bulb and of course extinguishing the lamp. Lighting Geissler
tubes held in the hand or even in the mouth by approaching them to the oscillators is an
experiment that never fails to bring forth the applause of those present. Perhaps the most
spectacular experiment, one which is unaccountable for by the every-day electrician who
does house wiring and has never been brought in touch with high-frequency- currents is the
lighting of an ordinary incandescent lamp with the current traversing the operator's body.
Before performing this experiment some few remarks on the quantity of current necessary to
bring the filament to full brightness on the no-volt circuits should be made, if those present
are ignorant of electrical matters. They can thus see that the energy required to light an
ordinary 16 C. P. lamp is equivalent to 55 watts, and that this amount
is therefore taken through the operator's body. At no \olts this means approximately \ ampere,
while according to the best authorities ^^ of an ampere is fatal to the average
human. The reason why this amount of energy can be taken with impunity is not definitely
known, but it is thought to be due to the fact that the high-frequency current does not
penetrate into the interior of a solid conductor, but follows the surface. This is known as the
skin effect. The operator, to get the best results, should stand on an insulated stool and grasp
one terminal of the coil with one hand, and approach with a piece of metal held tightly in the
hand or mouth one lead of a lamp, the other lead of which has been previously grounded. The
lamp will come up to bright red and if an assistant adjusts the primary sparkgap to its best
working distance, the lamp may be brought up to full brightness.
Chapter 9: DIMENSIONS OF 7" STANDARD COIL

For those amateurs who, ha\'ing read the previous chapters, think that an apparatus giving a
twelve-inch spark is too large for their limited uses, this chapter has been added. This coil is
by no means to be thought of as a toy, for the authors themselves used the very apparatus
described in this chapter in carrying out their first experiments with the X-ray and Geissler
tubes. Wireless messages were also sent successfully over a distance of three miles in wet
weather. This was the greatest available distance over which the authors could try the coil, so
that three miles should not be considered the maximum transmitting distance. In clear
weather messages could easily be sent a distance of about fifty miles, provided your antennae
is well insulated from grounds.
Because this apparatus is not as powerful as the other, does not mean that any less care
should be taken as regards insulation and mechanical construction, for it depends entirely on
this whether the coil builder is to get a thin, intermittent spark or a fat crackling one. The only
difference between this apparatus and the larger one, besides that ofsize, is in the construction
of the transformer and condenser and then they are only trivial. The core of the transformer is
2f" in diameter and is built up of pieces of No. 32 B. & S. gauge iron wire 13" long after the
manner described in Chapter II. The same care should be taken in annealing and insulating
the iron wires as was done before. The primary is wound in two sections adjacent to each
other, as seen in Fig. 38. Each section is wound towards the centre, starting i" from the ends
of the core, for a distance of 5". There are six layers of No. 16 B. & S. gauge double cotton

covered copper wire in each section. Each layer is thoroughly shellacked when put on and the
terminal wires are held by the same method as described in Chapter II. At least two feet
should be left for bringing out the terminals to the binding-posts. The secondary is wound in
two sections. No. 32 B. & S. gauge double cotton covered copper wire is used. The tube
Dimensions of 7" Standard Coil 99 on which it is wound has an internal diameter of 4" and
the thickness of the wall is ^^". It is 11" long and made from the best vulcanized fibre. The
bobbin heads are cut out of j" sheet fibre. They are 6" in diameter and have a hole 4g" in
diameter cut out of the centre.
On the one side of the frame two strips J" x i" are fastened; one at the top and the other at the
bottom of the frame. See figure. The frame is now laid on a flat surface with the side on
which the strips have just been fastened down. A glass plate is placed in the frame after
having been wiped dry and clean so that it touches the bottom of the frame. Then a brass
sheet is laid in so that there is i" margin of glass at the bottom and a V' margin on the sides.
The lip should just fit against the upper edge of the glass. Without displacing the brass sheet
place a sheet of glass on top of it. This is followed with a sheet of brass, but in this case
the lug on the brass is brought out on the opposite side to the previous one. Continue this
process until the 28 sheets of glass have been put in place. Two glass sheets are placed on top
of the last brass sheet. Also remember to bring the lugs from the brass sheets out on alternate
sides. Mortices should be cut in the top and bottom of the upper ends of the sides a little
deeper than the point to which the last glass sheet reaches. These are to receive two strips of
pine l" X i" similar to those on tjie other side. These should be screwed down so that they
press firmly against the glass. A piece of paper or cloth placed between the strip and the glass
will prevent the breaking of the latter. Set the condenser upright and solder a piece of copper
wire, which has already been tinned, to each of the lips in turn down the one side and another
wire to all the lips on the other side. About No. 16 bare wire will do. Enough extra wire
should be left to make all necessary connections. The oscillation transformer is constructed in
the same manner as the one for the 12" coil. The circular supports for the secondary are 6" in
diameter and are turned out of i" material. Eight equidistant slots are cut in the
perDimensions of f' Standard Coil 103 iphery Y square. The fibre strips are A" square and 11"
long. A rod 11" long is turned out to the size shown in the figure 41. It is i" in diameter and
has a A" shoulder turned on each end. This rod holds the two supports for the secondary
apart. If the method of winding the wire in grooves is to be used, the thread should be cut on

The fibre strips before mounting them on the supports. In the original coil the wire was
merely wound on three fibre supports 3" wide mounted on a hexagonal end piece. The wire
was wound so that no adjacent turns were in contact and the whole was thoroughly
shellacked. Although this method of winding gave good results while the coil was new, it was
found after some usage that the wires became loosened, thereby reducing the effecti\-e
sparking distance.
Any one of the forms of interrupters described in Chapter V can be used with the coil; the
coil in question being equipped with the motor interrupter. The connections between the
primary spark-gap and hinges can be run in glass tubes lying in grooves cut in the under side
of the cover. A piece of I" hard rubber sheet should be screwed over the grooves wherever
there is any danger of shorting to the core of the transformer or primary terminals. The
connections are shown in the wiring diagram. The oscillators consist of two brass balls J" in
diameter screwed on the end of two -^^" brass rods 7" long, which are to slide easily in two
holes drilled \" from the top of the standards, through both the fibre and the rod. A set screw
at the top of each standard will be of convenience in clamping the rods in any desired

position. The standards are constructed as follows. Two fibre or hard rubber bushings 2" in
diameter and iV' in length and having a flange \" thick and 2^" in diameter turned on one end
are set in two holes cut in the cover directly abo^•e the holes in the brass bushings on the
oscillation transformer. A I" hole is drilled through the centre of each bushing. Two f" brass
rods 8" long are enclosed in fibre tubes f" in outside diameter and 7 V' long. The tubes should
fit the rods tightly. The ends of the brass rods project from the fibre and should be slightly
tapered to fit the bushings on the oscillation transformer. In order that the discharge gap may
be adjusted while the coil is in operation, two ^'ulcanite handles I" in diameter are screwed on
the ends of the rods, carrying the oscillators, for about i\". The standards are now slid through
the bushings in the cover until they make good contact with the bushings on the oscillation
transformer. When the coil is now connected up to the alternating current mains, it will break
forth in a beautiful 7" discharge. If everything is not as it ought to be, the trouble may be
found in the manner described in Chapter VII.