Construction of a secondary induction coil

This is an article that appeared in Experimental Electricity for Boys by Willard Doan, published in 1959. It is a fairly simple project, however some of the materials may no longer be readily available, and the next best substitution will need to be found. WARNING! This project involves high voltage electricity, and is therefore very dangerous, so use extreme caution. This is part of a series of three pages, which also includes companion articles on building a Tesla coil and a leyden jar. All text and illustrations in this article are by Willard Doan. Links to the illustrations are listed at the end of this page. I am not expert in building these things, so I probably won't be much help answering your questions. This article is here primarily for historical reasons. Sorry it isn't in metric, but this is an old book from the U.S., after all!
The older and more experienced experimenter may be interested in constructing a high voltage induction or spark coil of the vibrator type. This may be used for many interesting experiments, to say nothing of the valuable experience gained in its construction. The coil described will produce a jump-spark 1 1/2 to 2 inches long when excited by an 8- or 10-volt toy transformer or by a G-volt storage battery. We do not recommend that it be operated from dry cells as it draws too much current for this type of voltage source. Study the construction drawings of each part as you go along, being sure you understand each one before beginning the construction. See Figs. 7-2 through 7-8 for construction details.

Before we start the constructions let us consider just how such a coil works we have learned that a steady magnetic field from one current-carrying coil will not induce a current in another coil. To get an induced current or voltage, the magnetic field must be changing. The vibrator, or interrupter as the complete vibrator assembly is sometimes called, serves to make and break the circuit through the primary coil rapidly, which causes a similar interruption and rebuilding of the magnetic field through the coil. This changing magnetic field is what causes the current to be induced in the secondary coil.

The vibrator is a flat, steel spring mounted at one end of the core of the coil in such a position that when the core is magnetized by a current flowing through the primary coil, one end of the vibrator is attracted to the core. A contact on the vibrator bears against a solidly mounted contact screw when the core is not magnetized. The vibrator and contact screw are in series with the battery and primary coil. Therefore, when current flows from the battery through the contact screw, vibrator, primary coil, and then back to the battery, the core is magnetized and attracts the vibrator spring to it. This pulls the vibrator contact away from the contact screw, thereby opening the circuit. When the circuit is opened, the core loses its magnetism, releasing the vibrator which springs back against the contact screw closing the circuit again. This action continues at high speed as long as the battery is connected. At each make and break of the primary circuit, a current is induced in the secondary coil by the changing magnetic field. The current induced by the breaking of the circuit is the stronger.

The materials required to build the secondary induction or spark coil are as follows:

  • 250 feet of No. 18 soft iron wire
  • 70 feet of No. 16 enameled-copper magnet wire
  • 3 1/2 pounds of No. 34 or No. 35 enameled-copper magnet wire
  • one roll best quality thin waxed paper
  • 3 square feet of "Empire" or varnished cloth
  • 2 square feet of heavy kraft wrapping paper
  • 2 square feet of thin cardboard
  • 8 inches of clock spring, 35 to 1 inch wide
  • one strap of brass, 3 1/4 by 3/4 by 1/16
  • one strap of brass, 2 1/2 by 3/4 by 1/16
  • one piece of fiber board or bakelite, 1/4 by 1/4 by 1/8
  • two automobile breaker-contact screws with nuts to fit, or two tungsten contacts from a breaker and one 8-32 machine screw, brass or steel, 1 inch long, and two nuts to fit
  • one 10-32 steel machine screw 3/4 inch long and nut to fit
  • four 8-32 by 1 1/2 inch brass machine screws
  • eight 8-32 brass nuts
  • assorted small wood screws
  • one or two 0.5-microfarad* 600-volt tubular paper radio capacitors, or 15 square feet of tinfoil or thin, smooth aluminum foil
  • one pound of paraffin wax
  • soft wood for case
  • radio service or model-maker's cement
* The farad, named for Michael Faraday, is the unit which measures the capacity of a capacitor to store electricity. Micro means "one millionth".

The core for the coil is made of No. 18 soft iron wire, though something near this size will do very well. A good type to use is the single strand No. 18 sold by hardware stores as stove pipe wire. Cut enough pieces 8 1/2 inches long to make a tightly packed bundle 3/4 inch in diameter.

After the wires are cut, they should be annealed or softened by enough heat to make them red hot, then cooled slowly. After cooling sandpaper each wire to remove the oxide coat formed while heating. The wires need not be bright and shining, but should have the excess oxide removed. Then file one end of each wire flat and straighten each one individually by tolling it under a board or flat piece of metal on a smooth, hard surface.

Roll up and securely cement a tube of two or three layers of tough paper, 8 3/8 inches long and 3/4 inch inside diameter. Pack this full of the wire, with the flattened faces of the ends together. Even up the surface at this end to make a smooth flat surface to attract the vibrator. Be sure the wires are straight in the paper tube and so tight that not another one can be forced in. See Fig. 7-2 for the core construction details.

Impregnate the paper with paraffin by brushing on hot melted wax while holding the tube in a warm place so the wax can soak in before hardening. The core is now ready for the primary coil.

The primary consists of two layers of No. 16 gage enameled copper wire wound for a length of 7 inches over the core, with one layer of heavy waxed paper between the layers of wire.

Start the primary winding 1 inch from the flat end of the core. To hold the end of the wire at the start, cut a piece of friction tape about 1/4 inch wide and an inch long and loop it around the wire about 6 inches from the end. Then start the winding with the ends of the tape lying parallel with the core. As the wire is wound around the core, the tape holding the end of the wire will be securely anchored by the winding.

About 1/2 inch before the finish of the second layer place a similar piece of tape, so that as the winding finishes, a loop of tape just large enough to put the wire through will be left uncovered. Do not place one loop of tape directly over the other but separate them and fan out the ends. Otherwise, there will be a bump in the primary winding which will interfere with slipping on the secondary. The primary will require nearly 70 feet of wire.

Over the primary coil wind 6 layers of varnished paper or cloth for insulation from the secondary coil. This material is known by the trade name of "Empire,' and may be obtained from a radio supply house. Paper is the less expensive but it must be handled carefully to prevent damage. The smallest crack on its surface will make it worthless for insulation purposes.

We are now ready to start the two secondary coils which are exactly alike and must be wound in the same direction. Make two paper tubes 3 1/4 inches long by rolling up three layers of heavy paper and securely cementing the paper with radio service cement. The diameter of the paper tubes should be such that they will slip over the "Empire" cloth covering of the primary winding (one from either end of tile core) and fit closely. A sure way of making them the right size is to wrap one layer of thick paper around the insulated primary and build the paper tubes tightly around this. Then when the inner layer of paper is removed, the tubes will be the correct size. The secondary coils are to be wound on these tubes.

It will be necessary to make a coil winder as each secondary coil will have thousands of turns of wire. The form of the winder is not important and will probably depend upon the materials available. A suggested form is shown in Fig. 7-3. If a round piece of wood is used as a mandrel inside the paper tube on which the wire is wound, it should be slightly smaller than the tube and the space between should be filled by wrapping paper around the wood. This makes it easy to remove the finished coil, which may be done by pulling out the filling of paper after which the coil will slip easily off the wooden mandrel.

The wire for the secondary should be enameled copper. Number 34 or No. 35 is recommended. This wire may be procured from old loud-speaker fields obtained at some convenient radio shop.

Begin the winding 1/2-inch from one end of the paper tube leaving free several inches of wire for connecting purposes. Wind the wire smooth and tight, without allowing any turn to cross over another. It may take a little time to get the necessary skill. Wind the wire for a length of 2 1/4 inches or 250 turns whichever occurs first. This will, of course, depend on the size of wire used. If 250 turns do not occupy 2 1/4 inches of space, measure the winding space it does cover, and wind subsequent layers this length. Locate the winding midway on the paper tube, that is, leave the same amount of tube exposed at each end of the layer of wire. It is not necessary to count the turns after the first layer, as approximately the same number of turns per layer is satisfactory. Too many turns per layer may cause the insulation between layers to break down.

After the layer is completed, wind on two strips of thin waxed paper, 3 1/4 inches wide. The grade of waxed paper sold in rolls by grocery and "five and dime" stores if free of holes is satisfactory. Be careful not to damage the paper by bending it sharply, as this may weaken its insulating quality. It will handle much better, with less danger of damage, in a warm room. This paper must be put on as tightly and smoothly as possible. Wrap two sheets together rather than wind on one long sheet for two complete turns. The beginning and end of the sheets should overlap by about 1/4 inch.

It will help winding if two or three rubber bands of different sizes are put around the mandrel of the winder at each end, before the winding is started. Then when a layer of paper is put on, or it is necessary to stop winding for a time, a rubber band of the proper size will hold the work securely.

The winding should continue for approximately 75 layers if possible with double strips of waxed paper wound between each layer of wires. However, the outside diameter of the completed coil must not be more than 2 3/4 inches. If a heavier-size wire than that recommended is used, it may not be possible to wind 75 layers. After the last layer of wire has been wound, 8 or 10 layers of waxed paper should be put on and then fastened by wrapping them securely with thread.

Two coils exactly alike should be made. Be sure each coil is wound in the same direction. In case the wire should break while winding, clean the ends carefully for a distance of 1/2 inch and twist them securely together. If the splice can be soldered without leaving the wires rough, it may be better to do so, but this is not really necessary. Be sure the ends of the wire lie flat so there is no danger of the paper insulation being damaged by the sharp points. Fold a narrow doubled strip of waxed paper around the splice and continue the winding. For the next two layers, skip a space in the winding over the splice to lessen the pressure on it.

Be extremely careful to prevent breaks in the wire. This is a reason for not using very small sized wire. On the other hand, if the size of the wire is too large not enough turns per layer can be put on. Also, not enough layers can be wound without making the outside diameter of the coil so large that it will lie outside the strong magnetic field of the core.

After the coils are completed, they should be completely immersed in hot melted paraffin wax and soaked for an hour or more. Cool the vessel slowly until the wax begins to solidify and then remove the coils. Cool them as quickly as possible to prevent the wax from draining out. When cool, the ends of the coil should be completely filled with wax where it may have drained out. This may be done by standing the coil on one end and pouring melted wax slowly into the other end until full. Work out all the air bubbles taking care not to damage the paper insulation. Allow the wax to cool, and treat the other end of the coil the same way.

Cut two disks of cardboard 3 inches in diameter with a hole in the center just large enough to fit over the primary coil with its insulation. Cut 6 "Empire" cloth or paper disks like the cardboard. The disks are to serve as insulation between the two secondary windings.

We are now ready to assemble the coils. Slip one secondary coil over the "Empire" cloth-covered primary. The beginning end of the coil should be approximately over the center of the primary. Slip one of the cardboard disks over the primary winding and butt it up against the secondary coil. Next slip on the "Empire" cloth disks followed by the other cardboard disk.

Clip the projecting end of the coil connection to about 2 inches in length and clean carefully for a distance of an inch. Prepare the corresponding connection of the other coil the same way. Slip this coil over the primary far enough to fasten the connections together carefully. Bend the connection down against the primary and tie with a thread to make it stay down ( Fig. 7-2). Push the coils tightly together with the insulating disks in between them. By connecting the beginning terminals of the secondaries together next to the primary, the high voltage terminals will be more widely separated from each other and from the primary, thereby lessening the danger of insulation breakdown.

Details of the interrupter, consisting of the vibrator and contact screw assembly, are shown in Fig. 7-4.

The vibrator is made of two pieces of clock spring carefully straightened and just alike except that only one has a contact. To make a hole in the spring: with a sharp punch, make a little dent in the spring, being careful not to split the steel. This will raise a little bump on the other side. Grind it off with an emory or whet stone. Punch and grind until a hole is made through the steel. After the hole is started, it may be enlarged with a smalll drill or round file until it is the proper size.

The contact bridge and vibrator holder are made of brass strap of the shape and dimensions given in Fig. 7-4. A nut to fit the contact screw is soldered on the contact bridge if a tap is not at hand to thread the hole for it.

The best thing for the contact screw is a breaker-contact screw like those used in some of the older-model cars. If it is impossible to find one, a tungsten contact from the present-type breaker may be used. Cut it off of its assembly with a file, leaving a little of the surrounding metal. Solder it on the end of a headless 8-32 machine screw taking the greatest care to get the face of the contact at right angles to the length of the screw. To get the solder to stick to the steel, it will be necessary to use either plain solder with a soldering paste or acid-core solder.

Whenever acid solder or a soldering paste is used, the parts soldered should be very carefully cleaned afterward or corrosion may result. Also, the soldering iron used should be cleaned before it is used again to solder any fine wire or delicate parts. For soldering fine wire never use anything but rosin or some rosin compound for a flux. To be safe, never use paste where a rosin compound will make the solder stick and never used acid or salts except where nothing else will possibly do, and then thoroughly clean the parts afterward removing every trace of the chemicals.

If it has been possible to find some contact screws, one may be used also for the vibrator spring contact. Reduce the diameter slightly with a file as shown in Fig. 7-4, beginning about 1/32 inch back from the tungsten contact. Cut off the contact with the shank attached, forming a small rivet with the tungsten contact for a head. Very carefully peen this into the hole in the vibrator at the point indicated.

If it has not been possible to obtain contact screws, solder a contact (like the one prepared before for the screw) onto the vibrator at the point indicated. It is necessary to use tungsten contacts because any other kind will burn up from the heat produced by the interrupted current.

The adjustment-screw holder is made of insulating material, preferably fiber. If a 10-32 tap is not available, the adjustment-screw hole may be drilled slightly smaller than the screws and the screw forced to cut its own threads. This screw must be steel; brass will not work properly. Under the influence of the magnetism from the corers the vibrator will stick to the screw, causing a quicker break when it releases. The front and side views of the complete interrupter are shown in Fig. 7-4 and Fig. 7-7.

The two vibrator springs are assembled together, the one without the contact next to the core to act as the armature. Having the contact on a separate spring also helps to make the break more rapid.

The dimensions of the case materials are given in Fig. 7-5.

The interrupter capacitor may be either purchased or made. If purchased, a 1/2- to l-microfarad, 600-volt, paper-insulated capacitor, such as used in radios, is the proper kind. If the condenser is made, refer to Fig. 7-6, which shows the essential parts and their assembly. The paper may be the regular waxed paper used in the kitchen for wrapping foods. It must be of the best grade and should be free of holes or thin spots which would certainly cause the condenser to break down under the comparatively high voltage it must withstand. The foil may be aluminum foil such as used in the kitchen if it is perfectly smooth. The kind with an embossed surface is unsuitable. Thinner tinfoil is better if it can be obtained.

To assemble the capacitor, or condenser, first dip each sheet of waxed paper into hot melted paraffin and drain off the excess wax. Be careful in handling the paper not to crack it. The assembly is much easier if done near a stove so the waxed sheets will stick when pressed together. A sheet of Cardboard the size of the paper should first be placed flat with a sheet of paper on it. Follow this with a sheet of tinfoil. Approximately 150 sheets of tinfoil will he needed, with a sheet of waxed paper placed between every sheet of foil. Be sure the tinfoil sheets project alternately from each end of the assembly as shown in Fig. 7-6. After all the foil and paper have been stacked, place another sheet of Cardboard on top for protection. Put the assembly in a warm place and press it under a heavy Hat-surfaced weight until it becomes a solid unit. Wrap the unit with a few turns of string to hold it together securely. Figure 7-7 shows the location of the capacitor in the induction coil assembly.

The connections are made as shown in Fig. 7-8. Wire terminals are attached to the foil leaves at each end of the capacitor. One goes to the contact bridge and the other to the vibrator holder. In other words, it is connected across the interrupter contacts. The purpose of the capacitor is to help stop the flow of current the instant the contacts are separated, causing a quicker collapse of the magnetic field. It also reduces sparking at the contacts, which burns them and makes them stick.

When mounting the vibrator on the end of the case, counter sink the screw heads on the inside of the case as deeply as possible and cover over with paraffin, filling the recess completely. After this has been done cut several sheets of "Empire" cloth the size of the case end with a hole just large enough for the core end to pass through. The purpose of the sheets is to reduce to a minimum the chance of the high voltage from the secondary jumping into the primary through the interrupter connections.

Arrange all internal connections with the greatest separation possible between all high and low-voltage parts. Cut several sheets of cardboard and soak them thoroughly in melted paraffin to separate the secondaries from the condenser. Or better, use a sheet of bakelite or glass l/l6 inch or more thick. After everything has been assembled in the case and connections made to outside terminals, pour hot melted paraffin in until the case is full, taking the utmost care to see that it gets into every cranny and crack.

When the vibrator has been properly adjusted by the two adjusting screws, this coil will produce a fairly steady spark 1 inch or more long when operated by a 10- to 12-volt alternating current toy transformer or a 6-volt storage battery. It is not recommended that dry cells be used for a power supply as they will be quickly run down. From the standpoint of convenience and low cost a transformer having an output of 8 to 15 volts and not less than 4 amperes is the best source of power supply. A toy train transformer of this rating will do nicely, or an 1l-volt radio-tube filament transformer with a center tap furnishes both 5 1/2 and 11 volts adapting it other uses. If a transformer is used for this or any other purpose, great care must be taken not to short or overload it.

Do not allow the secondary to become shorted while in operation as this causes a heavy current to flow which may result in damage if continued for any length of time. Bringing the secondary terminals very close together may also overload the coil. The greatest care should be taken not to allow the spark to jump to the body. It would be extremely unpleasant and some cases might he dangerous. Take care to keep all inflammable materials away from the spark when in operation. A 3/4-inch spark will quickly set fire to paper.


An interesting experiment demonstrating the heat of the spark may be performed. Arrange two vertical wires connected to the secondary so they are about 3/4 inch apart at the bottom and gradually spread to 1 inch at a height of 3 or 4 inches. The spark will start at the bottom and, as heat makes the air above the spark more conductive, it will rise to the top where it goes out. It then will start at the bottom again and repeat the performance.

If two pointed wires connected to the secondary are separated slightly more than the maximum sparking distance, a brush discharge or corona will take place, producing a crackling sound and considerable quantities of ozone gas. Do not breathe much of this gas as it is slightly poisonous,

Many interesting experiments which will be explained later, may be performed with this coil by using it to excite a Tesla coil.

The images.

Fig. 7-2 Fig. 7-3 Fig. 7-4 Fig. 7-5 Figs. 7-6 and 7-7 Fig. 7-8

© 1997- Brian Carusella All rights reserved.
Quotes and images not my own remain in the
copyright of the originator or else in the public domain.