This eight-part series gives directions for constructing a tuned spark transmitter plus a crystal receiver. This article doesn't mention anything about licencing, apparently under the assumption that the transmitter's range was limited enough not to require it. One feature omitted from this review, which likely would have come in handy, was some sort of setup to disconnect the outside aerial during lighning storms.
 
Popular Science Monthly, September, 1916, pages 449-452:

How  to  Become  a  Wireless  Operator

I.--Why  Wireless  is  Interesting

By  T.  M.  Lewis
NOBODY knows just how many amateur wireless operators and experimenters there are in the United States; the total number has been estimated as somewhere between twenty thousand and fifty thousand. Nearly ten thousand licenses for amateur stations have been issued by the Department of Commerce. Each one of these licenses is for an amateur station which contains both a transmitter and a receiver. No license is required for stations equipped for receiving only, and it is believed that there are many more of these than of the sending stations.
    Why have so many American boys and young men taken up this subject? What is there about it that interests them, and induces them to spend their time and money in buying, building and using wireless instruments? The answer to these questions is simply that wireless or radio telegraphy represents one of the latest developments of electrical science, and that it offers both amusement and profit to whoever cares to work upon its problems.
    Whether you wish merely to make a pastime of wireless experimenting or desire to study radio telegraphy with the intention of making some part of it your profession, you will find time spent on it well worth your efforts. In the first instance you will be able to receive messages through the ether from stations many miles away, getting press reports of important news items, and the results of races and ball games and so forth, before they are published in local papers. In the second case, you will be able to train yourself as a radio operator or installation engineer, or possibly you will make new inventions or discoveries of commercial value. Either way you will constantly be learning more and more about electricity and its applications, as well as getting a better knowledge of many important physical principles which may be used in almost any kind of work.
    In addition to all this, there lies before you the fascination of sitting at your receiving instruments and listening to wireless messages from stations located all about you. Soon after you begin it is possible to hear from distances of several hundred miles, and after you have gained a thorough knowledge of your instruments and their possibilities it becomes feasible to listen to the tremendously powerful transmitters even so far away as Germany and the Hawaiian Islands.

Elementary  Principles

    This article is the first of a series which will describe a number of really practical and useful instruments for use in radio telegraphy, both for sending and for receiving. The ways to make and use these various pieces of apparatus will be discussed in detail, but it is not proposed to go into the theory of wireless telegraphy at all. By going to your library you will be able to find books and periodicals which describe the principles of ether-waves and their uses in wireless; some of the books you will wish to buy and have in your own workshop for ready reference. Among the most interesting and valuable of these are the following, which are named in the approximate order of their complexity:
    "The Elementary Principles of Wireless Telegraphy," by R. D. Bangay.
    "Experimental Wireless Stations," by P. E. Edelman.
    "Wireless Telegraphy," by A. B. Rolfe-Martin.
    "Textbook on Wireless Telegraphy," by Rupert Stanley.
    "Wireless Telegraphy," by W. H. Marchant.
    "Elementary Manual of Radio Telegraphy," by J. A. Fleming.
    "A Handbook of Wireless Telegraphy," by J. Erskine-Murray.
    "Wireless Telegraphy," by J. Zenneck, translated by A. E. Seelig.
    The above list should be useful as a guide in hunting for technical information about radio telegraphy. There are many other books on the subject, a large number of which are excellent. Those named, however, include one or more of each type from the most elementary to the most advanced.

A  Simple  Transmitter
Fig. 1
    In beginning experiments on wireless telegraphy it is best to take up first the least complicated arrangements, which are suitable for very short distances, and then to work along gradually from these to the more important instruments. This first article, therefore, will describe the use of a complete wireless set which is capable of demonstrating the principles involved. By its use you should be able to send messages a distance of a few hundred feet, from one part of the house to another; by using long aerial or antenna wires, upward of a quarter of a mile may be covered.
    The sending station involves nothing more than a simple buzzer, a telegraph key, a tuning coil and a few cells of dry battery. These are to be connected together as shown in Fig. 1; a good kind of wire to use is No. 18 annunciator, since this has a strong waxed double cotton covering which is easily removed. The buzzer can best be purchased from any electrical supply store for about forty cents; the key may be bought, or simply improvised by cutting and bending some thin strips of brass as shown in Fig. 2; the dry cells will cost from twenty to thirty cents each.
    The tuning coil may easily be built by winding about fifty turns of annunciator wire on a cardboard tube approximately three inches in diameter. The ends may be fastened and at the same time made available for convenient connection by attaching them to binding posts let into the tube at the top and bottom. There is no need of building this tuning coil of any specific size. The diameter may be anything from two to four inches, and the number of turns from thirty to seventy. It is only necessary that two identical coils be built, one for the sender of Fig. 1, and the other for the receiver of Fig. 4. Fig. 2
    In setting up the sender it will be found that one end of the tuning coil must be attached to the contact post of the buzzer, which is marked 3 in Fig. 1; this can be done by removing the cover of the buzzer and wrapping a bare copper wire firmly about the post. Care must be used to prevent the contact wire from touching the metal base, however, or the operation of the buzzer will be stopped. Binding post 2 is to be connected with "earth" as indicated at E in the diagram. The earth connection is easily made by running a wire to a water pipe or steam radiator and wrapping the bare end tightly about a scraped or plated portion of the pipe. The upper end of the tuning coil is to be led to the aerial or antenna wire, at A. This antenna may be of any convenient size, but the larger it is the farther you will be able to signal. For transmitting from room to room within the house, it will be sufficient to string some twenty feet of wire around the picture moulding near the ceiling.
    If you have set up the apparatus properly the buzzer will hum strongly as long as you hold down the sending key and thus close the battery circuit. By pressing the key for short and long intervals you can produce short and long buzzes which correspond to dots and dashes in the Morse telegraph code; in this way messages can be spelled out letter by letter.

A  Microphone  Receiver

    For the receiving station you will need to make another ground connection by fastening a wire to the steam or water pipes, and then the next thing is to build a second antenna or aerial wire system exactly like that at the sender. The second tuning coil, an old dry cell (preferably one which has become very weak), a telephone receiver and the microphone detector are to be connected together as shown in Fig. 4. Any telephone receiver will do; you can buy a 70-ohm watchcase instrument from an electrical store for about 75 cents, but if you intend to continue with wireless experimenting it will pay you to invest several dollars in a pair of telephones of high sensitiveness. These will not only make it possible to receive messages from longer distances, but because of the headband with which they are fitted you will be relieved of the nuisance of holding the receiver to your ear and will have both hands free for manipulation of your apparatus. Fig. 3
    The microphone detector is to be made as shown in Fig. 3, which indicates how two large double binding posts are to be mounted upon a hard rubber or wooden base. Two sharp sewing needles are inserted into the upper holes of the binding posts, and between their points is lightly supported a short length of graphite from a soft pencil. The piece of graphite should be about one-half inch long, and should have its ends partially hollowed out so that it will hang easily upon the needle points. It is not to be damped firmly, but allowed to rest so loosely that it may be revolved freely and even slid a very short distance back and forth. Fig. 4

Operation  of  the  Apparatus

    After you have set up both stations according to the diagrams, have someone work the transmitter key, making regular test signals such as "V" or "D", and go to the receiver. Listen carefully to the telephone receiver, and move the graphite piece of the microphone around slightly. You will notice that you can hear every touch; when the microphone is adjusted to its most sensitive condition there will be a continuous slight hiss in the telephone receiver, and even the slightest taps on the table or instrument base will be dearly heard. When the apparatus is adjusted in this way you should hear the buzzes of the transmitter reproduced in your telephone, and so should be able to copy the signals sent out from the transmitting station.
    If you have any difficulty in getting good results, try again with the receiver nearer to the sending station. When you have once transmitted good signals, move the stations farther apart. Remember that it is necessary to have good ground connections, that the two tuning coils must be exactly alike, and that the sending and receiving antennas must be identical. If you are able to erect fairly large aerials for the two stations, such as, for instance, sixty foot lengths of wire supported by trees or poles, you should be able to transmit signals a distance of five hundred or a thousand feet; with larger aerials even greater distances can be covered. Begin in a small way, however, and make your progress a step at a time.
    If you are near a commercial or Naval wireless station you will be able to receive signals from it by using the apparatus of Fig. 4; better arrangements which will operate over longer distances will be explained in later articles, however. The microphonic detector of Fig. 3 is quite useful when connected to a commercial wireless tuner, and knowing how easily it may be built from material commonly at hand may be of value even to the commercial wireless operator, in times of emergency.
    You will find it important to become a good telegraph operator if you propose to continue wireless experimenting. There are a number of pamphlets and books published which explain methods of learning the Morse code; Chapter IV of "The Book of Wireless," by A. Frederick Collins, gives a good method to follow. Cards showing the International Morse Code in full may be obtained from the Radio Inspectors' offices at Boston, New York, Baltimore, Savannah, New Orleans, San Francisco, Cleveland and Chicago. It is only by constant practice that you can become proficient.
October, 1916, pages 627-629:

II.--Construction  of  a  One-Mile  Wireless  Transmitter
IN  AN article published last month, directions were given for putting together a little buzzer wireless telegraph set which would operate over a distance of a few hundred feet or even more. This small outfit was sufficient to demonstrate such of the principles of wireless telegraphy as should be known by every student and to send messages from one house to another nearby. The receiver was sensitive enough to pick up messages from commercial stations for some distance around, provided that a fairly long antenna wire was connected to it and properly tuned.
    The amateur who has built and tested the buzzer set will want next to own and operate an outfit with which he can signal over greater distances. It is the purpose of this article to describe the construction of a wireless telegraph sender which can be made cheaply and easily, and which will give good strong signals at a suitable receiving station located as much as a mile or more away. The apparatus for the receiver will be taken up in later articles; the experimenter may well spend the intervening time in building his sender.

Transmitting  Coil

    One of the first requisites in increasing the distance over which messages can be sent is to increase the effective power of the sender. The buzzer run from a couple of dry cells is not strong enough to make waves which will carry very far, so it becomes necessary to get an instrument which will do better. Such an apparatus is the ordinary induction or spark-coil. The amateur may build his own spark-coil by following the descriptions which are given in a great many books on experimental electricity, but in the long run he will find it cheaper and more satisfactory to buy one. An automobile jump-spark coil is about as good a small induction coil as can be obtained. Often it is possible to get one at a nominal price from a garage or an electrician in the neighborhood. Even if purchased new from an electrical supply house, a good coil capable of giving a 1-in. spark between needle points in air will not cost more than three or four dollars. Fig. 1
    There is also needed a Morse key, for sending the dots and dashes which make up the signal letters. This may be an ordinary telegraph key, which costs about seventy-five cents, or even a "strap" or signal key of the kind that sells for only twenty-five or thirty cents. If he desires, the experimenter may build his own key as shown in last month's article. For the heavier currents used in the spark-coil (as compared to the buzzer) it is a good plan to use larger key-contacts than those illustrated. They may be made by soldering copper washers on each of the contact screws.
    To furnish power for the coil, the best thing is a 6 or 8-volt storage-battery. This is quite expensive, however, and also requires occasional recharging. Satisfactory results may be secured by using 12 dry cells connected as shown in Fig. 1. With the battery arranged in this way the voltage is no greater than can be had from 6 cells, but the load is distributed between two sets of cells working side by side in parallel. As a result, the battery will last much longer than if only 6 cells were used. The vibrator on the spark-coil should be adjusted so that it buzzes freely, with a high-pitched sound, whenever the sending key is pressed. A spark-gap connected across the secondary winding will break down whenever the vibrator is started buzzing, and a singing, clear spark will jump across as long as the key is held down. Fig. 2

The  Spark-Gap

    A good spark-gap for the wireless sender can be made as shown in Fig. 2. Two battery zincs, Z, Z1, which can be bought from any electrician, are cut off to about 3 in. in length, leaving the connection screws at the head of each. Holes to fit these a trifle loosely are bored through two stubby brass standards or pillars, P, P1. A smaller hole is bored lengthwise through each pillar and tapped to take a 10-24 machine screw, such as S and S1, to clamp the zinc electrodes in any position desired. Similar screws, S2 and S3, pass upward through counter-bored holes in a hard rubber base, B, and serve to fasten the pillars in place. Hardwood boiled in paraffin may be used for the base, but rubber is better because it is a better insulator. The ends of the zinc rods, where they come close together, should be filed perfectly smooth and parallel. Fig. 3 & 4

The  Loading  Coil

    It is essential to use a "loading coil" with this outfit in order to get the best results, and to make the transmitter meet the requirements of the Federal laws governing the operation of wireless telegraph senders. This coil can easily be made by following the suggestions given in Figs. 3 and 4. Two square boards, about 12 by 12 in. with rounded corners, are first cut out of hardwood about 1 in. thick. A hole ¼ in. in diameter is drilled at the center of each, and counter-bored to about 1 in. in diameter in the bottom of the baseboard at C. Four ½-in. holes are then bored in each, at points a little less than 2 in. from the corners, along diagonals as shown in Fig. 4. Twelve porcelain insulators of the sort shown in Fig. 5 are slipped over each of four ½-in. hardwood dowels, whose ends pass through the ½-in. holes just referred to and are cut off flush with the upper and lower surfaces of the top and base. A long ¼-in. brass bolt is passed upward through the central holes, so that its head drops into the counter-bored space in the base and its threads project a short distance above the top. A washer and nut put on the upper end will then hold the entire framework together.
    Some No. 10 bare copper wire, or some stranded bronze tiller-rope or aerial wire, is to be wound spirally on the insulators. Referring to Fig. 3, the end is first wrapped around the upper front right-hand insulator A and spliced on itself. The wire is then led straight back to the top insulator of the back right upright, then across to the top back left insulator, as shown by the dotted line, then forward to the top front left insulator, and then to the next lower front right porcelain. The winding is continued as shown until the last insulator, B, is reached; there the wire is made fast by splicing, as before.
    Two connected clips must be made or purchased. The spring testing clips sold by electrical supply houses are admirable for this, though anything of the sort will do. Flexible wires are soldered to each of them, so that connection to any part of the bare wire-spiral may be made merely by clipping on the desired point. Fig. 5 & 6
    The laws permit amateur wireless stations to use any wavelength up to 200 meters, provided that the wave sent out is sharp and pure. This means that the aerial wire system to be used with the sending apparatus described must not be more than 75 ft. long, measured along the conductor from its top to the ground connection. It is a good plan to use two wires about 50 ft. long running side by side to the top of a tree or chimney or specially built pole, keeping the wires about five feet apart by fastening them at each end to a light wooden spreader. The top, and in fact the whole aerial, must be thoroughly insulated, if good results are to be secured. An excellent plan for preventing electrical leakage is to connect in series, with loops or rope, five or six porcelain insulators of the kind used in building the loading coil (Fig. 6). These are inserted between the spreader which carries the antenna wires and the rope halyard which is used to haul up the aerial. Similar strings of insulators must be used to guy out the bottom of the aerial. Where the lead-wire enters the house and connects to the instruments it should pass through a thick porcelain tube, as shown in Fig. 7. Fig. 7
    The ground connection may be made by wrapping several turns of bare copper wire tightly around a scraped water or steam-pipe. The connection should be made at a point near to the sending instruments.
    If no water pipes are available, a large copper or iron plate may be buried deeply in moist earth. As a rule, though, such earth connections are not as satisfactory as a pipe forming part of the town water system.

Connecting  the  Set

    The several instruments making up the complete sending set must be connected up as shown in Fig. 7. The spark-gap should be adjusted with its electrodes quite close together--never more than 1/8 in. apart and at least half of the loading coil is to be put in series with the antenna. Unless a large part of this coil is used the transmitter will not radiate pure, sharp waves, and its use will violate the law and make its operator liable to prosecution by the government. If the spark-gap is kept short and a considerable portion of the loading coil used, there will be nothing to fear so long as neither of the aerial wires is over 75 ft. in length.
    Whenever the key is pressed, if the set is properly connected and adjusted, a bright, snappy, singing spark will jump across the gap. Each spark starts a train of high frequency currents oscillating back and forth in the aerial wires, and a train of electromagnetic waves is radiated into space. A suitable wireless receiver located where a portion of these radiated waves will reach it, will pick up some of their energy and produce from it a sound which indicates the dot-and-dash buzzes of a Morse signal.
November, 1916, pages 789-792:

III.--The  Construction  of  a  One-Mile  Receiver
IT  IS most important for a student of wireless telegraphy to learn all about the operation of the various forms of receiving apparatus. The best way to become familiar with the instruments is to build and operate them. The simple buzzer-sender and microphone receiver which were described in the first article of this series served to illustrate the principles which are followed in all wireless apparatus, but were of such small signaling range that they could not demonstrate fully the details of modern instruments. The one-mile transmitter shown last month, however, when used in connection with the receiver now to be explained, is of sufficient size to approach the conditions of operation existing in commercial radio stations. By studying its action carefully the experimenter can learn much which will be of inestimable value to him in his later practice of wireless telegraph operation.
    The student should remember that the use of a transmitter as powerful as that described in the second article, even though it is a very small one when compared to some of the great commercial plants, may cause interference at nearby receiving stations. He should therefore be very careful to observe all of the regulations and courtesies as to transmitting, and should send only when he actually has a message which he wishes delivered to his communicating station. One of the first habits which a successful wireless operator should cultivate is to refrain from sending except when it is absolutely necessary. Testing of the spark-gap should be done with the aerial disconnected, and code practice should be carried on with buzzers. There is never any objection to the amateur who sends actual messages with a wavelength of less than 200 meters (the range assigned to amateur stations by the Government) but the man who keeps tapping his key and sending out interfering waves which hold up legitimate messages soon becomes extremely unpopular with both the serious amateurs and the professional operators. Fig. 1

The  Detector

    Probably the most important element of any receiving outfit is the detector, which is an instrument for converting the received high-frequency current into pulsations which operate the telephones. The microphone which was described in the first article is a wave-detector of a very easily constructed type, and is always worth remembering for use in an emergency. It is very delicate, but is not so reliable nor so sensitive as the crystal detector which is illustrated in Fig. 1. A well made crystal detector is about the best instrument for all around use that can be had. Apparatus of this type is installed at by far the greatest number of commercial radio stations, and every operator should be familiar with its adjustment and use.
    A side view of a detector-stand, which has been found very satisfactory in practical work, is shown in Fig. 1. The construction should be clear from the drawings, and a brief description. Dimensions are not given, since it is usually most convenient to modify them slightly to suit whatever material may be on hand. The base 1 may be made of hard rubber, fiber or hardwood, and should be about 4¼ in. by 2 in. by ½ in. thick. Four holes to take 8-32 machine screws are drilled in the positions shown in Figs. 1 and 2 and directly beneath the parts and are counterbored from the bottom to about ¼-in. depth to take the nuts and washers. A disk 2, of copper or brass about 1/16 in. thick and 1¼ in. in diameter, is soldered to the head of a machine screw 3 and forms a sort of table on which the crystal-cup may slide. The screw 3 is fastened in place by a washer and nut, as indicated, and is connected to binding post 14 through the channel shown in dash lines in Fig. 2. Fig. 2
    A flat brass cup as at 4, Fig. 1, 5/8 in. in diameter and ¼ in. deep, may be made by cleaning out thoroughly the cap of a shotgun shell. In this is secured a piece of fused silicon, galena, or other sensitive crystal 6 (which may be purchased from almost any wireless supply house) by melting and pouring in solder around it. The heat of molten solder will partially destroy the sensitiveness of some crystals, so it is better to use Woods' metal or a mercury amalgam if it can be obtained; solder will generally do for silicon, however. A hard rubber or fiber ring 5, about 1/8 in. thick, should be forced on over the outside of the completed cup, so that the crystal may be moved around without making contact between the metal cup and the operator's fingers.
    A needle point 13 is to be supported directly above the crystal, and this may best be done by the pillar arrangement shown. A long machine screw 7 is passed down through two bushings 9 and 8, and is fastened below the base by a nut and washer. Between 8 and 9 is clamped a tapered strip of spring-brass 10, to one end of which is soldered a binding-post from the zinc terminal of an old dry cell. The shape of this strip may be seen in Fig. 2, where the upper part of the detector is omitted.
    At the top of the pillar is fastened the adjustment arm 11, which should be made of brass about 1/8 in. thick cut as shown in Fig. 3. The left hole is threaded to take the pressure-adjusting screw 12, Fig. 1, and is slit as indicated at 15, Fig. 3. Thus the screw 12, Fig. 1, may be held snugly by the screw-threads. A hard rubber or fiber hand-wheel should be affixed to the top of 12 by a washer and nut, as in Fig. 1. Connection is made from the screw 7 to binding-post 14' by way of the shorter channel indicated by dash lines in Fig. 2.

The  Telephones

    Next to the detector, the most important thing in the receiving station is the telephone. Any ordinary telephone-receiver will give some sort of results, but to get the loudest signals for any particular set of conditions the best telephones should be used. There are on the market a number of head-receivers, designed for wireless telegraph use. These are usually mounted in pairs, one for each ear, on a flexible headband, and are wound for resistances higher than ordinarily used in wire telephony. Reasonably good results can be secured from two ordinary 75-ohm watchcase receivers, if they are connected in series and mounted upon an improvised headband. Thus, there is no need for any one to be discouraged by the high price of the most expensive types. It is good policy, however, for the student to invest as much as he can spare in good telephones, even if a saving must be effected by cutting down the size of the transmitter. Fig. 3 & 4

The  Blocking  Condenser

    Another essential part of the receiving apparatus is a blocking condenser, which is used to prevent the tuning coil from short-circuiting the detector or telephones. Such a condenser as shown in Figs. 4, 5, and 6 may easily be made. A "fixed condenser" may be purchased from any wireless supply store, but it is a good plan for the experimenter to make one. By doing so not only is the actual construction of the instrument learned, but the weak points which might cause trouble later are located.
    A pattern for the tinfoil sheets is cut as shown in Fig. 4, 2 in. square but each having a lug ¾ in. square at the corner. Thirty of these will be needed for the condenser. It is also necessary to cut out about thirty-five sheets of thin paraffin paper 2½ in. square, as shown by the dash lines in Fig. 4. The condenser is begun by placing a sheet of paraffin paper upon a flat surface, and putting on top of it one tinfoil sheet with the lug at the lower left corner, as shown by Y in Fig. 4. On top of this foil is placed a sheet of paraffin paper, and upon it a second sheet of foil; this time the lug is turned to project at the upper right corner, X (dotted lines) in Fig. 4. Then a sheet of paper is added, and upon it a third piece of foil with its lug in the Y (lower left) position. Another sheet of paper is put in place, and then a fourth piece of foil with its lug in the X position. Thus paper and foil are alternately added, and the position of the lug changed each time. The result is a pile of thirty sheets of tinfoil separated by thin paraffin paper, fifteen lugs projecting to the left and the fifteen alternate lugs projecting to the right. Care must be taken that none of the alternating sheets of foil touch each other, since this would short-circuit the condenser. Fig. 5
    A holder for the paper-and-foil condenser is made by cutting out two pieces of 1/8 in. or 3/16 in. hard rubber or fiber or hardwood about 2¾ in. by 4 in., and drilling four holes in each as shown in Fig. 5. An 8-32 machine screw is passed through each of these holes, washers being placed between the clamping pieces in such number that the condenser is firmly gripped. The upper right and lower left screws X' and Y' clamp the groups of tinfoil lugs X and Y, as shown in Fig. 6, and the binding posts X" and Y" mounted upon their upper ends serve to make electrical connection. The other screws 16 and 16' are merely for mechanical strength.
    When the condenser is finished, paraffin may be melted and poured in to fill the entire space between the two clamping plates. If the construction has been careful and if the condenser is in good condition, when a dry cell and telephone are connected in series with the binding posts X" and Y" only a very faint click will be heard as the circuit is made and broken. If the condenser is short-circuited (and therefore useless until repaired) the telephone will click as loudly with it in series as when connected directly across the dry cell. Fig. 6

Additional  Apparatus

    In the next article there will be described the buzzer-testing arrangement which is used to adjust the crystal-detector to its sensitive receiving condition, and a complete wiring diagram for the entire transmitting and receiving station will be given. The manipulation of the apparatus, the methods of calling and answering and of sending messages, as well as the construction and use of the antenna change-over and the detector-protecting switches will be discussed. The remaining instruments needed for the receiver are even easier to construct than the detector and condenser outlined here. It would be a good plan for the experimenter to complete his transmitter (as described last month) now and the apparatus shown in this article, so that he will be all ready to put his station into complete operation soon after the appearance of the next article. It should be remembered that for exchange of messages between two stations it will be necessary to build two of each of the instruments described, so that each station may be completely equipped and thus prepared both for sending and for receiving.
December, 1916, pages 949-952:

IV.--Simple  Adjustments  and  Connections
IN  THE article published last month there were given descriptions of a crystal-detector and stopping-condenser to be made and used in connection with the transmitting set of the October article, for sending wireless messages over a distance of a mile or thereabout. Both the detector and the condenser are of types which can later be used in receiving stations which will pick up the messages from large commercial or government plants not only nearby, but hundreds of miles away. With the small sender using a spark-coil, however, the range will be limited to a mile or so, unless the aerials at both stations are large.

The  Test-Buzzer

    In using a crystal-detector it is necessary to be able to find out instantly whether or not the adjustment is sensitive. When the needle-point bears lightly upon some parts of the crystal, the receiver is sensitive and able to translate messages coming from a distance; with the contact at other points, however, the instruments seem absolutely dead.
    Obviously, to be certain that messages can be received effectively, one must be sure that his detector is properly adjusted. The best way to do this, and the way which is used by the professional operators in most large stations, is to take advantage of the feeble signal-waves induced by a buzzer. By setting up a small sending-outfit, such as described in the September issue of the POPULAR SCIENCE MONTHLY in the first article of this series, the sensitiveness of the detector may be tested by listening in the receiving telephones and at the same time pressing the testing-key. Fig. 1
    Figure 1 shows how to wire up the buzzer, strap-key and dry cell described in the first article. The only difference from the little sender used to signal from one room to another is that the vibrator-contact post of the buzzer is connected to a miniature aerial wire only a foot or two long, instead of to a genuine, full-sized antenna. The miniature aerial is run along the table about 2 or 3 in. from one of the wires leading to the detector, as indicated in the illustration, Fig. 2.

The  Change-Over  Switch

    In order to shift connections from sending to receiving, there must be provided a good-sized double-pole double-throw knife-switch. The lever arms of the switch should be at least six inches long, and the jaws should be mounted upon a slate, marble, or fiber base a corresponding distance apart. If the switch used is too small it will not have enough insulation to prevent the sparks from the secondary of the induction-coil from jumping to ground by way of the receiving contacts.
    A second-hand knife-switch of this size and type can be bought for about one dollar or less; if none can be obtained, it is not difficult to improvise from 1/8 by ½-in. strip copper, an instrument which will work perfectly. It is only necessary to observe closely the construction of the big knife-switches of the double-throw type, in some central station, and to imitate them as accurately as possible. A number of brief articles have been published in the technical magazines, giving details of construction and dimensions for such switches. The connections for the changeover switch are shown in Fig. 2.

The  Detector-Protecting  Switch

    When the wireless station is completely equipped with detector and spark-coil, it is essential to make some provision which will protect the delicately adjusted crystal from the violent impulses set up by the transmitter. The simplest way to do this is to connect a small single-pole switch (either a knife-switch or a lever-switch of almost any sort will do) directly across the terminals of the detector. In the wiring diagram of the complete station, Fig. 2, the detector-protecting switch is marked S; the wires leading from it to the binding posts of the detector should be kept as short as possible; otherwise they may pick up enough current from the sending-spark to "knock out" or destroy the sensitive adjustment of the crystal-detector. When receiving, the protecting-switch S must be open, so that the detector can operate to rectify the currents produced in it by the incoming waves. When sending, the switch must be closed. In this position the heavy induced currents are shunted past the detector and the adjustment is not disturbed by them.

Connecting  the  Complete  Set

    In addition to the parts of the receiving station fully described in last month's article, the various elements of the transmitter illustrated and discussed in October will be needed for a complete sending and receiving station. In fact, a complete set of parts is necessary for each terminal of the proposed wireless "line." The following must, therefore, be at each plant:

STATION:
    1 Antenna and support    See September and October articles.
    1 Loading Coil       "  October article.
    1 Ground Connection       "  September  "
    1 Change-Over Switch       "  above.
    Necessary wire for connections.

SENDER:
    1 Set of dry or storage-cells    See October article.
    1 Sending Key       "         "          "
    1 Induction Coil       "         "          "
    1 Spark-Gap       "         "          "

RECEIVER:
    1 Crystal-Detector    See above.
    1 Stopping-Condenser       "      "
    1 Pair of Telephones       "      "
    1 Test-Buzzer       "  September article.
    1 Strap-Key       "          "            "
    1 Dry Cell       "          "            "
    1 Detector-Protecting Switch       "  above.

    The above-named elements of the complete station must be carefully connected together as shown in Fig. 2. It is a good plan to use No. 16 or No. 18 lamp-cord for the wiring of a set such as this. The twisted pair should be separated and smoothed out, and the single conductors used independently.
    It is necessary to keep the transmitting apparatus well away from the receiving instruments. The loading coil, for example, should not be nearer than two feet to the detector, telephones and stopping-condenser. As explained in the second article of this series, the lead-wire from the loading coil out to the aerial must be well insulated if good work is to be done. It is very important that the change-over switch be well insulated, also, for three of its contacts are subjected to the full sparking potential of the transmitter (see the diagram of Fig. 2). Fig. 2
    The best plan for beginning work is to have the two antennas, one at each station, as nearly alike as possible. If their form and height cannot be made identical, they should at any rate have exactly the same length of circuit. That is to say, there should be the same number of feet measured from the ground connection up through the spark-gap (but not through the loading coil) to the distant insulated end of the antenna, within a few per cent. In this case, i. e., with the lengths practically identical, the loading coils at the two stations can be put entirely in circuit, and the apparatus will be approximately tuned for the interchange of messages.
    If one of the aerials is longer than the other, less of the loading coil should be used at that station than at the other. The exact point to clip on to the wire of the loading coil can be determined only by experiment. By trying every turn, it will be found that some one position is best both for sending and receiving messages. The wire in the loading coil has the effect of lengthening the aerial; it is therefore perfectly clear that, since it is desired to have both antenna systems of the same total length, less of the loading coil must be included in circuit with the longer antenna wire. The coiled wire is more effective in increasing the station's wavelength than the straight wire in the aerial, however; so less of it needs to be added than one would imagine if he merely considered the difference in the lengths of the two aerial wires themselves.

Adjusting  and  Operating

    When the apparatus is set up as shown in Fig. 2, the first thing to do is to put the transmitter into operation. Throw the change-over switch to the left-hand or sending side, and set the spark-gap at about 1/16 in. separation. Making dots and dashes with the key, adjust the induction-coil vibrator to the position which gives a clear, sharp spark between the electrodes of the gap. The spark should be white and snappy, and should sing with the tone of the vibrator. If you cannot get this kind of spark, the set is not working properly and you must go over the antenna insulation to be sure that it is good. If the coil gives a good spark without the aerial connected with it, but won't spark when the antenna and ground are put in the circuit, it is proof that the insulation is not good enough, or that the spark-gap is too wide for the power of the coil. The gap should not be opened more than 1/8 in. at any time. Fig. 3
    Having adjusted the transmitter, swing the change-over switch to the right-hand or receiving side. Put on the telephones, see that the detector-protecting switch is open, and hold down the strap-key connected with the test-buzzer. Move the needlepoint of the detector around over the surface of the crystal, with light pressure, until the loudest signals are heard in the telephones. The detector is then adjusted and the receiver is ready for use.
    The next step is to arrange a sending schedule with your friend who operates the other station. At some fixed time, say four o'clock, let him close his detector-protecting switch, throw his change-over switch to the sending side, and send some predetermined test signal such as "B" in Morse, over and over again, for five minutes. During these same five minutes have your telephones on, your detector-protecting switch open, your detector adjusted to its best sensitiveness, and your change-over switch in the receiving position. If you have built your apparatus correctly and have set it up in accordance with the instructions of these articles, you should have no difficulty in recognizing the "dash-dot-dot-dot" signals being sent from the other station. Promptly at 4:05 your correspondent should stop sending, throw his change-over switch to the receiving side, open his detector-protecting switch, put on his telephones and adjust his detector. At the same time you should go through the opposite change-over, and begin to send him test signals for five minutes. If all is well he will "pick them up" at once, and when you stop at 4:10 he will be ready to reply to you by wireless that he has heard you; you can then give him the corresponding information and proceed to exchange messages.
    You must always bear in mind, however, that whatever your station or his sends out will be heard by other stations which happen to be within range and tuned to the same wavelength. Your signals may even cause interference, and prevent the other stations from reading important messages addressed to them. For these reasons, only such transmitting as is necessary should be attempted; and the Government regulations as to the use of a pure wave shorter than 200 meters should be strictly observed. As pointed out in the October article, if over half the loading coil is used at each station and if neither antenna is more than 75 ft. in length, the federal requirements will, as a rule, be met.

Station  for  Receiving  Only

    If it is desired to transmit messages in only one direction, the change-over and detector-protecting switches may be omitted at the receiver. A loading-coil will be necessary, but since it is to be used for receiving only it may be made as described in the September article instead of highly insulated in accordance with the October description. The transmitter should be connected as shown in October, and the receiver should be wired as in Fig. 3. The comments in this article as to the adjustment will still apply, except that the two switches need not be considered.
    The receiving apparatus described here will work one mile easily, and is capable of hearing signals much farther away. In the next article an adjustable receiving set will be discussed, by the use of which signals may be heard from stations located hundreds of miles away.
January, 1917, pages 153-156:

V.--Increasing  the  Range  of  the  Receiver  by  Tuning
THE December article of this series gave directions for completing and operating the receiver which was designed for working over distances of about one mile. The apparatus was intended for use with the small transmitter which was described in the October number of the POPULAR SCIENCE MONTHLY; when a larger transmitter is used, the possible working range is, naturally enough, considerably greater. It is necessary for each receiver to be adapted for the wireless waves sent out by the transmitter to which it is listening, however, if the best results are to be secured.

Production  of  High-Frequency  Oscillations

    Every transmitter for wireless telegraphy which is permitted to operate under the present radio laws must send out what is called a pure and sharp wave. That is to say, the sending apparatus must be so adjusted that its radiation has a single and definite wavelength or frequency. The main purpose of the regulations insisting upon this condition of sharpness and purity of wave is to enable a receiving station to "tune-in" a sending station without interference. Senders which emit waves neither sharp nor pure are the cause of interference, and sometimes prevent all other stations in their neighborhood from working effectively.
    To understand this matter of tuning, one must realize first of all that the currents in a wireless telegraph antenna are of the high-frequency alternating sort, which change in direction very rapidly. In a simple transmitter, with the spark-gap directly in series between the antenna and the ground, as described in the October and December articles, the effect of the induction-coil is to charge the aerial by storing in it a quantity of electricity just before each spark takes place. The induction-coil secondary tries to charge the antenna up to its maximum pressure, or to drive into it all the electricity possible. Before this top-point is reached, however, the spark-gap breaks down and the charge of electricity rushes across it to the ground. Because of the electrical property of the aerial wire (and of the coils in series with it) called inductance, the charge overshoots itself somewhat, and the antenna is left charged in the opposite direction for an instant. Therefore, in the natural attempt to restore equilibrium or electrical balance, the charge rushes back out of the ground into the aerial; this time it overshoots also, but not by so much. The electrical energy thus oscillates back and forth, like a swing left to itself, until it is all used up in radiation, or in losses in and near the circuits.

Period  and  Frequency

    A certain amount of time is required for the electrical charge to travel from the top of the antenna to the ground and back again, just as a certain time is required for a pendulum to swing from one end of its beat to the other and back again. This amount of time, measured in seconds, is called the period of the oscillation. The longer the wire, the longer the time for each trip of the current, and the longer the period. The number of times the electrical charge makes the round trip in one second is called its frequency, and this of course may be calculated by dividing the period, in fractions of one second, into one second. For example, if the period of oscillation of an antenna is one millionth of a second--which merely means that the charge takes that long to travel up and down the antenna once--the frequency is one-millionth second divided into one second, or one million. This is the number of trips the charge will make in one second.

Wavelength

    Knowing the frequency of any electrical oscillation or high-frequency alternating current, one can immediately compute the wavelength which it will produce if it flows in a suitable wireless-telegraph antenna. The rule is simply to divide the frequency per second into three hundred million. The answer to this little problem in Arithmetic gives at once the wavelength in meters. For example, taking the frequency of one million per second quoted at the end of the paragraph immediately above, it is found that the corresponding wavelength is three hundred meters. The following table of frequencies and wavelengths will be helpful.
TABLE  I.
     Period  in     
Seconds
Frequency
    Per  Second    
Wavelengths
in  Meters
0.00000033
0.00000066
0.000001    
0.000002    
0.0000033  
0.0000066  
0.00001      
0.00002      
0.000033    
3  000  000
1  500  000
1  000  000
    500  000
    300  000
    150  000
    100  000
      50  000
      30  000
    100
    200
    300
    600
  1000
  2000
  3000
  6000
10000
As the frequency of the sending current varies, the wavelength increases or decreases with it

    From an examination of this table it becomes very clear that as the period is increased, the frequency decreases and the wavelength increases. Remembering that the longer the antenna wire, the more time it takes for the charge to pass from the top to the bottom and back again, or the longer the period, it is easy to see that the longer the aerial wire (including the coils connected in series with it), the greater the wavelength will be. As a matter of fact, the fundamental wavelength of a simple aerial, which is its wavelength without any coils in series, is about 4.2 times its actual length measured from ground to top end. To use this rule, both height and wavelength must be measured in the same unit. A table showing the fundamental wavelengths of several heights of plain vertical antenna wires is given in the next column.
TABLE  II.
Height  (Length)
in  Feet
Fundamental  Wavelength
(in  Feet) (in  Meters)
  35
  50
  75
100
125
150
147
210
315
420
525
630
  45
  64
  96
128
160
192
The fundamental wavelength of a plain vertical antenna increases with its length

    The table is strictly applicable only to plain vertical antennas without any loading coils, but it may be used for the approximate fundamental wavelengths of L-shaped antennas if the total length of a single wire is used instead of that of the vertical lead alone. If the antenna is T-shaped, the length from the ground to the center of the flat-top and from there to one end should be used. The two parts of the flat-top should have the same length, as measured from their junction with the vertical lead. Where several wires are used in parallel, whether in a horizontal or vertical antenna, the length is taken as that of one of the wires, and not of the total amount of wire in the aerial system. Neither Table No. II, nor the simple rule must be used when loading coils are connected; for the wire on such coils is much more effective in increasing the apparent length of the antenna than is the straight-away portion.

Tuning

    The bearing of the foregoing discussion upon the adjustment of the receiver's tuning to get the greatest distance becomes clear on considering that, for this to be obtained, the receiving set must be adjusted "in tune" with the wavelength it is desired to receive. In the simple wireless set described last September, some small degree of tuning was secured by making the antennas and the tuning-coils alike at the sending and receiving station. This receiver, as shown in Fig. 3, has the detector right, in series between the aerial and ground connection. The result of this arrangement as regards tuning is that the high resistance somewhat spoils the sharpness of adjustment. If one makes material changes in the length of the aerial wire, or in the number of turns of coil used, a weakening of signals is noticed. The tuning is neither critical nor "sharp," however, and even approximate adjustments will give about as good results as exact ones. Fig. 3
    When the receiver described in the December article is used, as shown in Fig. 4. the adjustment is much more accurate. Here the detector is removed from the antenna circuit, and the aerial is connected directly to ground through a loading-coil. As a result the effect of the coil is considerably increased. It was pointed out that the coils and antennas at sender and receiver should be made exactly alike, if possible, but that if there should be any difference in the aerials the shorter one should have more turns of coil connected in series with it. This was, of course, to make the effective lengths of the two antennas the same, so that they would be tuned to the same wavelength.

Receiving  Various  Wavelengths

    With the senders and receivers limited in their activities to communication between a single pair of stations, it is usually not necessary to provide for variation in the tuned wavelengths of either. That is why the simple arrangement of Fig. 2 in the December article could be used. When it is desired to receive from a large number of outside transmitters, all using different wavelengths, it is necessary to provide apparatus whereby the effective length of the aerial at the receiver may be varied to suit the incoming wavelength. A large number of arrangements may be used for this purpose. Some of them tune the receiver very sharply, or in other words make it respond energetically to a very closely restricted range of wavelengths for each adjustment. Other sets of connections are less critical in adjustment, but easier to handle.
    The simplest variable tuning instrument for use at the receiver is the so-called "single-slide tuner." This is merely an inductance coil with a sliding contact whereby the number of turns in circuit may be varied at will. It may be used in place of the tuning coil shown in Fig. 3, and will allow some latitude of adjustment, though the tuning is very broad and unsatisfactory. A better mode of connection for the single-slide tuner is that of Fig. 4, in which the detector is put in a side or by-pass circuit; this gives sharper tuning and fairly strong signals. Fig. 4
    A still better tuning arrangement fuses the "double-slide tuner," which has two variable contacts. In the catalogs of manufacturers of radio apparatus there are to be found a large number of diagrams showing different ways to connect the double-slide tuner; but the best possible results are to be secured from the arrangement of Fig. 5. One end of the coil is connected to ground and one of the sliders to the antenna. The larger the amount of coil between the grounded end and the first slider, the longer the effective length of the aerial and the greater the wavelength for which it is tuned. The grounded end is also connected to one side of the blocking condenser described in the November article, and the other slider is connected with one terminal of the crystal-detector also illustrated in November. The telephone has one lead connected with ground, and the other joins the open sides of the detector and blocking-condenser. The test-buzzer, which is not shown in Fig. 5, is to be arranged as explained in the December article, so that the crystal-detector can be adjusted to its maximum sensitiveness without waiting for signals from outside stations.

Operating  the  Variable  Receiver

    In working the apparatus set up as in Fig. 5, the first step is to make sure that the detector is adjusted to a sensitive point and that the connections are all secure and in good condition. Then the slider connected to the detector is set at a position about half-way along the coil, and the antenna slider is moved back and forth slowly along the length of the entire tuner. When the station within range starts to send, his signals will be heard in the telephones; it will be noted that the dots and dashes are loudest with the antenna-slider at some particular setting. Leaving the antenna-slider at this point, the detector-slider is moved back and forth until the position giving the best signals is found. This is the tuned or approximately tuned adjustment of the receiver for the specific wavelength being received.
    It often happens with double-slide tuners of this type that there are several positions for both sliders which give good signals for a single wavelength. Therefore it is a good plan to try several settings of the antenna-slider, varying the other contact at the same time; thus one can sometimes find a single pair of settings which give markedly improved signals. This double-setting effect is more noticeable with other connections than that shown in Fig. 5, but with some combinations of antenna and detector it may be found in this arrangement also. The best thing to do is to have a scale, marked in number of turns or in centimeters of coil; fastened close to each slider, and then to make a tabulation of the best settings for each station as it is heard. Such a table makes it possible to leave the apparatus tuned quite close for any desired outside station, and to feel confident that its messages will be received whenever it starts to send. Fig. 5

The  Apparatus

    For receiving long distances it is merely necessary to combine the crystal-detector and blocking-condenser, recently described, with a pair of good head-telephones, a fairly long antenna, and a double-slide tuner, in order to receive at night from commercial stations hundreds of miles away. It is wise economy to buy good telephones; for, as a general rule, the more money invested in them (so long as they are purchased from a reliable dealer) the greater will be their effectiveness. For receiving from amateur stations, which are required by law to operate on waves less than 200 meters in length, it is not desirable to have an antenna longer than 100 feet or so, though longer wires may be used if a condenser is connected in series, as will be explained in later articles. To get the best results from the longer wave stations, such as the commercial plants which use 600 meters and the Naval stations on waves of 1000 and 1200 meters, it is best to have antennas about 200 feet long. Using the ordinary crystal-detector, a double-slide tuner, good telephones and a single aerial wire swung between chimneys 750 feet apart and 40 feet above the ground, it is not unusual to receive messages 600 or 800 miles at night during the winter.
    The loading-coil described for the one-mile sender, in the October article, may be used in the diagram of Fig. 5, if two clips are utilized. This will not give a very long range of wavelengths, but will do for experiments. For the best receiving, a modification of the double-slide tuner, which is easily made, will be described next month.
February, 1917, pages 309-313:

VI.--Simple  Adjustments  and  Connections
Fig. 1 SKILL in receiving, which includes both the ability to "copy" Morse and Continental signals rapidly and accurately, and facility in the adjustment of the receiving instruments, is one of the most important things for a radio-operator to develop. Knowledge of the transmitting apparatus and its manipulation, as well as a clear and firm method of handling the key in sending, is of course essential; but without knowing how to adjust the receiving tuners and detectors so as to get maximum strength signals with the least interference a wireless man soon finds himself in trouble. The only way to gain the needful familiarity is through practice with the apparatus itself; and at the same time the student must realize how and why his instruments work. Unless it is clear to him just what effect will be produced, and why, whenever he makes an adjustment or changes a setting, he will not progress very rapidly.
    In the January article of this series it was shown that the detector and stopping condenser previously described could be assembled with either of two types of tuning coil (the "one-slide" and the "two-slide" varieties) so that messages could be received from commercial and naval stations hundreds of miles away. Receiving practice gained through use of receiving apparatus in the ways indicated will form a valuable foundation for further advances in the use of more complicated tuning arrangements, and the experimenter should familiarize himself with the action of the circuits shown in Fig. 3, 4 and 5 of the last article. Fig. 2

Methods  of  Connecting

    The method of connection shown in Fig. 5 of the January article is the most effective of all the simple arrangements. It requires a tuning coil with two variable contacts or sliders, in addition to the usual detector, telephones, and stopping-condenser. This same tuner may of course be used for tests of the single-slider "hook-ups," by using only one of the movable contacts; such trials will demonstrate beyond doubt the fact that, when it is properly adjusted, the two-contact arrangement gives louder signals with greater freedom from interference. Somewhat more skill is required to get the best results from the two-slide than from the one-slide apparatus, but the effort is more than repaid.
    Sliding-contact tuning coils for use in any of the circuits described may be purchased from the wireless supply houses, and will give reasonably good results. It is very easy to make such tuners, since all that is required is a coil about 3 in. in diameter and 8 in. in length, wound with insulated wire of about No. 22 gage and fitted with two contacts which slide along rods supported above paths from which the insulation has been scraped. Such instruments have been used in commercial radio-telegraphy, and were very common in the stations of eight or ten years ago. It has been found, however, that the sliding contact upon the coil itself is not particularly desirable, since the slider finally wears through the wire, and there is always difficulty in maintaining good contact. Further the slider usually short circuits several turns at its point of contact, and thus causes mistuning and loss of signal strength. These disadvantages of the sliding contact, taken together with the poor circuit design often employed, have brought the direct-coupled circuit of the sort described into disrepute; there is really little choice between the direct and inductively coupled receivers, however, provided that both are properly built.
    By making the loading and transformer coils in separate units, and fitting each with switch contacts instead of sliders, it is entirely practicable to produce a receiving installation which has all the selectivity of the inductively coupled type and still avoid some of its disadvantages. Such coils are described below, and their mode of connection is shown in Fig. 7, which corresponds to Fig. 5 of the December article. Fig. 3

The  Loading  Coil

    This instrument is shown in Fig. 1. The basic piece is a paper tube of about 4 in. diameter and 8 in. length. Beginning about 1 in. from one end, No. 20 gage double silk-covered magnet-wire is wound on evenly for about 6 in., which will take 149 turns. A tap must be taken out for each of the first ten turns, and one for each tenth turn thereafter, as shown in Fig. 4. The best plan is to fasten the end of the wire, before beginning winding, through two small holes punched in the paper tube, leaving about 2 in. of wire free for the tap marked "9" in Fig. 4. Then a single turn is wound on, and a small loop twisted in as shown in Fig. 2. This twist and loop stand up from the surface of the coil, and the wire leading to contact "8," Fig. 4, is later soldered to the loop. The second turn of wire is then wound on the tube, and another twisted loop for the contact "7" is made. Thus a twist is put in for each of the contacts, at the end of each turn, till that marked "0" is reached. Then ten turns are wound without a tap, the twist for "10" being taken out at the 19th turn of the whole coil. Similarly taps for "20", "30", etc., are made at each tenth turn thereafter. Fig. 4
    The completed coil is to be mounted between two end blocks of wood or hard rubber, A, in Fig. 1. Three small pieces are fastened to the inside faces of each of the end blocks, as shown in dotted lines, to keep the coil from slipping sidewise; and the whole is held together by a piece 1 in. square passing through from end to end in the center of the coil. Screws B, with washers under their heads, pass through holes bored in the end blocks and clamp the tube by threading into the ends of this central stick.
    As further shown in Fig. 1, a switch-panel is mounted on top of the end blocks. This should be made of hard rubber or fiber, about ¼ in. thick, though hardwood 3/8 or ½ in. thick will do. It is fastened to the end blocks by means of two wood-screw binding posts D, D1, Fig. 1 and 3, and the screws E, Fig. 3. On it are mounted two rotary switches, one having 10 and the other 15 points, as shown in Fig. 3. Any type of switch-arm will do for these, but the easiest manipulation will be obtained if a center-knob type is used.
    The 10-point switch should be marked "Units" and the 15-point "Tens"; the buttons are to be numbered from zero to 9 for the former and from zero to 140 in tens for the latter. The taps from the coil itself are to be connected with these switch-points by means of short soldered leads, as shown in Fig. 4. Great care must be taken to see that no short circuits are made as this wiring is put in place. The central points of the switches, i. e., their arms, have wires leading to the binding posts D, D1. It is best to make the switch buttons marked zero the lowest; then as the switch-arms are turned toward the upper positions, more and more turns of the coil are cut into the circuit between the two binding posts. As is obvious from the diagram, the "tens" switch cuts in ten turns at a time, while the "units" switch gives steps of a single turn. By various combinations of both switches, any number of turns from 1 to 149 can be had. Fig. 5

The  Transformer  Coil

    Another coil very much like the first is now to be built. The mechanical features are exactly the same; but only 8 twisted taps are taken out from the 150 turns of No. 20-gage wire. The end of the wire is connected to a binding post F, Fig. 5, and taps at the tenth, thirtieth, fiftieth and each twentieth turn thereafter are led to the corresponding points of two nine-point switches mounted on the panel. These are marked "Coupling" and "Secondary" respectively, and take the places of the 10-point and 15-point switches on the loading coil just described. Their buttons are marked 0, 10, 30, 50, and so on to 150, in steps of 20 turns, and the soldered connections are made as shown in Fig. 5. The binding post F may take the place of one of the screws E in Fig. 3, and the posts G and H are located as D and D1 in Fig. 1.
    It is to be noted that in the loading coil the adjustment of inductance may be had in single turn steps by the use of two switches, but that there are only two points of connection. In this last described coil there are three points of connection, two of which are variable in steps of 10 and 20 turns from 0 to 150. Thus the two coils, while superficially alike, may be used for very different purposes.

Switches

    It is possible to buy lever-switches from the various supply houses, but usually the type with the knob at the outer end of the switch-arm is furnished. This is not nearly so convenient for tuning as the kind with central-knob shown in Fig. 1, 3 and 6, since considerably greater movement of the hand must be made in order to accomplish a given adjustment. Central-knob switches are not difficult to make; a simple and effective design is shown in Fig. 6. The switch-points I may be of any sort whatever; most supply houses will sell these for one or two cents apiece. The switch-arm K is cut out of thin spring-brass, and is formed of two pieces, each having one end punched to fit over the machine-screw O and the other end bent down to make smooth contact with the buttons I. The arms are fastened to the turned hard rubber knob L by means of the brad or escutcheon pin M, and are further clamped by the two nuts N, N1. The central machine-screw O passes through the switch-panel Q within a short sleeve P and through two washers shown but not lettered. Two additional nuts N11 and N111 are clamped on the inside, just enough play being allowed for free turning when the arm K bears upon the contact points. Connection to the arm is made through the spiral R, the free end of which is soldered to the screw O. Soldered connection is made to the switch-points at J. Fig. 6

Setting  Up  the  Receiver

    The loading coil and the transformer coil as described above are to be combined with a detector, telephone set and stopping-condenser as shown in Fig. 7. The antenna circuit passes through the loading coil from D to D1 and thence to binding post G of the transformer coil. This leads to the "coupling" switch, and the circuit runs from there through whatever part of the coil is cut in, and out to ground through binding post F. The secondary circuit includes that part of the transformer coil cut in by the "secondary" switch, and runs from binding post H through the detector and stopping-condenser to the ground binding post F. The telephones are connected across the stopping-condenser.
    A crystal detector and stopping-condenser suitable for this use were described in the POPULAR SCIENCE MONTHLY for November, 1916, in the third article of this series. The proper sort of telephones to use was also discussed in that article. A test-buzzer arrangement for adjustment of the detector was explained in the December article, and should be combined with the outfit of Fig. 7 in order to make the adjustment easier and more positive.

Tuning

    The circuits of Fig. 7, with the apparatus described, will give excellent results in receiving from the commercial and naval spark stations, if used with an antenna of from 150 to 250 ft. length. For shorter wavelengths than 500 meters the antenna used should be somewhat shorter, and for waves longer than 2500 meters (that on which time signals are sent) it is advantageous to use still longer wires. It is not necessary to erect multiple wire aerials for receiving, nor is great height essential. A 200-ft. single wire, of No. 10 gage copper, or even galvanized iron, swung horizontally between two 50-ft. masts or trees, will prove entirely satisfactory for most purposes. When it is desired to receive the short waves from amateur stations, which are restricted by law to wavelengths under 200 meters, a wire not much over 70 ft. in length should be used. Fig. 7
    In attempting to "pick-up" signals with the outfit of Fig. 7, the best plan is to set the "units" switch on 5, the "coupling" switch on about 30, and then to vary simultaneously the "tens" and the "secondary" switches. As the turns on the loading coil are increased in number, those of the secondary should also be increased. When signals are heard, the best point of the secondary is selected, and the loading coil and coupling switches adjusted to give the loudest responses. One must of course be careful that his detector is adjusted properly before starting to tune; for this purpose the test-buzzer is a great time-saver.
    It will be noted that more turns of the loading coil and of the secondary are needed for long wavelengths than for short ones, and that when the number of "coupling" turns is reduced, the number of turns in the loading coil must be proportionally increased. This is because the loading coil and the coupling turns in series form a primary circuit, whose effective length must be adjusted for the various wavelengths in the manner described last month. It will also be noted that all stations which have the same wavelength will "come in" best with approximately the same settings, and that the wavelength of any station may be estimated roughly by considering the number of turns in the loading and coupling coils which give the strongest signal from it.
    A thing which is very important in the operation of this tuner, however, is not likely to be evident from the first tests made upon it. That is the relation between the sharpness of tuning and the number of turns on the coupling part of the transformer coil. Careful observation will bring out that when the number of turns on the coupling section is reduced, and the loading coil correspondingly increased to the tuning point, better selectivity is obtained. Usually there is a best value of the coupling turns for every station or wavelength, and its use requires the proper corresponding settings of the loading coil and secondary switches. Often when there is interference it is best to use still fewer turns of the coupling section, correspondingly increasing the loading coil, so as to get sharper tuning in spite of a weakened signal. The judicious selection of values for these three coil sections (primary, coupling and secondary), and the proper balancing of signal strength against sharpness of tuning, is one of the items which is most important in commercial radio-telegraphy. Many operators fail to get the most out of their receivers merely because they fail to pick out the best adjustments, or because they do not retune primary or secondary circuits after changing coupling values. Practice with the apparatus of Fig. 7 should make the desirability of correct tuning evident to every experimenter.
    In the next article, of this series there will be discussed the secondary variable condenser. When properly used, this instrument is of great value in increasing sharpness of tuning.
March, 1917, pages 475-478:

VII.--The  Variable  Condenser  for  Tuning  the  Secondary
THE direct coupled receiving tuner which was described in detail last month will prove satisfactory for long distance working, up to the limit fixed by the sensitiveness of the detector and telephones used in connection with it. The loudness of signals, if the coils and switches are properly built, should compare very favorably with the best obtainable by using an inductively coupled receiving transformer of equivalent design. To get the strongest signals, in either the direct or the inductively coupled outfit, the operator must determine by trial the best settings of primary and secondary inductance and coupling, as previously pointed out.

Sharpness  of  Tuning

    Although the simple tuners without variable condenser in the secondary will give loud signals, and in fact about as loud as can be obtained by any arrangement of tuning circuits when working with spark stations, the best sharpness of tuning cannot be secured. When two transmitters are heard at about the same intensity and on only slightly different wavelengths, it is difficult to build up the signals of either at will by the mere adjustment of the coupling and primary and secondary inductances, when the circuit without variable condenser is used. For this reason the tuner, whether inductively or conductively coupled, which has the detector directly in series with the blocking condenser (i. e., which has no condenser directly across the secondary terminals for tuning) is called "broad tuned." However, if a variable condenser is connected as shown in Fig. 1 and 2, the selective powers of the circuit become very much greater and it is called "sharp tuned."
    The details of adjustment which are necessary in order to get the best results from the sharply tuned receiver will be taken up in full next month. The manipulation of this receiver should be second nature to all radio operators, since it is the arrangement of apparatus used by the great commercial companies. In the hands of an unskilled operator better results are sometimes obtained with simpler circuit arrangements; the reason for this is simply that the man does not know how to get the most out of the sharply tuned lay-out. The circuits of Fig. 1 and 2, when correctly adjusted, provide the maximum selectiveness which is to be had in the best commercial receivers in general use. Securing the correct adjustment, once the principles are clearly understood, is merely a matter of practice. It is essential for the student, therefore, to familiarize himself with the actions of such receivers under all conditions likely to be encountered in practice. Fig. 1

The  Variable  Condenser

    Since Fig. 1 is exactly the same as Fig. 7 of last month's article, except for the addition of the variable condenser and a single pole switch for cutting it out of circuit, all of the instruments described in the preceding articles of this series may be tilized. The switch of Fig. 1 is preferably a small single pole, single throw knife-switch, since this type almost invariably gives good contact, though any other form will be satisfactory if kept in good condition. The variable condenser is the important new instrument, and must be of good design if it is to be really useful. The amateur who has sufficient funds at his disposal will do well to buy one of the standard variable condensers now on the market; if he sticks to the ordinary semi-circular rotary type, having a capacity of about .001 microfarad and costing from $5.00 to $25.00, he will be likely to get a good instrument for tuning. The cheapest apparatus, as well as the various freak forms which appear from time to time, are less likely to be satisfactory.
    Since good variable condensers are expensive, as compared with the other parts of short-wave receiving apparatus, a simple and yet good design for making them at home will be described. The condenser made in this manner will prove rugged and suitable for continuous operation, will have good insulation between its terminals, and yet will be found easy and comparatively quick to build. Fig. 2

The  Fixed  and  Moving  Plates

    The plates of the condenser may be made of almost any conducting material. Aluminum about 1/32 in. in thickness is by far the best, since it is light, soft and easily kept flat. Soft brass or copper will prove suitable though heavy, and even sheet tin can be used if reducing the expense is of the greatest importance. The fixed plates are laid out as shown in Fig. 3. They consist merely of rectangles 2 by 5 in. in size. About 3/8 in. from each corner, on a 45 deg. line a hole is drilled to take the supporting uprights. The relation of the plates to a semi-circle of 2 in. radius, which corresponds to the active surface of the moving plates, is also shown in Fig. 3. The rectangular form of plate is shown for the reason that it is the quickest and easiest to cut out. Obviously, if material is of more importance than time, the corners may be cut off and the outer side of the plate held by a single vertical bolt passing through suitably placed holes at or near the center of the upper edge.
    The form of the moving plates is shown in Fig. 4. Essentially, these are portions of circles having a 2-in, radius; they are roughly semi-circular in form, and the exact relation to a half-circle is shown by the dotted lines. A hole to take a 5/32-in. bolt is drilled at the central point where the radii meet, as shown. This design of plate is about the easiest to make of all that have been suggested, and yet is not particularly wasteful of material.
    For the tuning condenser, 13 fixed and 12 movable plates may be cut out. The simplest way is to make a full size templet or pattern out of pressboard, and to scratch the outline of each plate on the material by running a needle around the edge of the pattern when held tightly on the surface of the metal sheet. The plates are then trimmed out with sharp shears, clamped together in a vise and filed to exactly the same size and shape. They must be flattened by hammering, preferably between perfectly flat metal surfaces, until no dents or warping can be seen when the plate is held edgewise to the eye.

The  Top  and  Base

    Hard rubber or horn fiber, from ¼ to 3/8 in. in thickness is the best material for the top and base. If these cannot be used, hardwood about ½ in. thick will do. In the illustration, Fig. 5, are shown the two pieces that form a 5-in. square, and how they are drilled at each corner, in the center, and at the inner corners of the fixed plates, to take the several bolts. The base must also be drilled and tapped near the center, to take the foot or base bearing of the vertical shaft, to be described later. In Fig. 5 there are shown the outlines of both the fixed and moving plates in their proper relative positions, so that no difficulty should be experienced in laying out the holes once the plates have been finished. Fig. 3, 4, 5

Other  Materials

    Practically nothing else is needed except a number of 8-32-hexagon brass nuts (machine screw size No. 8 with 32 threads per inch), a considerable quantity of copper washers or burrs which will slip freely (but not too loosely) over an 8-32-machine screw, six threaded brass rods of 8-32 size and about 5 in. long (the excess is cut off after assembly), a 5 in. by 5/32 in. threaded brass rod with 6 hexagon brass nuts to fit, a number of perfectly flat brass or copper washers of about 5/8 in. outside diameter, which will slip over the 5/32-in. bolt, a little soft brass strip, two small screws and a hand wheel. In Fig. 6 is shown how one of the 8-32 bolts is used to support the top of each corner of the group of fixed plates in proper relation to the top of the condenser itself. As indicated, the completed assembly begins at the top with a hexagon nut and washer, after which comes the 5-in, square top plate. This is clamped in place by means of a second washer and nut, after which come separating washers to space the uppermost fixed plate the proper distance from the lower side of the insulating top, and then that plate itself. After the first plate, enough washers to make a space of ¼ in. are put on the threaded rod, and then the next plate. Thus the thirteen fixed plates are placed on and clamped by the nuts of four of the threaded rods. At the bottom, as shown in Fig. 7, the third nut is screwed upward to hold the plates firmly, and a fourth nut turned on to the point which will hold the lowest fixed plate the correct distance from the upper side of the insulating base. After adding a washer, the base itself is placed on, and then a threaded rubber, wood or fiber foot is screwed on to clamp it all in place. Of course, only the corner posts of the base itself take these feet; the two rods at the inner edges of the fixed plates are fastened by screws and washers on the lower side of the base.
    The two corners of the base, shown at the lower part of Fig. 5, where no fixed plates reach, are supported by the fifth and sixth threaded 8-32-rods in the manner shown in Fig. 6 and 7, with the obvious exception that the plates themselves and their separating washers are omitted.

The  Rotary  Part

    The moving plates are assembled upon the 5/32-in. threaded rod, clamped by the hexagon nuts and separated by the larger washers, as shown in Fig. 8 and 9. Beginning at the top of Fig. 8, it is seen that the hand-wheel, which may be a knurled disk of ¼-in. hard rubber 2 in. in diameter, is clamped at the top of the 5/32-in. rod or shaft between two washers and nuts. Between the lower nut and washer there is gripped the end of a bent metal indicator-hand which points to a degree scale (a cheap protractor makes a good one) mounted on the upper side of the insulating top. Just below this lower nut is a little brass bushing or tube, forced into the insulating top and having the right inside diameter to act as a bearing for the shaft. Below this come separating washers--one or more--and then two nuts which lock each other and hold the moving plates in place. The uppermost moving or semi-circular plate comes next; immediately below it, and separating it from the next plate, are enough of the larger washers to space the plates just ¼ in. apart.
    As shown in Fig. 9, the moving section is built downward, and the lowest plate is clamped in place by two more nuts. The shaft itself continues for about ½ in. farther, and is tapered off to a blunt point so as to reduce friction. Fig. 6, 7, 8, 9
    

The  Lower  Bearing

    In Fig. 9 is also shown how two pieces of 1/16-in. soft brass strip are bent and drilled to form a thrust bearing for the lower end of the shaft. They are secured to the base or lower 5-in. square insulating plate by the two machine screws. Electrical connection is made to the moving plates by soldering a wire to one of these screws, running it out to one of the corner screws which is not associated with the fixed plates, and once more soldering it there. A binding post takes the place of the top nut of Fig. 6 on this outside screw. Similarly, a second binding post substituted for the uppermost nut on one of the screws supporting the fixed plates provides a convenient means of connection with the fixed plates.
    For continued service it will not do to depend upon the electrical contact in the bearing of Fig. 9 alone; a small piece of thin spring wire (brass or phosphor bronze) should have one end soldered to the shaft just above the upper plate of the base bearing, its other end being fastened to the bearing itself after one or two spiral turns are made loosely about the shaft. This will prevent any possible trouble from oxidation of the contacts within the condenser. A stop should be provided to keep the plates from turning more than 180 deg. and twisting off the spring wire.

Assembling  and  Adjusting

    Manifestly the foregoing descriptions, with reference to Fig. 6, 7, 8 and 9, apply to the completed apparatus. The best way to assemble the instrument is to clamp together the fixed plates on their four rods, independent of either the base or top plates of insulating material. When the conducting plates are all set parallel and with the proper ¼-in. spacing, they are laid aside and the moving plates similarly assembled upon the shaft. The base, hand wheel and top are left apart until the moving plates have been correctly spaced and adjusted. Then the moving plates are slipped interleaf-wise between the fixed plates, and the insulating top put in place. By trial the proper number of washers and the best position of the clamping nuts is determined, and the top is then fastened. Next the base is adjusted to the proper position, and all the fastening nuts tightened. Last of all the hand-wheel and pointer are added, and the scale set to the correct position. It is a good plan to cut a strip of celluloid film to just the width of the space between the insulating top and base, and to bend it around the metal parts so as to exclude dust. When this is to be done, the top and base may be made 5½ in. square to give a little additional edge-space in which the celluloid strip may be fastened by gluing bits of 1/8-in. felt next to it.
    The next article of this series will describe fully the use of the variable condenser in the sharp-tuned secondary, as well as methods of adjusting the primary circuit by using the same instrument.
April, 1917, pages 629-633:

VIII.--Tuning  with  the  Variable  Condenser  in  Primary  and  Secondary
LAST month's article described in detail the construction of a simple variable condenser which could be built at home and which is entirely suitable for use in either the primary or secondary circuit of any normal radio receiving apparatus. It is the purpose of the present article to show how one or more of these variable condensers may be used to the greatest advantage, both for increasing sharpness of tuning (and thereby lessening interference difficulties) and for increasing signal strength. The great advantage which is secured immediately upon the use of a variable condenser is in the flexibility of circuit adjustment which can thus be secured. Practically all the effects of mere tuning to wavelength can be obtained from variable inductances alone, using fixed condensers. However, to change the ratio of inductance to capacity, which is one of the important factors governing sharpness of tuning, it is necessary that both the condensers and the coils be variable; if only one element can be altered, it is not possible to secure any desired ratio at any desired wavelength.
    A second advantage gained by using variable condensers is the mechanical simplification of the inductance coils connected with them. Since the condensers are continuously adjustable, and can therefore be set at any desired value of capacity without the limitation of switchpoint steps, it is unnecessary to take out many coil taps in order to get close variation of inductance. Any wavelength can be tuned to with accuracy, with only a few steps or connections on each inductance coil, because the condenser fills in the tuning range. Fig. 1

Advantages  of  the  Sharply  Tuned  Secondary

    There has been some dispute as to whether a secondary tuning condenser, connected in the usual way, shown in Fig. 1 and 2, aids in producing a receiver more effective than the simple "broad tuned secondary" circuit. Extended trials have shown beyond a doubt that the sharply tuned circuit of Fig. 1 and 2, when properly adjusted, will give greater selectivity for the same strength signals than the circuit without variable condenser in the secondary. By building the secondaries specially to suit each case, about the same maximum signal strength may be had with both types. With the secondary variable condenser, however, the maximum signals are secured with looser coupling between primary and secondary. As a result, the tuning is sharper and interference is much reduced. The practical actions of such tuners as Fig. 1 and 2 represent should be studied in detail by every operator, and the differences in operation dependent upon closing and opening the condenser "switch" in the secondary circuit should be particularly noted.
    The receiver of Fig. 1 is set up by combining the tuning coils described in the February article with the variable condenser of last month. The resulting arrangement is capable of excellent signal intensity and tuning sharpness. Although it is a directly or conductively connected auto-transformer type of receiver, it has variable coupling between primary and secondary, and is therefore not open to much of the criticism applicable to the ordinary "two-slide tuner" combinations. For the same reason it requires more careful handling in order to produce the best results. In operation of the tuners described, in operating any of the tuners described, it must be borne in mind that the instructions of this article apply solely to the proper manipulation of the inductances and condensers. Before beginning to tune, the operator must have made sure that his connections are properly fastened, that his aerial and ground leads are correctly arranged, that the telephones are in good condition and that the detector is adjusted to a sensitive point. The two last-named items are all-important. Each time the receiver is operated they should be tried out by a "test buzzer" equipment such as described in the December, 1916, article of this series. Once these preliminary steps are gone through, the entire attention may be devoted to tuning. Unless you are certain that the rest of the apparatus is in working order, however, much time and effort will be wasted in trying to build up signals by tuning alone. Fig. 2

Tuning  the  Coupled  Receiver

    The tuner of Fig. 1 has five tuning adjustments. The positions of switches A, B and C govern the wavelength to which the primary or antenna-to-ground circuit is tuned. The third adjustment (switch C) also determines the coupling between primary and secondary. The fourth switch D, and the secondary variable condenser, fix the wavelength to which the secondary is tuned. The more turns of loading coil included in circuit by the switches A and B, and the more turns of transformer coil cut in by switch C, the greater the tuned wavelength of the primary circuit. Also, the more turns cut in by switch C, the closer the coupling between primary and secondary and, consequently, the broader the tuning of the set. The more turns of the transformer coil cut into the secondary circuit by the switch D, and the greater the active capacity of the secondary variable condenser, the longer the tuned wavelength of the secondary circuit.
    Having noted the effects of the various switches and the condenser, as above, and bearing in mind that the object of tuning is to get the primary tuned wavelength and the secondary tuned wavelength to be as nearly as possible equal to the wavelength being received, it is not hard to see how the various elements must be adjusted. There is an additional object, however, which complicates matters a little; that is, the coupling between primary and secondary must be made as loose as possible without sacrificing strength of signals. Since reducing the number of turns cut in by switch C loosens the coupling, it is clear that this switch must be kept as near to zero as possible without weakening the signals too greatly. Fig. 3

First  Operations  in  Tuning

    It is almost impossible to adjust the five variables to their best points simultaneously, so the best plan is to eliminate certain of them from the preliminary operations. In beginning to tune in a signal, therefore, open the switch which thus disconnects the secondary condenser. This gives the secondary a broadly tuned character, and makes it much easier to "pick up" a strange, incoming message. Next, cut in nearly all of the secondary inductance by setting switch D to a high value of turns. Since the secondary condenser is cut out, this has the effect of bringing the secondary circuit more nearly in tune with the usual wavelengths than would be the case if fewer turns were used. Third, set the coupling switch at forty or fifty turns (unless the wavelength you expect to receive is very short, when fewer coupling turns will be used) so as to make the coupling quite close and therefore to aid in hearing signals to which the apparatus is not accurately tuned. Fourth, set the single-turn switch B at the middle of its arc, and finally try to pick up your desired signals by moving the ten-turn step primary switch A. As soon as signals are heard, reduce the number of coupling turns by switch C, at the same time compensating for those cut out of the primary circuit by correspondingly increasing the turns of the loading coil. When the best settings of switches A, B and C are found (that is, the points upon which the signals are loudest), close the secondary condenser switch and set the secondary contact D to the button giving the loudest signal. Then adjust the condenser itself until the signals are loudest; and the receiver will be about at its best adjustment.
    Further improvement in signal intensity can be secured by readjusting the primary single-turn switch, B, after the secondary is sharply tuned. Sometimes it is best to change the position of the coupling-switch C after the secondary condenser is cut in. All this must be determined by trial for each particular station heard. The important things to remember are, to make the rough primary adjustments first, the secondary adjustments second, the final coupling adjustments third, and to try both the primary single-turn switch B and the secondary condenser last, in order to make further improvement in signal strength or tuning sharpness, every time any other adjustment is changed. Fig. 4
    

The  Inductively  Coupled  Receiver

    The circuit of Fig. 2 is the usual inductively coupled type used in most receiving sets. It is entirely satisfactory for general receiving, but must be handled carefully in order to give the best results. Its manipulation corresponds very closely with that just set forth in detail for the auto-transformer tuner. The only radical difference is that coupling is changed by moving the primary coil physically with respect to the secondary. As before, the more turns of primary cut in between the single-turn switch B and the ten-turn switch A, the greater the tuned primary wavelength. Similarly, the more turns of secondary coil cut in by switch D, and the greater the active capacity of the variable condenser, the greater the secondary wavelength. Again, the farther apart the primary and secondary coils, the weaker (looser) the coupling between them and, consequently, the sharper the tuning and the less interference difficulties.
    The objects of tuning are, as before, to adjust the primary and secondary tuned wavelengths to agree with that being received, while at the same time keeping the coupling as loose as possible without destroying the signals. The plan to follow, therefore, is to open the secondary condenser switch and cut in a large portion of the secondary coil as before. Then close the coupling by sliding the coils well together. Finally, search for the desired signals by making rough adjustments of the primary. When signals are heard, open the coupling by sliding the coils farther apart and at the same time adjust both the primary switches to the point where signals are loudest. Then reduce the number of secondary turns and cut in the secondary condenser, setting these to the points which give best signals. For the last adjustment, open the coupling still further and tune the primary and secondary still more accurately. Always bear in mind that the three groups of settings (primary, coupling and secondary) are physically interlinked and that whenever you change any of them you must try the single-turn primary switch and the secondary condenser in order to find out whether you are hearing the loudest possible signals.
    In picking up very weak signals it is sometimes necessary to adjust the large primary and secondary steps together, since in that way the circuits are kept more nearly in tune all the time, and weaker signals can be heard. Under such conditions, it is evident that when few primary turns are in circuit the secondary switch must be near its zero point, and that as the number of primary turns is increased the number of secondary turns must be increased correspondingly. Such detailed handling of the apparatus can only be learned by experience, however, and for cases of this kind these instructions can do little more than suggest a line of action. Fig. 5

Receiving  Long  Waves

    For tuning to very long waves it is sometimes necessary to add loading coils to primary and secondary circuits, as shown in Fig. 3. The addition of active turns to the primary of course increases its wavelength range; a similar effect is had in the secondary. The wavelength of the secondary circuit may be increased by enlarging the size of the secondary variable condenser, and when this is done the secondary loading coil is of course unnecessary. Nevertheless, it is a good plan to keep the inductance of the secondary large, and its condenser correspondingly small; the additional loading coil is therefore the preferred method of extending the tuning range to include the very longest waves. When the main transformer is very small and the received waves are exceeding long, it may not be possible to get close enough coupling unless the primary and secondary loading coils are placed near together so that they act as an additional transformer. When this is done, care must be taken that the direction of connection is correct; otherwise the inductive effect between the loading coils may neutralize instead of helping that between the original primary and secondary coils.
    Another way of getting long wavelengths with a comparatively small receiving transformer is shown in Fig. 4. Here a second variable condenser is shunted across the primary coil terminals, which has somewhat the effect of increasing the capacity of the antenna. This arrangement has a number of advantages, among which is that by its use the fine tuning of the primary can be accomplished by varying the condenser, and that, as a result, it is unnecessary to build the primary coil with a single-turn switch. Further, and especially when very small receiving antennas are used, signals may be louder with this shunt primary condenser than when the same long wave is received by adding the primary loading coil. In Fig. 4 the secondary loading coil is shown, but, as before, its effect of increasing the tuned wavelength may be obtained by enlarging the secondary condenser. Fig. 6

Tuning  to  Short  Waves

    When it is desired to receive wavelengths which are short compared with the fundamental wavelength of the antenna, it is very convenient to insert a variable condenser in series with the aerial connection, as shown in Fig. 5. This has the effect of reducing the size of the receiving antenna, and makes tuning to short waves a very simple matter. As when the primary variable condenser was used in shunt (Fig. 4), small primary inductance steps are not needed, for the sharp tuning may be secured by means of the series condenser. With the arrangement of Fig. 5 it is also possible to tune to wavelengths of the medium range, since the inductance in series may be increased to give the period desired. When used in this way the signals are sometimes weaker than those obtained from the arrangement of Fig. 2, but the tuning is usually sharper, because the ratio of inductance to capacity is increased.
    The primary circuit is shown in Fig. 6, which may be arranged with two single-pole double-throw knife switches so as to connect the variable condenser in series or in shunt with the primary, or to cut it out altogether. With both switchblades at the left, in positions W and Y, the condenser is in series, as shown in Fig. 5. When both switchblades are thrown to the right, in positions X and Z, the condenser is in shunt to the primary coil, as in Fig. 4. When the upper switchblade is at the right and the lower at the left, i.e., in positions X and Y, the variable condenser is cut out and the circuits are connected as shown in Fig. 2. This same switching arrangement may also be used with the arrangement of Fig. 1. Practically the same results will be secured, except that the primary turns are changed whenever the switch C (Fig. 1) is moved to vary the coupling.
    Experienced radio operators, as well as beginners, will find it worth while to study the principles explained in this article. They are the basis of successful operation of the coupled receiving sets.
(The  end)