Radio broadcast .. (1922-30)

KADIO BROADCAST. No. 25 Radio Broadcast's Home-Study Sheets BUZZERS IN RADIO EXPERIMENTS July, 1929 A GREAT DEAL may be learned from a buzzer, although many never think of it in this connection. There are several ways in which a buz/er may be used to generate radio waves. In Fig. 1, a few turns of wire and a large con- denser of one or two microfarads are con- nected across the contact points. When the contacts are separated, the large condenser is charged by the battery, and when they are closed it suddenly discharges through the few turns of wire. On account of the very high ratio of capacity to inductance, this discharge is highly damped and probably results in only two or three oscillations. This sudden impulse, however, starts oscillations in the secondary circuit, which continue for some time as they are but slightly damped, due to the fact that this circuit has plenty of inductance and very little capacity. The wavelength of the result- ing radio currents in the secondary, circuit will, accordingly, depend on the values of L and C, and may be determined from the equation \\.ivHength = 1884) L X C where L — induc- tance in microhenries and C = capacity in mfd. Another Method In Fig. 2 another method of generating radio waves is illustrated. The connections are the same, except that the condenser is smaller and the inductance larger, which reduces the damp- ing effect. High-frequency currents of consider- able strength may be generated in this manner by applying several cells to the buzzer, and it has the merit of not seriously affecting other portions of the circuit by inducing stray cur- rents therein. Probably the most satisfactory circuit for general laboratory work is that shown in Fig. 3. When the contacts are closed, the coil, L, is surrounded by a magnetic field due to the Fig. 1 L i O o o Fig. 2 direct current flowing therein. When the con- tacts are open, this field collapses rapidly and thereby induces a voltage in the coil, which charges the condenser, C. When the magnetic field has ceased to generate further voltage, the condenser discharges through the coil, and so on until the resistance of the circuit com- pletely dampens the oscillations. At the next movement of the vibrator, the process is re- peated. The buzzer simply functions as an electrical hammer that applies a blow to an electrical weight, at regular intervals, the weight con- tinuing to oscillate at its own natural period between the blows. If the value of L and C were such as to have a wavelength of 300 meters, and the buzzer had a pitch of 1000 per second, there would be time for 1000 radio oscillations for each oscillation of the buzzer. By means of the buzzer and the inductance and capacity standards referred to in "Home- Study Sheets," Nos. 20 and 24, we are now in a position to set up a radio current of any desired wavelength, but before undertaking this it is desirable to consider the mechanical features of buzzers, and also to provide some additional measuring apparatus. Unsatisfactory Types There are several high-pitched buzzers on the market that will answer very well in this connection. These are made specially for radio work, such as code instruction and testing crystals. The ordinary house buzzer is of too low a pitch to be satisfactory. Also, the kind that is connected up as indicated in Fig. 4 should be avoided, for the reason that when the contacts are open there is still a fairly large capacity across them due to the fact that one of the binding posts is in contact with the case and the magnet cores, while the other is con- nected to the surrounding coils. Sometimes buz- zers are provided with a resistor across the Fig. 3 Fig. 4 contacts to prevent sparking. Such a resistor would have to be removed to make the buzzer suitable for radio work. Making a Buzzer It is quite surprising to learn what a satis- factory radio current may be generated by a single dry-cell battery and a very small buzzer. Suitable ones of high pitch are not very diffi- cult to make, and they lend themselves admi- rably to the construction of bridges, wave- meters, etc. Fig. 5 illustrates the general plan of construction, and to what extent the di- mensions may be reduced. Buzzers of this type may be made that will vibrate as high as 5000 times a second. The magnet core is simply a piece of soft iron wire about an eighth of an inch in diameter or a bundle of soft iron wire (stove-pipe wire will do) of the same size. In making the hair-pin turn, care will be neces- sary to keep the two legs straight and parallel. On a straight piece of the same material a thin paper tube is formed by wrapping it with a strip of paper to which glue has been applied. When the tube has hardened, two suitable lengths are cut, and each is wound with num- ber 32 or 34 enameled wire. Shellac must be applied during the winding, and if consider- able tension is used in placing the turns, each successive layer may be stepped back half a space, as indicated in the drawing. In this manner the necessity of making two very small bobbins is obviated. When the shellac has hardened thoroughly, the coils are then pushed onto the iron core and the connection between is soldered. The vibrator is a one-eighth inch strip of the thinnest commercial tin, from which the tin coating has been polished off. While this may be supported by soldering it to one of the magnet poles, it is preferable to provide a sep- arate brass block. This should be heated in a clean flame just enough to permit the tinning of the space where the vibrator is to be ap- plied. The latter may then be coated with flux, and steadily held in position until cool. In this way there should be no danger of drawing the temper of the vibrator. Considerable care will be necessary in provid- ing the contacts. A small piece of platinum wire should be procured from the jeweler, say about an inch of No. 30 guage. Hammer out one end to form a little spring, not over three- sixteenths of an inch long. Using a piece of heavy tinned copper wire as a soldering iron, attach the platinum to the vibrator with the least amount of solder possible. When the contact screw has been provided, a short length of the platinum wire is soldered on the end, care being taken to remove all surplus solder. The base should be cut from a piece of panel material, to which the magnet is secured by means of a thin brass yoke held down by two 4-36 screws. That such a buzzer is working properly can- not be determined by its sound, although it is essential that the note be clear, musical, and steady. It is only when the resulting radio currents are detected with a crystal and are listened to in a telephone receiver that the final result can be ascertained. Not infrequently the strength and quality of the note can be improved materially by placing a rather large fixed condenser across the magnet coils. Determining Frequency There is a simple way of determining with a fair degree of accuracy the frequency of high- pitched buzzers. It is also interesting and instructive from the standpoint of radio. In Fig. 6, L is a 1500-turn duo-lateral or honey- comb coil, the inductance and distributed ca- pacity of which we will accept from the manu- facturer's statement. Li is 8 or 10 turns of wire wound on a wooden core and slipped inside of the large coil, L. C is a calibrated condenser reading up to 0.001 mfd. and to one side is connected a crystal detector with a telephone receiver across it. If the buzzer is now started, the sound should be heard clearly in the re- ceiver, and, as the condenser C is varied, the character and intensity of the note will pass through a series of changes, alternately becom- ing clear and loud, and then indistinct and confused. The points at which these changes 1500 Turns 0.001 mfd. 1 mfd. Fig. 5 Fig. 6 occur will be found close together at the lower dial readings and further apart toward the up- per end of the scale. In the circuit with which we have been ex- perimenting each impulse of the buzzer starts a series of oscillations in the circuit L C, and the frequency of these oscillations is given by the formula f =s 15920 ° " ^ITc When the note in the telephone receiver is strong and clear, the frequency of the radio oscillations is an exact multiple of the fre- quency of the buzzer. To illustrate by figures, assume that at 100 on the condenser dial, C indi- cates a capacity of 0.001 mfd., and that the lower readings are strictly proportional, as would be the case if the condenser capacity curve were truly a straight line. Let it be further assumed that the coil, L, has an induc- tance of 175,000 microhenries. With these values a critical point might be noted at, say, number 23.1 on the condenser dial, a second one at 'M\, a third one at 41.1, and so on. The corresponding capacities would be 0.000231, 0.000315, and 0.000411 mfd. Multiplying these by 175,000 we find the value of L C, and the corresponding frequencies may then be found from the formula or by consulting the wave- length and frequency table presented in "Home- Study Sheet" No. 19 (April RADIO BROAD- CAST). In the present case, these will be found to be frequencies of approximately 12,000, 14,000, and 16,000 cycles per second. The uni- form difference of 2000 is the frequency of the buzzer. In actual practice, depending on the values involved, five or ten critical points may be identified, in which case it is necessary to divide only the difference between the first and last- radio frequencies by the number of steps in- volved, which will, of course, be one less than the number of readings taken. It is also neces- sary, if accuracy is desired, to add to the ca- pacity of the condenser at each reading the distributed capacity of the coil L, which will be of the order of .00003 mfd. When a radio-frequency galvanometer, to be described later, has been made, it may be inserted directly in the oscillatory circuit L C, and the crystal and receiver may be dispensed with. The experiment is more striking when carried out in this manner, as the galvanometer shows a marked increase in deflection whenever a critical point is passed. JULY -1929 169