UHF technique in 1927

By H. E. Hollmann
Physical Institute of Technology, Darmstadt, Germany.
From 'Radio News'.

75 cm transmitter

The telescoping horisontal rods A form the only aerial of the 75 centimeter radiophone transmitter. They are slid up and down the vertical mast to decrease or increase the length of the parallel wires included in the oscillatory circuit of the tube V1, and connecting the latter to the batteries.

"Low wavelengths" and "high frequencies" are radio terms which may be considered as entirely relative, as they have no particular significance of their own. Several years ago all channels below 600 meters were classed as "short waves"; and, when the American radio law of 1912 was passed, the legislators thought that the amateur transmitter could be eliminated by assigning stations of this class to wavelengths below 200 meters, as this field was considered valueless. When broadcast stations started operating on waves near the amateur band, the amateur was forced to use still shorter wavelengths; and at that time it was discovered that satisfactory results could be obtained on less than 100 meters. Later amateurs found that waves as low as 40 meters provide valuable channels for communication. Today there are many high-power commercial stations operating on wavelengths which were considered entirely impractical less than six years ago. It has been found that the very short waves possess many unusual characteristics which make them ideal for certain types of communication, particularly over long distances. In addition to the stations which are actually operating on short waves, there are hundreds of experimenters endeavoring to obtain data on the ultra-short wavelengths which have not yet shown their commercial value. While engineers are busy trying to find some commercial use for the wavelengths below 13 meters or so, scientists are still seeking shorter wavelengths to conquer. In this field the latest development is a new transmitter which operates on a wavelength of 75 centimeters (three-quarters of a meter, or 29½ inches. Of course, no practical use has yet been found for such a transmitter, but the inventors of the system are very much pleased with the fact that they have been able to produce oscillations at such high frequencies, and are equally elated over the development of a receiver which will detect the oscillations. Not only have they been able to transmit and detect continuous wave signals on a wavelength of 75 centimeters, but they have been able to send and receive I.C.W. (interrupted continuous wave) and even telephone signals on these high frequencies. These signals have been received over distances of approximately a quarter of a mile.

In this article no attempt will be made to give constructional details of a receiver or transmitter for operation on 75 centimeters, but the experiments with this apparatus which have been conducted in Germany will be described for the benefit of the qualified experimenter who may wish to do some original research work with ultra-short wavelengths. However, it must be remembered that the results described were obtained with foreign equipment and, therefore, entirely different results may be obtained from American tubes and apparatus. Two German radio engineers, Barkhausen and Kurz, are responsible for the first research work attempted in this direction. The discovery which pointed to the possibility of operating radio apparatus on wavelengths of less than one meter came as a result of a study of the characteristic curves of vacuum tubes. It was found that in these graphs certain irregularities appear, which are similar to the resonance curves obtained in high-frequency phenomena. In the experiments which followed this discovery it was proved that high-frequency oscillations can be produced by applying a high positive potential to the grid of a vacuum tube and a negative potential to the plate. Measurements were made which showed that the oscillations which were produced in this way had a frequency greater than one

Resonant circuit

The 75-cm resonant circuit comprises only the parallel wiresof length (l) varied by the sliding bridge B, and the tube's internal capacity.

hundred million cycles per second, corresponding to a wavelength less than three meters. Barkhausen and Kurz explain the results described above as follows: the electrons emitted from the incandescent filament are so strongly attracted by the positive grid that they move at high speed across the space between the filament and the grid, and into the space between the grid and the plate. As the electrons approach the plate, they are strongly repelled by the negative potential and, as a result, are driven back toward the grid. In this way a pendulumlike movement is established around the grid; or in other words, electrical oscillations are produced. The frequency of the oscillations has been found to be governed by the characteristics of the vacuum tube and the value of the plate and grid potentials. The oscillations are accelerated by an increase in the potential producing them, and vice versa.

The research of Barkhausen and Kurz has been confirmed by Scheibe, another wellknown German scientist, who has in addition developed an exact formula by which the frequency of the oscillations may be determined in every case. With the method described above, Scheibe found the oscillations very feeble, and only with the most sensitive indicator was he able to detect their existence. However, he found that by connecting a closed oscillatory circuit to the tube it is possible to obtain a considerable increase of energy. The circuit which he used for this purpose is shown in Fig. 1. Two parallel wires (P and P1) are connected to the grid and plate terminals of the tube, and these wires are connected by a sliding bridge, B. The grid-plate capacity of the vacuum tube, plus the inductance of the two parallel wires, composes the closed oscillatory circuit, which may be adjusted to the desired frequency by the sliding bridge. The grid and plate potentials are brought to the tube through the parallel wires and the by-pass condenser (C) is used to prevent a short- circuit of these power-supply wires. Because of its relatively large capacity, the presence of the condenser in the bridge has little or no effect upon the high-frequency circuit.

The operation of such a circuit as shown in Fig. 1 may be explained as follows: with a negative potential applied to the plate, under ordinary circumstances, it would seem that there should be no current between the plate and the filament. On the contrary, in a poor vacuum tube it is possible to detect an "ionic" current flowing in the opposite direction from that of the normal stream of electrons. If oscillations are now produced in the tube, in the manner previously described, a positive stream of "ions" will be found to exist; and this stream will reach its maximum when resonance is produced between the outer oscillatory circuit and the oscillations inside the tube. The ion represents a particle of gas which has lost an electron (unit negative charge) and therefore has a unit positive charge; it tends to move toward a negative potential, therefore. With a milliammeter (Ip) connected in the plate circuit, it is possible, by observing the current registered on the scale, to determine with certainty the best position for the sliding bridge. It is possible also to measure the frequency of the oscillations. Fig. 2 shows that the plate current of the tube (the solid line of the graph) varies with the length of the wire in the bridge circuit. The dotted line of the graph indicates how the R.F. current of the closed oscillatory circuit varies with the length of wire in the bridge. This measurement was obtained by connecting a thermo-coupled milliammeter in series with the condenser (C) and in this way the high-frequency current in the bridge circuit was directly measured. By comparing the two curves of the graph, it may be seen that the maximum intensity of oscillations is approximately at the point of maximum plate current. Therefore, by adjusting the bridge until maximum plate current is obtained, it is possible to determine the frequency of the oscillations by measuring the length of wire in the bridge; and, at the same time, be assure of maximum intensity of oscillations. The curves shown were obtained with a Schott transmitting tube (type N) under a plate potential of 240 volts.

In order to convert the oscillator described above into a short-wave transmitter, it is necessary only to connect a bi-polar antenna to the bridge in the proper manner (see Fig. A). Such a transmitter will send out a considerable amount of energy, and, with proper modulation, will serve for radio telephony. The bi-polar antenna consists of two straight wires (each half as long as the wave produced), which are fastened directly to the bridge in a horizontal position, in the manner shown in the pictures accompanying this article. In Fig. 3 the complete schematic wiring diagram of a transmitter of this type is given. The bipolar antenna is shown at A and is connected directly to the bridge C. The two inductors in the filament circuit are choke coils which prevent undesired self-oscillations.

Development of apparatus for receiving signals from transmitters operating on a wavelength of less than one meter presented another problem which scientists did not find easy to solve. With signals having a frequency of approximately 400,000,000 cycles per second, it may be seen that the usual method used for receiving continuous-wave signals {i.e., the audio beat-note system) would be entirely impractical; as

it is almost impossible to adjust the frequency of the receiver so exactly that the oscillations received from the transmitter, when "beating" with the locally-generated oscillations, would produce a steady audio note. Therefore, another system was found necessary and I. C. W. and phone were tried; both of these proved satisfactory. In their original experiments Barkhausen and Kurz endeavored to modulate the output of the set with a buzzer connected in the plate circuit of the tube. However, this proved entirely unsatisfactory because, when the plate circuit was opened, the grid received the entire emission current and, as a result, quickly overheated. In Fig. 3 the buzzer (B) is shown connected directly in the grid circuit and this has been found to be very efficient. Barkhausen and Kurz also attempted to modulate their transmitter for telephone, but the results have been more or less unsatisfactory. A modulation transformer was used to replace the buzzer in the plate supply wire, and a microphone was connected in series with the primary winding. This method, however, modulates only a small percentage of the output.

The only experiment which has given satisfactory results employed a circuit such as is shown in Fig. 3, in which an extra tube is used as a modulator. In this circuit the fluctuating current of the microphone (M) is carried through the microphone transformer (T) to the grid of the modulating tube (V) and this changes the internal resistance of the tube so that low frequency potential variations exist in the choke coil (L). As a result, the grid potential of the transmitting tube (V1) is increased and diminished in accordance with the voice frequency. Figure A gives a complete view of the radio-telephone transmitter which has been designed to operate on a wavelength of 75 centimeters. The two parallel wires (P and P1 in the diagram) are supported by the wooden pole and the oscillator tube V1 is mounted on the top. The sliding bridge and the bipolar antenna, (A-C) which are adjusted to tune the closed oscillatory circuit to resonance with the transmitted frequency, are provided with a telescoping brass rod. B is the buzzer which is used to modulate the system for telegraphy, and the telephone modulation equipment is located to the right. The designating letters of all parts correspond with those in the diagram (Fig. 3.) The 120 storage cells located under the table are used for providing the grid potential to the oscillator tube.

The apparatus used for receiving signals on wavelengths of 75 centimeters is shown in Fig. B. The construction of the receiver is very similar to the transmitter, although the circuit is somewhat different. The long wooden rod supports a crystal detector in an insulated tube (D) at the top, and the two parallel wires are carried up the sides of the rod. The bi-polar antenna (A and Ai) and the sliding bridge are also mounted in much the same manner. The wires to the headphones go up the inside of the rod and the escape of R.F. currents is prevented by small choke coils (L). If, in place of the headphones, an extremely sensitive measuring instrument is connected to the detector circuit of the receiver, the transmission of short waves through space can be observed and investigated. Here we must refer to an observed phenomenon; if the waves from the transmitter or the receiver are thrown back or reflected by a metal plate, they will then interfere with the direct transmission and, in this way, there are produced "standing waves" which can be recognized by the rising or falling of a galvanometer needle. The apparatus is so sensitive that waves reflected at a distance of several meters can be observed. The intensity of reception is at maximum when the transmitting and the receiving antennas are parallel. If one of them is turned through 90 degrees, the reception disappears completely, and then by further turning is brought again to maximum. Plotting the galvanometer, changes with polar coordinates produces a graph for a bi-polar antenna similar to that shown in Fig. 4.

For reception over greater distances, we have to design an efficient vacuum-tube receiver. After various experiments the circuit shown in Fig. 5 was chosen, and this has given very good results. V is a vacuum tube, with a grid condenser (C) and a grid leak (R) of one megohm. The oscillatory circuit is formed just as it is in the transmitting circuit, with two parallel wires on which slides the bridge extending to the antenna. Now, as only the plate potential is carried over the parallel-wire

system, the condenser portion of the bridge is superfluous. The tuning in of the signal from the transmitter must be accomplished by sliding the bridge and by lengthening or shortening the antenna. The tuning of the oscillatory circuit with these methods is extremely difficult and the highest sensitivity is finally obtained by variation of the plate potential. Fig. C shows the receiver as it appears completed. As in the detector circuit previously described, the high-frequency section must be withdrawn as far as possible from the disturbing influence of the ground, and supported on the vertical insulating tube. The circuit constants with ultra-short waves are very small; so that the entire length of the system of parallel wires is only about 50 millimeters (two inches). The connections to the battery lead are brought down through the rear of the tube. To increase the amplification of the receiver, a Loewe triple tube (three sets of elements in one glass bulb) is connected in an audio amplifier circuit. With this apparatus, it is possible to demonstrate to a large number of listeners the phenomena of fixed waves, interference, reflection, etc., by means of a loud speaker. The experimenters have covered a distance of about 300 meters; but the intensity of the sound was still so great that the limit has been by no means reached.

UHF technique anno 1927.

Remarkable Work of German Experimenters With Radiophone on Ultra-High Frequencies Develops Interesting Circuits