Radio broadcast .. (1922-30)

24 RADIO BROADCAST NOVEMBER, 1927 limitation in common: They can transmit only one shade and one unit area of the picture at a given instant and therefore transmission must be accomplished by dividing up the subject to be transmitted into thousands of small areas. The " Rayfoto" and many other systems trans- mit the signals for each unit area in rapid suc- cession and the resultant signal varies in ampli- tude in accordance with the shading of the pic- ture. The speed at which the impulses are trans- mitted depends largely upon the ability of the receiving apparatus to reproduce rapidly the electrical impulses on the recording medium. The corona method of printing used by the Cooley "Rayfoto" system is capable of printing faster than any other system the author knows of, but for simplicity and low first cost we are using a signal frequency of about 800 cycles per second, which does not permit printing as rapidly as is possible with the system when higher fre- quencies are used. The possibility of operation at higher frequencies has been taken into consider- ation in designing the present equipment so that the speed of transmission can gradually be in- creased without necessitating any radical changes in equipment. As explained in the October issue of RADIO BROADCAST, the picture or subject to be trans- mitted is placed, at the transmitting station, upon the drum of the picture transmitter, which we will hereafter call the "convenor." A small spot of the picture is illuminated and the re- flected light from this spot actuates a photo- electric cell, the signals from which control the radio transmitter after the photo-electric cell currents have been sufficiently amplified. Each time the drum is revolved, the spot of light traverses a different path an eightieth of an inch wide across the picture. The line is broken up into 480 sections by the optical system so that 480 electrical impulses are transmitted every revolution of the drum and each impulse cor- responds in intensity to the reflected light from a small area of the picture. The result is that 480 electrical impulses are transmitted for each revolution of the drum, or about 800 per second when the drum is making one hundred revolu- tions per minute. Running at this speed the drum feeds along the shaft 1, Fig. i, at the rate of one and a quarter inches per minute. The drum is two inches in diameter and about five inches long. This will give us an operating speed of four minutes for a five-by-six-inch picture. The beginning of each revolution is marked by an impulse made up of twenty strong 800- cycle signals in succession. This impulse is used at the receiver to start the recording drum off at exactly the same time as the transmitter drum, for it is necessary that the two drums start off together. To accomplish synchronism in this way, known as the "stop-start" method, the recording drum must start a revolution at the same instant as the transmitter drum. It is neces- sary that the recorder drum run slightly faster than the converter drum, then stop at the end of the revolution for an instant until the con- verter drum completes its revolution. A trip magnet operated by the strong synchronizing impulse releases the recording drum at the proper time. This trip magnet is operated through a relay which is connected to the rest of the system only after the revolution of the recording drum has been completed. Between the time the recorder drum stops and the time the synchronizing im- pulse is received, there must be no strong signals received, so we paste a strip of white paper at the end of the picture being transmitted so the signals will be weak while the recorder drum is stopped. Should a crash of static or some other disturbance be received during this waiting period, or "recorder lap," as it is called, the re- corder drum will be released in advance of the synchronizing signal. By making the recorder lap very small, the danger of such a " static slip" will be reduced proportionately. The wider the white strip on the picture being transmitted, the greater will be the chances of a good start after a static slip so that the only marring effect will be one line slightly out of place. RADIO BROADCAST Photograph SOME EARLY "RAYFOTO" EQUIPMENT IN RADIO BROADCAST LABORATORY The apparatus on the right of this picture is a Cooley photograph transmitter and in the center is an amplifier and "corona" apparatus. The picture receiver at the left has been redesigned in many ways to make its operation as simple as possible. This apparatus was photographed four years ago We will consider here a few of the principles involved that affect the characteristics of the received picture. In picture work, we wish to reproduce at the recorder shades of light and dark corresponding exactly to those of the transmitted subject. The light reflected from the subject varies the current through the photo- electric cell in a ratio almost directly with the intensity of the light, and this current, after amplification, is made to control the power input to the radio transmitter modulator which there- fore varies directly proportionately to the re- flected light, due to the characteristics of the Heising modulator. The final modulated radio signal sent out over the air will vary as the square root of the reflected light. The received signal is am- plified lineally in the radio-frequency stages of the receiver. The detector output varies as the input squared, however, and therefore the cur- rent in the plate circuit of the detector will be directly proportional to the reflected light. The signal then can be amplified in the audio ampli- fier and delivered to the " Rayfoto" printer with an intensity directly proportional to the reflected light at the transmitter. Limited by the data available at the present time, this is as far as we can go with the signal and know definitely what we are doing in the way of maintaining the proper signal ratio through the various circuits. We have no exact data on the relation of the input to output of the Cooley " Rayfoto" printer. Also we do not know the relation of the power delivered by the " Rayfoto" printer to the effect it has on the re- ceiving paper. This factor is quite flexible and can be controlled considerably by the selection of the printing paper and its time of develop- ment in the photographic solutions. The printing paper we recommend today probably will not be the paper you will be using next year. It is therefore necessary to have some control over the system so we may match our amplification characteristics to conform with those of the re- cording paper we may choose to use. For ex- ample, if the received picture does not show sufficient contrast in the lighter shades but too much in the darker shades, we must adjust our amplification characteristics to correct for this. One way it can be done is to reduce the filament voltage on one of the amplifier tubes so that the strong signals are cut off somewhat by running over the top knee of the characteristic curve while the signals of lower value are on the straight portion of the curve. Additional correction may be obtained by reducing the time of development in the photographic solution. The most convenient place for signal charac- teristic control is in the detector circuit, because of its "squared" characteristic. This character- istic may be varied considerably by proper pro- portioning of the grid condenser and grid leak. If the grid leak can be brought down to a very low resistance, say 500 ohms, and the plate voltage made adjustable over a range of from 4 to 40 volts, additional control of considerable value will be gained. Instead of varying the plate voltage, a variable grid battery may be used. EFFICIENT AUDIO STAGES NECESSARY A GOOD picture must not only represent exact shadings of the subject but it must also show up most of the small details of the original. A poor audio amplifier system will blur up the details in black shades and will not permit any of the details in the light shades to appear. The amplifier must not oscillate at any audio or super-audio frequency or even tend to oscillate. Oscillations in audio amplifiers most generally occur because of feed-back through the B bat- teries from one stage to another and can be pre-