Radio Broadcast (May 1929-Apr 1930)

RADIO BROADCAST No. 301 Radio Broadcast Laboratory Information Sheet October, 1929 Output Transformer Ratios /"lUTPUT transformer ratios required for ^ use with cone or horn type loud speaker are shown on the chart on Sheet No. 302. This chart, reproduced from a recent bulletin of Ferranti, Inc., covers the range of loud speaker imped'ances from 0 to 6500 ohms and tube impedances from 0 to 15,000 ohms. On Sheet No. 297 in the Sept., 1929, issue, a chart was given of output transformer ratios for use with moving-coil type loud speakers. In the case of horn and cone type loud speakers the bulletin suggests that the impedance used in determining the transformer ratio should be the impedance of the loud speaker at about 200 cycles. The impedance at this frequency may with fair accuracy be taken as 2.5 times the d.c. resistance of the winding. That is, a loud speaker with a d.c. resistance of 2000 ohms has an impedance at 200 cycles or approximately 2000 times 2.5 or 5000 ohms. Owing to the fact that the impedance of cone type loud speakers fre / quency varies from 1000 to 30,000 ohms between 200 and 5000 cycles, ideal results are not possible but transformer ratios determined as indicated above will give best results. As an example of the manner in which the impedance of a magnetic loud speaker varies we show on this sheet an impedance curve the Western Electric 540 AW. An example will help to make perfectly clear the use of the chart on sheet No. 302. Example 1. A loud speaker has an impedance of 4000 ohms at 200 cycles. It is to be used in a push-pull circuit using two tubes each with a plate impedance of 5000 ohms. What transformer ratio is required? Two 5000-ohm tubes in pushpull give a total impedance of 10,000 ohms. Referring to the chart we find the horizontal line corresponding to a loud speaker of 4000 ohms impedance intersects the vertical line corresponding to a tube impedance of 10,000 ohms at a point corresponding to a transformer ratio of 1.6 to 1. No. 302 Radio Broadcast Laboratory Information Sheet October, 1929 Output Transformer Ratios 6500 6000 5500 5000 4500 4000 3500 3000 2500 2000 1500 10C0 500 0 y 2000 4000 6000 8000 10,000 TOTAL PLATE IMPEDANCE 12,000 14,000 No. 303 Radio Broadcast Laboratory Information Sheet October, 1929 "Power" and "Linear" Detection Explained T)OWER detection" and "linear detection" ' are two terms frequently used to describe certain characteristics of new receivers. On this sheet we explain briefly what these two terms mean. A power detector (according to the popular definition) is one* operated at fairly high values of input voltage. Power detectors are frequently followed by a single stage of audio frequency amplification because sufficient a.f. voltage can be obtained from them to load up a power tube. It should be realized, however, that power detectors are not always followed by single audio stages but may be followed by a two-stage audio amplifier. The important factor is the signal level at which they operate. Accurately defined, however, the power detector is one from which sufficient power may be obtained to operate a loud speaker directly. Ordinary detectors operate on what is called a "square law," that is the a.f. output voltage is proportional to the square of the r f . input voltage. Such detectors produce some distortion especially if the r.f. input is modulated at fairly high percentages. The distortion reaches a maximum of 25 per cent, when the input signal is modulated 100 per cent. The present tendency in broadcasting is to increase the modulation to 100 per cent, so as to utilize as completely as possible the output of the transmitter. Linear detectors will produce very little distortion on 100 per cent, modulated signals and it is for this reason that this type of detector is increasing in use. A detector is "linear" when its a.f. output is directly proportional to the r.f. input. Such detectors produce a distortion of about 10 per cent., with 100 per cent, modulated signals in comparison with the 25 per cent, distortion produced by square-law detectors. The decreased distortion (lrom 25 to 10 per cent. ) due to the linear detector is readily noticeable to the ear. It should be pointed out that the fact that the detector operates at high signal levels does not necessarily mean that it is linear. C-bias detectors are linear over only a small portion of their operating characteristic. STRAYS FROM THE LABORATORY {Continued from page 3k9) the courage to make this plain and straightforward statement in the face of what is apparently a distinctly different tendency. "For the past eight years the writer has had charge of the radio market news broadcasting for the United States Department of Agriculture and I feel that anything said in the interest of the farmers by technical radio men indicates an appreciation of a situation with which we have been dealing for a long time. "At the present time there are more than a hundred radio stations broadcasting crop and market reports for the benefit of farmers. This information is handed to the radio stations by the branch office representatives of this Bureau located in different parts of the country. All of these offices are connected by leased wire and all of the wire lines lead into the Washington office, so that every branch office is in touch with every other branch office and with headquarters. In a sense our service is that of chain broadcasting although the information is not poured into a single microphone but the information is disseminated over our leased wires and broadcast by each station. " In order to avail himself of this vital market information, upon which the success of his farming largely depends the farmer must have a radio set that is selective enough to make certain selections among radio stations and sensitive enough to bring in clearly stations during the daytime that may be 100 or 150 miles away. "Although the aim of the electric power interest is to provide the farmer with electric light and power as rapidly as good business will permit, nevertheless there is a potential market of nearly six million homes not provided with the facilities for using modern electric sets. "In the last eight years this office has tried out a great many types of battery operated radio receiving sets and I believe that if a three or four-tube set capable of meeting the requirements which I have briefly sketched could be devised there would be a very good market for it." GRID LEAK VS. BIAS DETECTION (Continued from page 352) with very small signals. The lower plate voltages give somewhat better smallsignal sensitivity. The sensitivity of the c-324 bias detector is slightly less than the sensitivity of the c-327 grid-leak detector. Fig. 6 shows that the sensitivity of c-324 bias detector exceeds the sensitivity of the c-327 grid-leak detector when the effective amplification in the preceding radio-frequency stage is included. In conclusion the writer wishes to thank D. F. Schmitfor helpful comments and suggestions and to acknowledge the assistance of members of the Cmiriingham Laboratory. Bibliography 1 — F. B. Llewellyn — "Operation of Thermionic Vacuum Tube Circuits," Bell System Tech. Jour., 5, 433; July, 1926. . 2 — E. L. Chaffee and G. IL Bbowning — "A Theoretical and Experimental Investigation of Detection for Small Signals," Proc. I. R. E. 15, 113; February, 1927. 3 — F. M. Colebrook — "The Rectification of Small Radio-Frequency Potential Differences by means of Trade Valves," Experimental Wireless, 2, 865; 1925. 4 — Stuart Rallantine — "Detection by Grid Rectification with High-Vacuum Triodes," Proc. I. R. E., 16, 593; May, 1928. 5 — L. P. Smith — "Theory of Detection in a HighVacuum Thermionic Tube," Proc. I. R. E., 14, 649; October, 1926. 6 — J. R. Nelson — "Detection with the FourElement Tube," Proc. I. R. E., 16, 822; June, 1928. 364 • • O C TOIIER 1 9 2 9 •