Radio Broadcast (May 1928-Apr 1929)

Data on Filament-Type Tubes CHARACTERISTICS OF POWER REOTFIERS THE first socket-power device sold in any quantity was the "B eliminator" which at the time these devices were first offered, some three years ago, commonly utilized the general-purpose tube, type cx301a, as the rectifier. Some of these were single-wave rectifiers, capable of supplying 10 to 15 mA., at 90 volts, while others were full-wave rectifiers using two tubes and with an output of 20 to 30 mA. The outputs mentioned were obtainable only with tubes in which the filament was in the best of condition, and the slightest drop in filament emission became noticeable immediately in a reduction in output current and voltage. These "B eliminators" gave good service at the time they were introduced, the conditions differing considerably from present-day requirements in several important respects, the most important one being the plate current drain of the sets then in use. Fivetube sets were just becoming popular, with six-, seven-, and eight-tube sets quite rare. The average receiver was of the three or four-tube type, requiring from 10 to 20 milhamperes, at a maximum potential of 90 volts. The rapid increase in the number of tubes per receiver soon raised the maximum demand to 30 mA. in the case of the five-tube receiver in which the C battery was omitted altogether. Two years ago, power-amplifier tubes were introduced, the 371-type tube adding nearly 15 mA., to the total receiver current drain when operated 135 volts, and 20 mA. when operated at 180 volts. As a result of this increase the maximum current requirements became 50 to 60 mA. Other developments were the type of receiver in which the filaments of cx299 tubes were operated in series from socket power, requiring 65 to 75 mA. and those using cx-301a in series and requiring 250 to 300 mA. The rectifier using the cx-301a, soon became inadequate in the face of the rapidly increasing current demand, and the cx-313, a full-wave rectifier, and cx-316b, a half-wave rectifier, were introduced, the cx-313 providing an output of 65 mA., the maximum allowable transformer voltage being 220 volts a.c. per anode. The cx-316b also has a rating of 65 mA., but the design is such as to permit the use of higher transformer potentials, 550 volts. With an efficient filter the cx-313 provided a maximum voltage of 180 volts, and the cx-316b, 450 to 500 volts. These types are now being superseded by the cx-380 (fullwave) rated at 125 mA. and 350 volts a.c. transformer voltage per anode and the cx-381 i 3 (One I ilamentE ndOne inode) / / / , T13 1/ 1/ 1/ rCX30 -A til Ef = 5.C Volts By ROGER WISE Formerly, Chief Engineer, E. T. Cunningham, Inc. The author of this article on the whys and wherefores of filament-type rectifier tubes and circuits was, until recently, Chief Engineer of E. T. Cunningham Inc. He is now associated with Grigsby-Grunow and actively engaged in putting into shape a tube manufacturing plant that in size will rival the largest now in existence. Mr. Wise points out the development of the present rectifier tubes of the 280 and 281 types. He shows that with certain types of filter circuits the instantaneous currents that must be supplied by the filament of the tube are much too great for the tube to have long life, and suggests another filter arrangement which increases tube life. It has been the experience of the Laboratory Staff that the suggested filter arrangement, whereby the first filter condenser is eliminated, produces too much hum for use with good amplifiers and good loud speakers. The Staff's lack of success may be accidental, and should be glad to hear from readers who experiment with these two filter systems. — The Editor. 'GRID* PLATE) VOLTAGE Fig. 2 Fig. 1 (half-wave) rated at 110 mA., and 700 volts a.c. The latter two tubes have been in production only a short time, and the data presented in connection with them is subject to slight modifications. The circuit diagram of one of the early types of rectifiers designed to use cx-301a tubes, is shown in Fig. 1. This circuit diagram also shows the placing of the oscillograph vibrators used in testing the performance of this unit as shown in the following figures. Vibrator V3 is arranged to show the transformer secondary voltage, which, with 110 volts applied to the primary, was 230 volts r.m.s., or 115 volts per tube, the applied voltage to each rectifier being one half of the total transformer voltage because of the full-wave connection. Vibrator Vi reads the instantaneous current through the tube, while V2 by means of a special circuit arrangement reads the voltage across the tube only during the time it is conducting current. The back voltage across each tube is blocked off by means of the auxiliary full-wave rectifier tube, T3 in the diagram. With this arrangement it is possible to get a complete record of the performance of each rectifier and to determine the exact point at which overloading, due to limited filament emission, begins. The power lost in the tube can also be computed, and the efficiency determined. In taking the records, high sensitivity vibrator elements requiring about one milliampere per millimeter deflection, were used and a 10-millimeter deflection was satisfactory for most of the work. The current required, 10 mA., could not be disregarded in taking some of the voltage readings, but the effect of this current, for instance, the current flowing through the vibrator V>, was eliminated by making an extra exposure on the film, the vibrator Vi being opened when V2 was being read, and vice versa. Tubes Tu and Ti3 were cx-301a's while T3 was a cx-380 rectifier tube. In Fig. 2 the emission characteristics of the cx-301a tubes used in the tests on this older type of socket-power device are shown. The tubes were prepared for service by an ageing treatment so that the emission would remain entirely steady during the taking of the film. A voltage of 100 volts was applied to both plate and grid, under which voltages the full emission current of 40 to 60 milhamperes was drawn over to these electrodes. Under this very severe overload the emission dropped slowly, and after the current had dropped to 20 mA. the voltage was cut off and the filament reactivated at 7.0 volts for a few minutes. After a few cycles of this treatment the emission remained steady for a long period of operation at the high voltage. The curves show that practically the full emission current flows when a potential of 50 volts is supplied to the anode (that is, to the grid and plate connected together as an anode) and that there is very little increase as the applied voltage is raised to 90 volts or above. The dotted line gives some data for a cx-313 tube. When a cx-313 was substituted for the two CX-301A tubes in taking the record shown in Fig. 3 B InJA Fig. 3 • april, 1929 page 393 •