International projectionist (Oct 1931-Sept 1933)

A "LOW-HUM" A. C. SOUND PICTURE TUBE J. O. McNally >Ii::\IBER OF THE TECHNICAL STAFF, BELL TELEPHONE LABORATORIES Supplementing that data ivhlch appeared in the October, 1932, issue of LP. under the heading "The W^.E. 262-A 'HeaterType' A.C. Vacuum Tube" is the accompanying article by Mr. McMally. The previous presentation cited only the highlights of this new tube development, further detailed information concerning which is included herein. — Editor. WHEN the filament of a vacuum tube carries an alternating current, disturbance currents of the frequencies of the power supply and its harmonics are found in the plate circuit. In voice frequency amplifiers, with sufficient gain following a vacuum tube so operated, these disturbance currents produce an undesirable or even intolerable hum in the amplifier output. In tubes where the electron emission is obtained directly from the filament these currents are so great that such tubes can be employed only in amplifiers giving very low gains. Tubes with indirectly heated, or equipotential type, cathodes may be employed for certain radio receiver and general amplifier uses, although they cannot be employed in the early stages of high-gain amplifiers. To provide for the need of an all a-c operated audio-frequency amplifier of high gain, such as are used in public address and announceing systems, sound picture projection, and speech input equipment for radio broadcasting. Bell Laboratories have developed a low-hum tube known as the Western Electric No. 262-A vacuum tube. Study of Disturbances Before undertaking the direct development of a low-hum tube, it was necessary to study the various ways by which disturbances enter the plate circuit, and to evaluate the contributions from each source. Equipment was assembled with which it was possible to measure separately the disturbance currents of different frequencies. Output currents as much as 120 db. below one milliampere, or .001 microampere, could be measured with satisfactory accuracy. With this measuring equipment available, numerous experimental tubes of the indirectly-heated cathode type were made and tested. From the results it was found that disturbance currents are introduced into the plate circuit in three ways: through the electric field due to the potential of the heater, through the magnetic field due to the current flowing through the heater, and through leakage current flowing through the resistance and capacitance between heater and grid, and heater and plate. Disturbances introduced through induction between parts of the circuit external to the tube are not considered because they are not a part of the tube problem. The electric field due to the potential of the cathode heater acts on the plate current in a manner similar to that of the field of the grid. Since the electric field due to the filament varies with the frequency of the heater supply, a corresponding disturbance current is introduced into the plate circuit. Effect of Shielding The space between cathode and plate in any directly heated cathode tube is partially shielded from the electric field of the heater by the cathode cylinder. Below the cathode cylinder, however, there is an unshielded section of heater conductor. Experimental tubes were made, therefore, in which various forms of shielding were applied to this lower section. It was found that commercially practical amounts of shielding would reduce the disturbance from this source by very appreciable amounts. In some of FIGURE 1 Assembled 262-A tube without bulb and with part of plate cut away to shoiv the grid construction [21] the experimental tubes the disturbance current produced by the electric field in shielded tubes was only about a hundredth of that from similar tubes not shielded. The shield employed in the 262-A tube may be seen in Figure 1. It is in the form a flattened bell covering the section of the heater wires below the cathode cylinder, and is supported by two short mount wires projecting upward from the glass press. Although not in contact with the cathode cylinder, it is connected to it electrically by a small wire. Plate Disturbance Component A component of the disturbance curlent in the plate circuit arises from a deflection of the electron stream from cathode to plate — and a consequent reduction in current — by the magnetic field produced by current flowing in the heater conductor. These reductions in plate current occur twice for every cycle of heating current, and thus produce a double frequency component — 120 cycle for the usual 60 cycle supply — in the output. If the two reductions of one cycle of heater current are unequal, a fundamental, or 60 cycle component, is present. The level of such a fundamental output is generally less than that of the second harmonic, and because the human ear is 18 or 20 db. less sensitive at 60 cycles than at 120, the disturbance of fundamental frequency due to the magnetic field usually is relatively unimportant. In the 262-A tube the effects of the magnetic field of the heater have been reduced by employing a filament of comparatively high voltage, and thus lowcurrent, and by arranging it in a closely spaced "U" so that the field due to the current passing up one side of the ''U partially counteracts that due to the current flowing down the other side. The heater conductor is wound in a spiral and then threaded up and back through two longitudinal holes in a small ceramic cylinder. The two holes are made as close together as is mechanically possible so that the neutralization of the field due to current in one leg by that due to current in the other will be as great as possible. This ceramic cylinder is mounted with