Radio Broadcast (Nov 1926-Apr 1927)

A Five-Tube Nom-Oscillsitine: Receiver Wherein an Effective System of Receiver Stabilization is Described — The [Causes of Mil THIS article is concerned with a method of stabilizing radio-frequency amplifiers and the application of this system to a five-tube receiver employing two tuned r.f. stages, a non-regenerative detector, and a transformer-coupled audio amplifier. The receiver has only two tuning controls because a gang condenser, C2 is used to tune the second and third coils. It is capable of giving excellent tonal quality when worked with a good quality loud speaker, and is absolutely stable in performance. The receiver is easily and permanently stabilized, and will not squeal once it has been adjusted. It tunes sharply and will satisfactorily separate local stations. In designing any tuned radio-frequency receiver, one of the most important problems is to properly stabilize it, since any ordinary highfrequency amplifier has an inherent tendency to oscillate, and when this occurs, reception is ruined by the howls that are created. Stabilization can be obtained in various ways, and it will be worth while to orient very briefly our thoughts regarding these different methods. Perhaps the ideal way of preventing oscillations in a high-frequency amplifier is through the use of a bridge system, since such a method does not introduce any loss, and is, theoretically at least, capable of being arranged so that its effect is practically constant over the entire band of broadcast frequencies. However, a bridge system is difficult to adjust, necessitates that the apparatus be laid out very accurately, and that the wiring be carefully done. There are indeed few receivers on the market efficiently designed and completely neutralized. Diametrically opposite to this method, we have the simple "losser" systems. Resistance in series with or across the tuned circuit very effectively prevents oscillations, but it also precludes considerable amplification. Such systems are generally easy to adjust but are not constant in effect over the entire broadcast spectrum. Consequently, if adjusted so as to prevent oscillations at the high frequencies (short wavelengths) where they are most likely to occur, the loss in amplification at low frequencies is considerable. Between these two methods (the bridge systems, constant over the broadcast band but difficult to adjust; and the "losser" system, very simply adjusted but also very inefficient and unequally effective over the broadcast band), there lies a third system, also quite easy to adjust and fairly constant in effect at different frequencies. This leads us to the Phasatrol, a stabilizing device used in the receiver illustrated in this article. The Phasatrol is easily adjusted and will completely prevent oscillations. Also its effect is fairly constant with frequency so that only a slight loss of amplification at low frequencies Circuits — The Triple By T. H. NAKKEN (long wavelengths) occurs. To properly understand the action of the Phasatrol, one should fully realize why a radio-frequency amplifier always tends to oscillate, and it will be attempted to give here a simple, understandable explanation of the reasons causing oscillation. We all realize that a vacuum tube is an extremely sensitive device, which amplifies almost completely without any distortion any kind of electrical disturbance which is experienced by its grid — the one element which controls its actions. Now, in a radio receiver there are many electrical currents flowing in various circuits, and each one of these circuits creates its own electrical disturbances, which easily may act upon one or more of the grids of the different vacuum tubes. This reaction of the various circuits is called coupling, and only too easily gives rise to the familiar squealing, always an indication of oscillations in the receiver. This oscillation is the nightmare of every radio manufacturer and experimenter, and ceaseless labor is being expended everywhere to prevent the occurrence of these oscillations. Oscillation will almost invariably occur when there exists coupling between the various stages of radio-frequency amplification, and, in general, we may say that there are two ways in which circuits can be coupled, i.e., inductively and capacitively. By judicious design we may construct receivers in which extraneous coupling has been reduced to an absolute minimum, by using parts that show the least tendency to act as coupling devices, and placing them in such a way that the danger of coupling is largely prevented. But even when in the receiver itself, not the slightest chance exists for coupling of any kind whatsoever, there would remain one coupling device, which always tends to cause oscillations and instability. This is the amplifying tube itself, in which there exists an inherent capacitive coupling factor due to the proximity of its elements, notably its plate and grid. Any change in the potential of the plate reacts capacitively upon its grid, and these changes in plate potential almost invariably occur in such a way that their disturbing effect on the grid coincides with the original disturbance on the grid. This means, then that this inter-element capacity has the Radio Broadcast A FRONT ASPECT OF THE RECEIVER The milliameter serves as an excellent check on the quality, as explained elsewhere in this article effect of building up the original disturbance, and when this happens, oscillation is sure to occur, unless proper action is taken in the design of the receiver to prevent this inter-element capacity from having such an effect. When we analyze why this influence of the plate upon the grid has the effect of causing oscillations, we will understand the method employed in the Phasatrol to overcome this. Generally, in a radio-frequency amplifier, the coupling device between two success've stages is a tuned r.f. transformer. Such a transformer almost invariably consists of an untuned primary and a secondary, the latter being tuned by means of a variable condenser. The plate circuit contains a coil and therefore the circuit obeys the laws that govern the behavior of alternating currents in inductive circuits. Now it is a well known fact that, in such a circuit, the current lags behind the voltage, as it is expressed. This means that, as there is a difference in the time of maximum voltage and the maximum current in such circuits, the current reaches its maximum value after the maximum voltage has been developed. This then can also be expressed by saying that the voltages are ahead of the current, which in this instance is very important. When we now consider that the current fluctuations in the plate circuit are due to disturbances on the grid, it follows that, when the currents lag behind the voltages, the potential changes in the circuit, and of the plate, therefore, are advanced with respect to the plate-current changes. When the plate changes in potential, which change of potential is always larger than the original impulse on the grid, the plate reacts upon the grid, due to the fact that the two form a small condenser. The back action of the plate on the grid is therefore a capacitive one, and there is an electrical law which states that in capacitive circuits the voltages lag behind the currents, meaning that the voltages developed reach a maximum only after the current has reached its maximum, or rather, when the current has already passed its maximum. This is exactly the reverse action of an inductive circuit. When we look at these facts, we will see that the voltage variations of the plate are in advance of the current, but that the voltage variations of the grid, due to those on the plate, experience a lag. These voltage variations of the grid due to the plate circuit variations, will almost coincide with the disturbances on the grid which caused the changes in plate current in the first place. In other words, the back coupling, due to the capacity between the tube elements, reenforces the original impulse and this condition causes the further Photograph building up of the original impulse, and either regeneration, or, if the back action is strong enough, oscillation, occurs. It follows that, if we only