Radio Broadcast (May 1928-Apr 1929)

NEW TRENDS IN RADIO DESIGN FOR 1929-30 15 so badly clipped at present, and yet to prevent crofs-talk from adjacent radio channels. Dr. Vreeland's other suggestion has occasioned a lot of discussion. It consists in using in the receiver a band selector composed of — this is our interpretation, not Dr. Vreeland's — two tuned circuits coupled closely enough so that the resonance curve of such a system has a flat top, a top with a dip in it, or a sharply peaked top depending upon the constants of the circuit. According to the discussion, this is not a new idea — we do not know that Dr. Vreeland said it was — but the fact is that in the old spark days, the bane of radio inspectors was the closely coupled antenna-transmitter system which emitted a broad wave. The effects of staggering are two: (i) to broaden the top of the response curve and thereby to improve fidelity ; and (2) to reduce the r.f. gain somewhat. With deliberate staggering it should be possible to use very highly selective circuits — "low loss," if you will. Then the gain per stage ought to be somewhat greater. The gain in fidelity, however, will not be noticeable on the average amplifier and loud speaker. On the other hand, with a flat audio amplifier, a good transmitting station, and a dynamic speaker, the higher frequencies as heard should be materially improved. We must now record a visit to the Technidyne Laboratory, where we again had the pleasure of chatting with Mr. Joseph Jones who is business manager of this organization, of which Mr. Lester Jones is the engineer. This was not our first meeting or conversation, and only convinced us the more that this is an organization which will be heard from more and more in the next few years. The Technidyne group has a band selector, too, but it differs in several respects. The Jones group has in addition a self-shielded coil, a loop of similar characteristics, and an untuned r.f. amplifier of considerable gain. How these work into a modern receiver will be indicated below. The coil has two windings, one inside the other. The high potential ends of the coil are inside, so that one can grasp hold of the coil, or wrap a short-circuited turn of heavy copper wire about it, without destroying signals or even detuning the set. In our opinion this is an extremely useful invention and has resulted in a very valuable patent. So long as coils are used in receivers, there must be means of keeping their respective fields from getting out of bounds. One method is to encase the coil in a metallic can — shielding it — and another is to use a self-shielded coil. The Jones band selector will probably be used in several receivers, notably the Sparton. It works out as follows. A band selector is made up of several coils and condensers and encased in a container. This has but one wire coming from it, the connection to the following unit, the r.f. amplifier, which connects to the detector and the audio system. This amplifier is untuned, has five stages in it, with the gain varying from 3000 to 15,000 from the short to the long waves to offset the lack of coupling on these waves between set and antenna. When an antenna is attached to the amplifier input, the mix-up of signals is worse than anything the Radio Commission ever imagined. This circuit has several obvious advantages. In manufacture there are three belts on which the selector, the r. f. amplifier, and the a.f. amplifier are placed for inspection. This is to be contrasted to factories in which the completed receiver is placed upon a single belt. If anything goes wrong with a selector unit, it is removed ; the same thing happens with an r.f. or a.f. am FIG. 2 The circuit diagram of the Vreeland hand selector; note the inductance, Xs, which couples together the two tuned circuits composed of X\ and X? and their associated tuning condensers plifier. When any good selector is attached to any good r.f. and a.f. amplifier units a good receiver results. In service a similar occurrence takes place. The service man takes with him an extra selector and an r.f. amplifier. If the customer's selector is out of order, a new unit is slipped into place — the entire set is not placed out of commission. The Hammarlund-Roberts engineers have incorporated the band selector idea in their 1929 Master receiver already described in this magazine (October, page 341). In this receiver both the primaries and the secondaries of the r.f. transformers are tuned, the two circuits being coupled together by the mutual inductance existing between the two coils. This is very loose mechanical coupling. The Vreeland and the Jones system use other types of coupling, it being possible, of course, to couple two tuned circuits together with mutual or self-inductance, or with FIG 3 The hand selector used in the HammarlundRoberts " Hi-Q 29" receiver. In this case the tuned circuits are coupled together by the mutual inductance existing between primary and secondary capacity or resistance or combinations of all of these. Here, then, is the second trend toward band selectors, electrical contrivances which cut out of the broadcast spectrum a swath of the desired width. Such circuits have been suggested as a relief in the present overcrowded ether conditions. We believe, however, that another year will see the band selector the rage in set construction. The Jones system is a preselector, that is, the signals are first selected and then amplified. In the Vreeland and the Master Hi-Q circuits amplification and selection go on at the same time. THE ABBREVIATED AUDIO THE third noticeable trend is toward the elimination of the first stage of audio, and working the power amplifier by the detector. Let us see what this means. First of all it means that the detector must supply a much higher output a.f. voltage and therefore must be supplied with a higher input modulated r.f. voltage. The Jones system does not use the usual first stage of audio. The detector is a C-bias affair with input r f. voltages of the order of 10 to 15 volts. A new R. C.A. super has also been put on the market which uses a.c. tubes and has only one audio stage. The elimination of the first stage of a.f. reduces cost, makes a set simpler to construct and should reduce not only tube noises and the tendency toward severe microphonics, but eliminate considerable a.c. hum, which is a great advantage when the dynamic speaker is used with a good amplifier. One well-known physicist-radio engineer states that in his opinion the proper place for the loud speaker is in the detector circuit. Whether the set of the future will have no audio amplification at all cannot be debated at present. A NEW TUBE IN THE OFFING IN ENGLAND there is considerable talk of the ' new special-purpose tube with five elements, the Pentode. This is a power valve built along lines similar to our present screen-grid tube. The idea is to get much greater output power with given input voltages. It is a tube with a large amplification factor and a high plate impedance; with our present low-impedance speakers it may require new coupling devices. A number of articles have appeared in Wireless World (England) recently on this tube, which lead us to believe that the Pentode will do much toward making unnecessary all the r.f. amplification that eliminating one audio stage demands at present. WHAT ARE WE COMING TO? HERE, then, are the trends in sight. The first, the dynamic speaker, is here now. It is one more step toward better reproduction, greater fidelity of voice and music. The second, the band selector, is another step toward fidelity, with the possibility of an increase in selectivity. In the hands of Lester Jones the problem of selectivity has been separated from that of amplification — a feat we predicted months ago. The superheterodyne is such a circuit, although not to the degree the Jones system is. The third trend is the elimination of the first stage audio. In the Jones system this is done by using greater r.f. amplification and detectors more heavily biased than those of the present. In the super-heterodyne the amplification is at intermediate frequencies. The Pentode tube may make it possible to eliminate some of the additional r.f. gain now necessary, with an obvious advantage from the standpoint of cost and simplicity. It looks as though it is never safe to predict that the time has come when there is nothing new under the radio sun. The home constructor, however, need not feel it unwise to construct a set at present on the suspicion that next year's circuits will make obsolete his present gear. Receivers built to-day according to recognized engineering principles and equipped with good amplifiers and speakers will be standards of comparison for some time to come.