What Effect?

Radio Record, Volume II, Issue 1, 20 July 1928, Page 2

What Effect?

amount and type of publicity that the | — SHIELDED-GRID VALVE rn INFLUENCE. ON DESIGN WHEN the $22 type valve was first introduced more than one engineex was heard to remark, "It’s an interesting and unusual tube, but what for?-what practical application can be made of it’? ‘This attitude (says a New York exchange) has been considerably altered, and it is interesting in the light of this to attempt to predict just what effect it will have on set design. The biggest "objection" to the valve is that it is a recent product and that relatively little is known about its possibilities and limitations. It is largely this fact that has led some American manufacturers to postpone the introduction of receivers using it. ‘There is very definitely a danger of the recurrence of the A. C. valve flurry. The A. C. valve caught the public’s fancy in America, with the result that the obsolescence factor was serious in the battery-operated set field. There seems to be considerable justification for the belief, | expressed by some American engineers, that the $22 type valve has received will result | in a negative reaction to the sets that. appear using it. Superlatives and hyperboles regarding it have been all too. prevalent.

Manufacturers’ Experiments. NUMBER of the leading American * manufacturers have experimental models of receivers using the 322 valve. | It is safe to assume that every labora-. tory worthy of the name has a num--ber of experimental models vf this. type. There is a natural reluctance to placing these receivers in production until the trend of the consumer demand is better defined, however. : If even one large American manu-_ facturer announces such a receiver the others will almost have to fullow suit. Then, too, there is the possibility of pressure from the "kif," or set construction field, where such receivers are already appearing. Failure on the part of the large manufacturers to bring ont a 822 valve receiver should result in better business for the kit manufneturers. since they can capitalise the publicity this equipment has received. From 2a technieal viewpoint the valve has very good possibilities, even though they fall quite a bit short of what the first articles on the subject Jed one to believe. The possibilities are best brought out by considering what was considered good performance in the radio-frequency stages only a year ago. A well-designed, neutralised stage usually gaye 2 yoltage ampliticatiun of abont 7 at 550 ke. and i4 or so at 1500 kc. The selectivity per stage while satisfactory a’ 550 ke, was frequently very poor at 1500 k. e. At the litter frequency the band passed was frequently 2.5 or 8 times that at 550 k. ¢.

| Careful Shielding. AREFUL shielding was necessary to permit a stable high gain amplifier, and this shielding complicated the use of the balancing circuit with its associated leads. The problem of stability was in itself serious enough so that the introduction of a new yalye permitting the same gain without a neutralizing circuit would have been very welcome. Eyen those who are pessimistic about the 822 admit that an average amplification of about 20 a stage may be secured. Unfortunately, as in the case of the usual balanced stage, the gain is about twice as high at 1500 ke. as it is at 550 kc. ‘This is due to the higher equivalent dynamic resistance of the tuned circuit to this frequency. Any losser method which tends to overcome this, such as a gridlosser, results in yery ununiform selectivity at the various frequencies. Even 20 a stage is twice that obtained from good radio-frequency circuits last year. And the stabilisation problem is not acute at this gain a stage when the over-all gain is not unusually high. To effectively use this added gain better selectivity is necessary, But this should not be serious, since the trend is toward the use of band selector circuits, in which a higher order of selectivity is possible without discrimination against the higher modulation: frequencies, , Direct Filament Current. NE electrical objection to the valve for use in A.C, sets is that it requires direct current for the filament. This difficulty has been overcome both by the introduction of suitable rectifiers capable of supplying the necessary current and by the development of heater type 822 valves. Some of the independent manufacturers already have announced such yalvyes. . Another objection to the valve arises from its physical construction. It is a dificult valve to manufacture uniformly on a large seale. ‘This trouble was also found in making valves of the heater type, however, but this problem is said to have been solved. On the whole, then, it seems that the 822 type of valve, or its A.C. filament type equivalent will have a marked effect on the design of radiofrequency cirenits henceforth. Its use should vesult in more gain 2 stage, with less difficulty due to stabilisation when moderate over-all gain is used. Its use should nof make present A.O, type receivers obsolete by any means,

It should, however, result in more extensive investigation into its performance, with the assurance of its complete acceptance in the near future. By-Pass Condenser Necessary. Dte to the fact that the shield-grid of the new 222 type valve is connected directly to the 45-volt tap of the "B" battery, it has commonly been assumed that it played a negligible part in causing instability and other troubles which arise from the use of a common "B" battery. Actually this shield-grid has a mu or amplification constant with respect to the plate of between forty and fifty. This means that any variation in its voltage produces a change in the plate current forty or fifty times as great as the same change in plate voltage would produce. For this reason it is important to see that any common impedance through the "B" battery is eliminated. This should be done by connecting a by-pass condenser from the shield-plate terminal of the socket to the negative filament terminal. ‘The fact that the circuit is not unstable does not indicate that coupling is not taking place, since any disturbance in the "B" circuit ef one particular valve will cause a variation in the shield of that valve which will produce "negative" regeneration or tend to reduce the amplificafiom,