The Marconi Screened Valve Type S625

Radio Record, Volume I, Issue 35, 16 March 1928, Page 10

The Marconi Screened Valve Type S625

vr ‘These valyes will very shortly be} en sale in New Zealand, and there appears te be quite an impression that this will prove superior in use to the | screet-grid valve of American amanufacture. Certainiy the arrangement of the electrodes in the British pattern appears to be based upon thoronghly sound principles, and so far as can be judged from articles in recent magazines to hand, is giving good service. An American radio journal considers that the mechanical arrangement im the British valye is the better, as it seems, in the opinion of that journal, to have been worked out with more thought toward ease of use. However, | we shall see. This valve has four electrodes, but must net be confused with the foxrelectrode valve. The valve is designed so that the capacity of control grid to plate is quite negligible when the screen is connected to a point of fixed potential such as earth or the plate circuit battery, and its plate impedance and amplification factor are of quite .a different order of magnitude from those usual for the ordinary types of valve. The harmful effects of grid to plate’ capacity in the three-clectrode valve are well known as causing ai almost chronic state cf instability in radiofrequency amplifiers. However we reniove stray coupling between leads, coils and condensers, there would always remain this comparatively high inter-electrode capacity in the valve, requiring neutralisation that might or might not be very nearly perfect. In order to obtain stability it is frequently necessary to sacrifice a certain amount of amplification, and perhaps sclectivity also, The screened valye does away with the necessity for neutralisation, The impecance is about 300,000 ohms, and | the amplification factor 135, but this_ fizure must not be taken as the amount ef amplification obtained, as this depends entircly upon other factors in the circuit. Published curves show that as the plate voltage is increased, the plate current rises from a small value until, at a certain point, it deereases rather rapidly, only to rise avail and to remain practically steady

for the high¢r j-late voltages. It is clear that, given an output circuit of suitable design, the voltage amplification obtainable is very large indeea. Macnification depends upon the impedance of the plate circuit, and the aniplification increases with the 1mpedance, so the question is how to obtain the necessary high impedance in a practical cirenit, Tinpedance, Calculated connected to plate. magnification. G00,G08 ofS wuscecescerreeeeeee 90 HOO 000 O1s ..reecseeececeeeee 67 200,000 olims ...... thaeseneeeesene 54 100,000 ohms ......... te ceseeeense o4 HO,00) oftg ve. cece eee = 19 25,000 Ohlins voc. cece eeeeceeee ee 10 The above table shows how magnification fails off as the impedance connected to the plate circuit is reduced. If the impedance of the plate circuit is made equal to that of the valve, the yollage amplification shonld be 135 divided by %, or, say, 67. | Four firms now produce the shielded grid valve-Marconi, Osram, Cossor, und Mullard, the arrangement of the latter being slightly different to the others in regard to leads, which appears to be rather a mistake. ‘This latter has an ordinary four-pin base, the usual plate pin being connected to the screening grid, and the plate counectien to a single terminal ut the top of the valve. Normal filament current is 0.25 amp. at 6 volts, aud for an average may be taken 120 volts on the plate, 80 on the shield, and 9 volts on the control grid. As the internal resistance is high, it is desirable to use "litz’’ wire for shore wavelet:gths (up to 600 metres) coils to obtain imaximum impedance when UI DRE rene en oe _

tuned, and toe reduce losses to a minituum. The-best type of coil is the single layer solenoid, 24 te 3 inches in diam-' eter. A single stage of R.F. annplifi-’ cation, it is said, presents no construcfional difficulties, and for broadcast reception does not require elaborate screening, a partition through which the valve passes beiug in many cases sufficient. Where reaction is used for range | the detector tube shonld be of low impedance (6060 ohms normal), followed by a low ratio transformer atid a power

tube. Sour: kind of ‘tieldless’? winding is advisable for the coils. Captain Round is of pinion that the adoption of this valve in an R.T, stage preceding the detector will reduce the possibility of- radiating interference to a negligible minimum. We reproduce a circuit suitable for this yalve as an R.T. amplifier, but only experimenters will be interested in Corie | structing it at present. (eT eLTEEEDELEL ED EC ULE ELT EEL DCL ELD DECREE LL EAE ET

PURROUPURE RRO RR eRe ee Shori-wave Receivers. A British journal, commenting on tlie construction of short-wave receivers, says: "It is one thing to draw a circuit diagram and another to build an efficient receiving set from it. For this reason readers who contemplate the construction of short-wave receivers are strongly advised not to attempt to make their own designs, unless they have had a good deal of experience of short-wave work, but to copy faithfully a published design. If this is done, and no alterations or ‘improvements’ are made in the design, the necessary smoothness of reaction control will be obtained without difficulty.’ ‘