When I Think Back...

by Neville Williams

Vintage receiver design 3 Mains supplies usher in a new order

The late 1920's and early 1930's saw the establishment of a vigorous Australian radio industry and the emergence of a large factory-trained workforce, giving rise to a whole new generation of dedicated hobbyists. On the technical front, an initial wave of locally produced mains powered TRF receivers were superseded, in short order, by simplified but efficient superheterodynes.

In the early 1920's, many home fostered an appropriate awareness of the and it was recommended only for urban handymen had become involved in need to be careful with anything con- use (I quote): 'within 20 miles or so of 'wireless', primarily because it of- nected to the power mains — m point the broadcast stations'. Even so, inter- fered access to information, news and that needs to be borne in mind by the ference between adjacent stations could entertainment at a time when all three present generation of vintage radio still be a problem in difficult locations; were in short supply — particularly in receiver enthusiasts. e.g., close to one or more transmitters. the country. Compared with an equivalent two- With no relevant background and Small AC receivers stage battery set, husky valves and a limited back-up in the way of technical As happened in the battery set era, high tension supply voltage in the range literature, such newcomers to the new designs for a whole range of elementary 425-450V DC could provide loud and unfamiliar technology largely had to mains powered receivers appeared in the reproduction of such as the set learn by trial and error. Fortunately, it technical press, circa 1930, intended could successfully tune. In an optimum was a relatively safe hobby in terms of primarily for home construction. They situation, with an efficient dynamic life and limb. were welcomed by the rising generation (moving coil) loudspeaker, the construc- Operating purely from batteries, of industry-based enthusiasts who, while tor was promised reproduction to rival receivers of the day posed no threat to handling commercial receivers by day, that available from a much more preten- unskilled experimenters. The real risk were frequently too poor, in the shadow tious design. was to the equipment, with beginners all of the great depression, to afford any- too prone to confuse the battery connec- thing quite as pretentious. Dynamic loudspeakers tions, invoking the ultimate disaster for By courtesy of Mr H.D. Burraston of Elsewhere in the booklet, readers were an impecunious experimenter — Murrarundi, NSW, I have to hand a reminded that many existing dynamic 'blown' valves! copy of a booklet 'Modern Radio Cir- loudspeakers were of the 'AC powered' However, when the focus later shifted cuits for AC Operation', issued as a electrodynamic type. Introduced about from battery to mains-powered supplement to EA's predecessor Wire- 1926, a step-down mains transformer receivers, the supply voltages lurking less Weekly for August 14, 1931. It of- and copper-oxide rectifier attached to in the wiring jumped from 135 at fers very helpful glimpses of the housing provided low voltage DC to most to more then double that figure, contemporary receiver design. energise the field coil. plus a decidedly lethal 240V AC direct Fig.1 shows the circuit of the WW Looking back, I recall that discarded from the power mains. Rather than the 'Direct Coupled 2', using a 224 examples of the breed were often picked components being at risk, the greater screened-grid tetrode as a regenerative over by hobbyists, in the forlorn hope concern was that a chance high-volt- feeding a type 245 power that the transformer and rectifier would age encounter might 'blow' the un- output valve. The 280 rectifier is not in- be in good enough shape to double as a wary experimenter! cluded in the valve count. modest home battery charger. But while many enthusiasts have ex- (Perhaps I should also mention here The next generation of electro- perienced salutary 'bites' from mains that the initial digit, indicating the dynamic loudspeakers (Fig.2) were powered equipment, I'm not aware of manufacturer was later dropped from less cumbersome and generally more ef- any local reader/hobbyist who has American valve type numbers, so that ficient. Fitted with field coils of much paid the ultimate price for their inter- they became known simply as 24, 45, higher resistance, they were capable of est in radio. 80, etc). being supplied with current from the It may well be that the infusion of Guide notes in the booklet warned that receiver's own HT power supply, re- academically and/or industry-trained the sensitivity and selectivity of such quiring from 7 to 10 watts for adequate people into the ranks of radio hobbyists simple two-stage circuits were limited, field energisation.

50 ELECTRONICS Australia, August 1991 Coupled 2' assumed that omission of the coupling capacitor would effect a noticeable improvement in the overall quality of reproduction. I subsequently contested this simplis- tic — but not infrequent — assumption in the very first instalment of 'Let's Buy an Argument' (now 'Forum% maintain- ing that the omission of a lone coupling capacitor in a simple non-feedback merely complicated the rest of the circuitry to no good purpose. In the circuit of Fig.1, the grid of the output valve is tied to the positive poten- tial at the anode of the 224 detector — a fact that may have influenced the choice of the anode-bend configuration. For normal class-A operating condi- tions to apply, the filament of the 245 would need to be about 50V positive with respect to the grid, with the anode supply 250 volts above that again. Ac- cording to the descriptive text, the design calls for a total supply in the range 425-450V — an incongruous fig- ure for such a small set. Fig.1: Redrawn from a Wireless Weekly booklet issued in August 1931, here is To obtain this voltage, the power the circuit diagram of the 'Direct Coupled 2", an early and modest mains transformer/rectifier system has to operated receiver Intended for home construction. operate in half-wave mode, as shown. With a ripple frequency of 50Hz and Subsequent generations of Australian- ing in the transitional period between the the filtering relying on a nominal 30 made Amplion, AWA, Rob and Mag- technology of the 1920's and 1930's. henry choke and two 4uF paper navox loudspeakers looked very much Thus, while the configuration of the capacitors (Fig.3), I do wonder about the like the Jensen pictured. After World detector is reminiscent of the traditional residual hum level. (Electrolytic War II, field coils gave way to per- Reinartz reaction circuit, there is no grid capacitors had yet to emerge as a routine manent magnets. capacitor or 'grid leak' resistor. Instead, choice for HT filtering.) With hindsight, various aspects of the the 224 is shown as an over-biased I also wonder about the presence of 'Direct Coupled 2' circuit reflect the 'anode-bend' detector — an innovation the loudspeaker field coil in the fila- sometimes immature technical reason- more appropriate to larger receivers, ment/earth return circuit of the output subject to possible detector overload. valve. Superficially it might appear to be For a simple receiver, the traditional and a neat way of energising the field, but a reputedly more sensitive 'leaky grid' complex and potentially resonant im- circuit would normally be a more logi- pedance in a path common to both input cal choice. and output must introduce random nega- tive current feedback around the output The output stage stage, affecting its output impedance. As Again, the output valve is a directly well it could divert audio power from heated type 245, normally requiring the voice coil into the field coil. the hum to be balanced out by return- To invoke an old saying, the complica- ing the negative feed from the HT supp- tions involved in eliminating one lone ly to a filament circuit centre-tap — coupling capacitor strike me as the tech- apparently overlooked in the circuit nological equivalent of 'straining at a diagram. In fairness to the designer, use gnat and swallowing a camel'. of the 45 probably has more to do with the valve manufacturers, who clung Different approach tenaciously to directly heated output val- Interestingly enough, the same supple- ves long after all other functional mains ment offered readers a quite different types had been developed around 2/3-valve receiver called the 'The Hi- FIg.2: A then-current American sleeved cathodes. Power Two'. Two separate coils and two Jensen dynamic loudspeaker, But, to me, the most debatable aspect ganged capacitors provide band-pass Imported by the International Radio of the design — a carry-over from the tuning (see Fig.4) ahead of a Mullard Company Ltd of Sydney. The field much publicised 1920's-style Loftin- 354V indirectly heated triode, operating magnet is bolted to the rear of the White amplifier — is the use of 'direct as a conventional regenerative leaky- cone, with the cable tagstrip and coupling' to the grid of the output valve grid detector. voice call matching transformer from the anode of the preceding stage. As a result and as distinct from the suspended underneath. Clearly the designer of the 'Direct- Direct Coupled 2, this alternative design ELECTRONICS Australia, August 1991 51 tiometer in series with the capacitor so WHEN I THINK BACK that the tone could be varied at will be- tween 'bright' and 'mellow'. was credited with 'astounding In fact, such lop-cue tone controls be- selectivity' (for a small set) and said to came a routine fitment on domestic be 'at its best when located right in the receivers from the early 1930's onward, midst of the powerful local transmitters'. being commonly left in the 'mellow' set- The detector was transformer coupled ting. As a result generations of listeners to a British-based Mullard PM24A became conditioned to 'woolly' music power output pentode — as distinct and muffled voices. But back to the from the American 24/24A RF tetrode. original theme. The specified loudspeaker was again a Jensen dynamic with 8000-ohm field Extra stages coil but, in this case, wired directly As had happened in the early 1920's, across the DC HT supply. small regenerative receivers like the Requiring only a routine 250-odd foregoing remained largely the province volts, the power supply system involved of hobbyists — of fmancial necessity a 280 type rectifier fed from a normal and/or because they derived a certain centre-tapped power transformer, plus a Fig.3: A typical paper dielectric satisfaction from achieving impressive couple of filter chokes and the then capacitor, used for HT filtering before results with a minimum of circuitry. routine `Chanex' or 'Hydra' 4uF paper the general adoption of the electrolytic On the other hand, commercial dielectric capacitors. can type. Many discarded block' capacitors found their way into early receivers, intended for family use, were Also worthy of mention in this other- home-built amateur station invariably of more ambitious design, wise poor man's 2/3 valver is the in- transmitters. less reliant on operator skill and able to clusion of an audio volume control cope routinely with a greater range of ahead of the output stage, and a top-cut reasonable quality, the overall sound can reception conditions. From the view- tone control across the anode circuit. be relatively smooth. point of both supplier and customer, the Both warrant explanation at this point; By contrast, power output tetrodes and ideal receiver was an affordable model first the volume control: pentodes have a much higher output that could be delivered to any ordinary In urban areas, as already indicated, resistance — typically 60,000 ohms in address, connected to an aerial and a small mains powered regenerative the case of a 247. In consquence, the power point and tuned, forthwith, to set could provide quite high output frequency response tends to vary with the full gamut of stations available in from strong local stations, necessitat- the loudspeaker impedance, resulting in the area. ing some means of reducing the sound a rather `boomy' quality at the Fortuitously, the abovementioned to an acceptable level. The seemingly loudspeaker's bass resonance and a Wireless Weekly booklet features a range obvious course was simply to back off pronounced accentuation of the upper of receiver designs using three, four and the regeneration (or 'reaction') control, treble. The difference between the two five valves plus rectifier — all TRFs, but this could adversely affect the was quite noticeable but, while listeners with one, two or three tuned RF stages selectivity, leading to possible inter- tended to tolerate the extra bass as a ahead of the tuned detector. All feature ference problems. novelty, they disliked the strident treble. direct coupling to the output stage, By providing a supplementary Designers countered by installing a which complicates the audio circuitry. volume control, the regeneration could suitable bypass capacitor across the But more importantly for our present be set for maximum detector and anode circuit of output pentodes, to at- purpose, the notes document how well selectivity, the volume control being tenuate the upper treble response. In- the respective front ends coped with the adjusted separately to produce the deed, translating a vice into a virtue, less crowded broadcast scene in desired sound level. While a routine they commonly provided a poten- 1930/31. procedure for a technically inclined, lis- tener, the critical manipulation of two knobs, both affecting volume, would have been potentially confusing for other members of the household. The tone control Most receivers up to this point in time had used triode output valves having a characteristically low output (or anode/plate) resistance; e.g., around 1600 ohms for a type 245. In such a case, the natural frequency response of the system is not greatly af- fected by the wide variations in im- pedance which loudspeakers typically Fig.4: Double-tuned band-pass' circuits were often employed to Improve exhibit over the audio range. Hence, front-end selectivity. Arrangement (b) is the logical choice when the tuning with a clean and a loudspeaker of capacitors share a common, earthed frame. 52 ELECTRONICS Australia, August 1991 Again, helpfully, a Wireless Weekly rated output pentode originally specified the excessive signal level may be evi- `Call Sign' booklet issued a few months and with unreliable resistors. Again, dent as distortion caused by overload of earlier (July 1930) lists the 33 AM with hindsight, I am not surprised when the final RF stage or the detector itself. broadcast band stations across the the article goes on to mention that the Either way, a gain control in the audio Australian continent — compared with a HT supply voltage provided for the system is of no help, since the overload present tally of around 250, crowded direct-coupled circuitry ended up at has already occurred ahead of where the into channels only 9kHz apart. around 625-650VI control can have any effect. The Wireless Weekly 'Midget 3' was Notwithstanding this, the fact remains Superficially, the gain of RF tetrode so called because it was fitted on to a that the two-RF stage configuration, can be be reduced by simply chassis compact enough to be housed in with single-dial triple-gang tuning increasing the negative grid bias — a table-top cabinet. It used a 224 as a (C1/C2/C3), won wide acceptance as the either directly or by using a wire-wound tuned RF amplifier followed by another best compromise for the reception con- potentiometer as a variable cathode bias 224 as a tuned anode-bend detector ditions that obtained in 1930. resistor. The problem is that this simul- feeding a 245 power output triode. A 280 Not surprisingly, many contemporary taneously reduces the plate current and rectifier in half-wave mode produced the commercial receivers adopted this further limits the ability of the valve to requisite HT voltage for the direct coupled general approach, with a 3-gang tuning handle high level signals. Thus, only ouput stage, the 8000-ohm field coil being capacitor and two RF stages, followed partial control is possible, with cross- connected across the HT in series with a by a detector, an ordinary capacitance and/or distortion remaining 5000-ohm heavy duty resistor. coupled output stage and a conventional a potential problem. Using a 224 tetrode in the RF stage 250-odd volt power supply. In the '1930 Four', the designers have ensured much greater amplification of opted instead for an 0.2 megohm the incoming signal than was normally potentiometer varying the screen volt- available with a triode. No less to the age of the first two valves. While this point, the presence of a screening was often used at the time, the idea electrode between grid and plate, and ac- would appear to suffer the same limita- cess to the grid via a top cap connection tions as variable bias. reduced the grid/plate capacitance by That it did so is evidenced by the fact around 100:1. that the design specified a 'Local- This obviated the need for neutralising distant' switch which introduced a low circuitry, the only precaution in the in- value resistor (typically 10-25 ohms) terest of stability being the use of metal across the primary winding of the aerial shield cans over the 224 valves and the coil. Intended to reduce the level of all respective coils (Fig.5). incoming signals, the resistor could be Significantly, the compilers of the switched in or out of circuit, depend- booklet classified the 'Midget 3' as ing on whether or not it provided a the smallest class of receiver which cleaner result. could offer reasonable gain and Gain control posed a problem for selectivity without having to rely on many medium to large receivers about the use of regeneration (quote) this time, both TRF and superhets and, `which is considered objectionable while a local-distant switch was a poten- by many enthusiasts'. tial source of confusion for non- techni- Fig.5: Manufactured by the cal users, it was commonly fitted to both `Most popular' set Metropolitan Electric Co, Sydney, commercial and home-built designs as a The sequence of small receivers leads shielded, matched aerial and RF coils matter of necessity. up to what the booklet presented as the like this were available for TRF One other point worthy of mention is then-current ideal receiver: 'The 1930 receivers, priced at 9/11d (990) each. the provision of a 'jack' socket (J) in Four', whose circuit is shown in Fig.6. It series with the earthy end of the detector was described as: 'the most outstanding Gain or volume control grid coil. For the most part ignored, its design ever known to the Sydney radio One other point about the '1930 Four' purpose was to allow a phonograph trade', for which 'more kits of parts have warrants special comment, namely the pickup to be plugged into the grid cir- been sold ... than any set previously matter of a gain (or 'volume') control. cuit of the detector so that the phono described in Australia'. Because of their high intrinsic gain signal would be fed through the Why? Because it had unmatched tone, and consequently low grid bias, RF amplifier and loudspeaker. `enough volume to fill a small hall' and tetrodes like the 224 could readily ex- It could be argued that a detector sufficient range to give excellent inter- hibit signal overload effects close to one would not be optimally biased to state reception. or more broadcast stations, therefore in operate as an straight amplifier, and Another TRF design, it used two 224 many urban areas. that there was no provision anyway tetrodes as tuned RF amplifiers, a third A powerful local signal could con- for an audio control to vary the sound as a tuned anode-bend detector, fol- ceivably penetrate the first tuned circuit level from discs. lowed by a Philips P443 — a directly- and be cross-modulated onto other car- Both observations are legitimate but, heated high-power output pentode. The riers by the first RF stage. Even though at the time, the majority of pickups were rectifier was a high-voltage half-wave the interfering carrier itself may be relatively crude magnetic types, and 281, fed from a 575V transformer. rejected by subsequent tuned circuits, its fitted with their own loudness poten- The covering article indicates that audio component could still break tiometer anyway. Most listeners were there had been problems with the design, through as spurious modulation on other aware that electrical phono amplifica- when first published, with the lower carriers. Or, again, in a larger receiver, tion was possible, but the current em- ELECTRONICS Australia, August 1991 53 local-distant switch, the set may indeed with its consequent cross-modulation WHEN I THINK BACK have been unduly vulnerable to front- and distortion. end overload and mains-borne RF inter- Had 235 type valves been available in phasis was on broadcast programs. ference in urban areas. time for the '1930 Four' and the 'De- Family radiograms did not really be- Front-end gain control remained an Luxe Five', the designers could simply come 'trendy' until around 1935. urgent problem until the introduction of have specified them for the RF stages, `super-control' or 'variable-mu' screen arranged for variable cathode bias and A 'deluxe' TRF grid valves, in mid 1931. The Interna- ended up with much smoother control The most ambitious receiver presented tional Radio Company of Sydney an- — minus the local-distant switch. in the Wireless Weekly booklet is The nounced local release of the National `De Luxe Five' — or 'Six', counting the Union variable-mu 235 in Wireless Selectivity vs quality rectifier. An elaboration of the foregoing Weekly for April 13,1932. That aside, while receivers such as 4/5 valve receiver, it used three tuned The variable-mu characteristic was those above offered predictable gain and RF tetrodes ahead of the 224 anode- achieved by fitting a special tapered- selectivity for the listening situation bend detector which, in turn, drove a pitch grid, or by simply snipping circa 1930, the industry was well aware Philips F443 high power output pentode. selected half-turns from a fixed that, as far as domestic receivers were The rectifier was a half-wave 281, fed pitch grid. concerned, such designs epitomised the with a 600V transformer. As distinct The effect was to change the grid con- practical limits of TRF technology. Two from the smaller receivers, the circuit trol characteristic such that while it ap- and three-gang tuning capacitors were called for an 'AC Jensen' loudspeaker, peared quite normal with low values of acceptable, four-gangs were manageable providing its own field energisation with bias, it required a very high value of bias but anything beyond that would be too an in-built mains transformer and cop- to cut off the plate current altogether. clumsy and too expensive. per-oxide rectifier. The abrupt plate current cut-off In any case, the selectivity curve ex- Summing up the receiver, the booklet `corner' in the characteristic was com- hibited by a TRF tuner varied unduly claimed that, irrespective of valve count, petely eliminated. across the broadcast band. According to there was nothing that could outperform Virtually identical in appearance to the one set of figures to hand, when tuned to the De-Luxe Five 'except for some 224, the variable-mu 235 offered essen- 600kHz, a typical receiver using a three- forms of superheterodyne'. It was par- tially the same transconductance — and gang capacitor exhibited a total ticularly recommended for use in 'up- stage gain — as the 224 at -3V. But bandwidth of 100kHz at 60dB down. country towns' involving a range of whereas the 224 plate current curve cut However, when tuned towards the high several hundred miles, but was not off sharply at around -6V, the 235 plate frequency end, its bandwidth, as a recommended for major cities because current curve trailed out to an ultimate simple proportion of the resonant fre- `it picks up all kinds of interference cut-off at around -50V. quency, widened to around 250kHz. noises and amplifies them to an extent In announcing their equivalent valve, Unfortunately selectivity was worst at which will make reception unpleasant'. the 335 in QST for July 1931, Cunnin- the very end of the band where it really This is an off-putting statement, to say gham listed the transconductance as needed to be at its best — thereby com- the least, but probably confirms the fact 1.050mA/V at -3V, reducing to paring very unfavourably with a super- that, at the relevant time, it was one 0.015mA/V at -40V — implying a huge heterodyne, which could provide a thing to provide high RF gain but potential reduction in stage gain. With narrower passband which was substan- quite another to control it smoothly that degree of control available and a tially uniform over the whole tuning and effectively. plate current curve free of abrupt range. Arguing with as much spirit as Relying on screen voltage control for corners, it became much easier for characterised their later forays, audio the three 224 RF amplifiers and a drastic designers to forestall front-end overload buffs of the period stoutly maintained

F443 RFC

AERIAL 224 224 I 221.

L2 OUTPUT Li TRANSFORMER

LNIF LOCAL "t1pF DISTANCE 30H SWITCH 02M II VOLUME *SPECIAL TAPPED RESISTOR OR 'VOLTAGE DIVIDER' U. 281 4 F 01pF tJA

240VAC ►

Fig.6: The '1930 Four', said by Wireless Weekly at the time to be their most popular project to date and considered to be the best current compromise between performance and cost for the average listener.

54 ELECTRONICS Australia, August 1991 7-8 VALVE T. RF. BROADCAST RECEIVER

6U7G Om) 6U7G 0061 6J7G We) 6J7G lecR) 6J7G feta) 2A3

SOOT LOAD P - P — scoo p

FIELD COL EA01 2'.,D:042 21.1. 30.1. 'DOA OW. —' ----. TOTAL RED.RED. , p"OO 240V. DO'. .2oo v 04 .1)

RADIOT PON CIRCUIT NO. oT,

Fig.7: Designed especially for audio buffs and featured in `Radlotronics' No.82 (1937), this receiver combined a three-valve TRF tuner with a classical push-pull 2A3 triode amplifier. that emphasis on receiver selectivity was bypassed only for RF. It did not load the inappropriate; that the vast majority of input circuit and had notably low distor- families listened only to their local sta- tion by reason of the cathode negative tions, and that extra selectivity would feedback. Because the design did not diminish rather than augment their lis- lend itself to AVC, the user had to tening pleasure. manipulate two manual gain controls — Receivers with a narrower bandwidth one for the front end, the other for the had to be tuned with greater accuracy for audio system. High impedance primary minimum distortion, they suggested. windings in all coils and a capacitive What's more, they attenuated the high coupling loop adjacent to the active end frequency sidebands, diminishing the of the respective secondaries helped to clarity of voices and robbing music of its equalise the gain across the tuning natural overtones. range. Fitted with a push-pull power The arguments gave rise to an audio triode output stage and an appropriately cult which supported TRF receivers on baffled hifi loudspeaker system, it was principle, and equated them with `hifi' very much a receiver/amplifier for hifi radio reception. enthusiasts of the day. In all fairness, it probably has to be In May 1941, the Editor of this admitted that the best sounding receivers magazine (by then called Radio & Hob- of the era were TRF tuners with two bies) the late John Moyle, presented a variable-mu RF stages, an anode- TRF tuner very like that used in the bend detector and a power triode out- AWV receiver, with the idea that it could put stage. The modest front-end be used with an existing R&H amplifier provided a reasonably balanced signal using push-pull 2A3 output . For and the power triode(s) delivered it to sound quality, he said, the combination the loudspeaker with much less distor- would be 'almost unbeatable'. A year tion than characterised the louder — later, in May 1942, I personally and more strident — non-feedback described the 'TRF Quality Six' in power pentodes. Radio & Hobbies. With parts scarce, due As we shall see in the next chapter, to the war, it was a distinctly different vital design initiatives enabled the su- economy design, slanted to take ad- perheterodyne system to capture the vantage of possible alternative com- mass market from 1930 onwards. But ponents. Since then, TRFs have been TRF tuners retained lingering support remembered mainly 'in absentia', with throughout the decade from specialist hifi-orientated engineers more intent on suppliers and audio buffs. dreaming up ways and means of creating Fig.7 shows the circuit diagram of a wideband or variable-selectivity super- TRF receiver for hifi enthusiasts hets. Only recently, with the arrival of developed in the Applications solid-state technology and AM-stereo Laboratory of the A.W. Valve Company has such technology come of age. and published in Radiotronics No.82, But that's much too recent to qualify dated December 1937. It used two tuned as history. In the next chapter I will be variable-mu RF amplifiers and a 'reflex' looking at the evolution of 1930's- detector — essentially an anode bend style superhets. detector with a large cathode resistor (To be continued) •

ELECTRONICS Australia, August 1991 55