Fullscreen HTML5Med Res (??MB)HTML4

Vintage Radio By RODNEY CHAMPNESS, VK3UG The Fisk Radiola 50G battery receiver Housed in an “Empire State” style Bakelite cabinet, the Fisk Radiola 50G battery-powered receiver is capable of both broadcast band and shortwave reception. It was first produced in 1939 and has a few unusual design “quirks”. R ADIO ENGINEER Sir Ernest Fisk play­ ed a vital role in founding Amalgamated Wireless Australasia (AWA) and later served as its Managing Director and Chairman. In practice then, the Fisk Radiola 50G was really an AWA receiver. Produced in 1939, this batterypowered receiver used either a 2V wet cell or an air cell (via a resistor) to power the filaments. These had a total current drain of 0.6A but this increased when the dial lamps are switched on. However, this extra drain was only This view shows the excellent condition of the loudspeaker. The 761 bias battery measures 2V (down from 4.5V) and is overdue for replacement. 94  Silicon Chip short term, as a switch has to be held in to actuate the lights. There’s not a lot of information on the air cells that were used in this set, other than that a resistor was needed to drop their output voltage to 2V. What little information I can glean indicates that they have a terminal voltage of around 1.4V, so two would have been used in series in this receiver. The high tension (HT) of 135V was taken care of by three 45V batteries connected in series, the current drain on this line being 13-15mA. In addition, there was a separate bias battery (type 761) which applied -1.5V, -3.0V and -4.5V to various sections of the receiver. By the way, a variant of the model 50G, designated the 50GV, was also available. Its circuit was almost the same but was modified to use a 6V vibrator power pack. It’s worth noting that radios at the start of the 20th century were almost exclusively battery powered. This was due mainly to a lack of suitable filtering components to produce the pure DC required for the filaments and plates of the valves. However, as time progressed, valves with indirectly heated cathodes made it easier to design sets for AC mains operation. At the same time, valves with lower filament current and voltage requirements were developed specifically for use in battery-powered receivers. Eventually, the later dry battery valves that were developed led to portable receivers using the 1R5, 1T4, 1S5 and 3S4 valve line-up. In the mid-1930s, valve manufacturers produced many valves that worked well from a single 2V lead-acid cell (battery) and three series-connected 45V dry batteries for the HT. In fact, valves were really coming of age at that time, with many of the octalsiliconchip.com.au The Fisk Radiola alongside its shipping carton. It’s quite rare to find the original packaging for old receivers. based types turning in quite good performance. The Fisk Radiola 50G The Fisk Radiola 50G described here is as complete as any set I have come across. It has the original shipping carton, ancillary items including an air-cell series resistor and a small aerial lead, and all the literature that came with the set. These items make this particular 50G all the more valuable and the literature makes interesting reading. The 50G and the 50GV (vibrator version) sets used five valves and would have been quite sensitive receivers for the time. The valve line-up starts with a 1C7G converter, followed by two stages of intermediate frequency (IF) amplification on 460kHz using two 1D5G valves. This then feeds both a detector diode and an automatic volume control (AVC/AGC) diode within a 1K7G valve. Audio amplification is then achieved using the pentode section of the 1K7G valve which in turn feeds a 1L5G audio output valve. These were all octal-based battery valves, used from the late 1930s through to the early 1950s. Dual-wave operation An interesting feature of the 50G is that it is a dual-wave radio. The origisiliconchip.com.au The chassis easily fits into the tall cabinet and is still in good condition. April 2006  95 nal versions covered 550-1500kHz on the broadcast band and 6-18MHz on the shortwave band but the broadcast band was later extended to cover 5301600kHz. Dual-wave operation was quite pop­ ular in those times, with many Australians interested in listening to radio stations in Europe and to the BBC – especially when the test cricket was on. Battery sets were mostly used in areas where a mains electricity supply was unavailable and that usually meant on farms and on remote stations well away from towns and cities. Receivers like the 50G were quite reasonably sensitive although they would have been a little noisy due to the noise generated by the 1C7G. Pentagrid converter valves generate more noise internally than triode hexode valves such as the 6AN7. To overcome this problem, a substantial outside antenna would have been required. However, large outside antennas were installed as a matter of course in those days, so the noise generated by the 1C7G would usually have been swamped by strong radio signals. As with many other receivers, there was no RF amplifier stage in this set. This meant that the band-change switching could be kept simple and that in turn meant lower cost. Main features The above chassis (top) and under chassis views. No components appear to have been replaced in the set. 96  Silicon Chip As shown in the photos, the receiver is installed in a brown bakelite cabinet, with Empire State styling. This cabinet was made using three separate pressings. The front plate of the set is one pressing, the back is another and the main pressing (body of the cabinet) holds the chassis. The cabinet is in extremely good condition and will need little if any work done on it (the reflection off the front panel gives a false impression of its condition). In fact, the set’s owner, Brian Lackie, is in two minds as to whether he will restore this set, as it has had so little work. It is so complete that it may be best to leave it in its original state. The front-panel controls, from left to right, are “Tone”, “Volume”, “Tuning and Dial Lamp” (concentric knobs), and “Off-On and Wave Change”. The idea of the wave change and on/off control being on a 3-way switch has me wondering why AWA did this, as there are nine sections to the switch. siliconchip.com.au Silicon Chip Binders REAL VALUE AT $12.95 PLUS P & P Fig.1: the circuit of the Fisk Radiola 50G. This battery-powered set used five valves and was capable of both broadcast band and shortwave reception. H SILICON CHIP logo printed in gold-coloured lettering on spine & cover H Buy five and get them postage free! Price: $A12.95 plus $A7.00 p&p per order. Available only in Australia. Just fill in the handy order form in this issue; or fax (02) 9979 6503; or ring (02) 9979 5644 & quote your credit card number. As well as its original shipping carton, the Fisk Radiola 50G also came with all its supplied literature. However, one advantage of this switch mechanism is that in the off position, the antenna was isolated from the aerial coils. This meant that with a big high antenna, there was less likelihood of damage to the aerial-coil due to nearby electrical storms. The only error they made here was that the aerial was not earthed when the set was switched off (there is a spare contact that could have achieved this). The dial lamps were actuated by pressing the concentric knob on the tuning shaft. This illuminated the secsiliconchip.com.au tion of the dial that related to the band in use. In practice, however, pressing this control and tuning at the same time is quite awkward. The 50G, like most battery sets of the era, used a “C” bias battery to provide the necessary operating conditions for the valves. In this receiver, a 4.5V 761 battery is used with taps at -3V and -1.5V. The -4.5V rail is used to bias the 1L5G only. For some strange reason, when the set is switched on, a 600W resistor is placed across the battery, giving a cur- rent drain of 7.5mA for no good reason that I can think of. By contrast, the vibrator-powered version (the 50GV) does not have a resistor loading the bias battery. In fact, bias batteries usually have no load placed on them and are simply there to provide the bias voltages. They will last for years when used that way but not in this receiver. A bias of -1.5V is applied to the grid of the 1K7G, while the -3V line is connected to the 1C7G’s signal input grid via a resistor when the set is on shortwave (instead of the AGC control voltage that’s applied on the broadcast band). Conversely, the second IF stage has -3V applied to it when the set is tuned to the broadcast band but has AGC applied to it when on shortwave. Dismantling the receiver This set is easy to dismantle. First, the five knobs are removed, followed by five screws which hold the back on and four screws that attach the chassis to the cabinet. Fortunately, the speaker and dial-scale are all attached to the chassis, which makes the job simple April 2006  97 Photo Gallery: AWA 516MZ 4-Valve Receiver Manufactured by AWA, Sydney, the AWA 516MZ series was introduced in 1947 and continued in production until 1949. These 4-valve mains-operated receivers were available in a number of different colours, with the brownwhite mottled set shown here being one of the less common. The valve line-up was as follows: 6A8-G frequency changer; 6G8-G reflexed IF amplifier/first audio amplifier/detector/AVC rectifier; 6V6-GT audio output; and 5Y3-GT rectifier. Photo: Historical Radio Society of Australia, Inc. – and it makes service and alignment so much easier. Once the chassis is out, the components on top are relatively easy to access. Similarly, access to the under-chassis components is quite reasonable, although access around the wave-change switch is a little tight. It appears that no components have been replaced during the life of the set and it has had very little use over the years. In fact, this set has been very well looked after and has obviously been stored in ideal conditions during the last 66 years. Even the loudspeaker is in pristine condition, which is rare in a set of this age. Restoration The cabinet will not need anything doing to it as it has already been polished and is looking good. There are a couple of small holes in the speaker cloth though and replacement may be the only real answer here. More seriously, the wiring in the battery loom has perished and if the set is to be fully restored to working condition, this will also need to be replaced. I faced a similar dilemma with the 98  Silicon Chip battery leads when I restored an AWA battery receiver last year. In that case, I was unable to slip the woven brown sheath off the wires, probably because the perished wiring had adhered to the inside of the sheath. Whether Brian will be able to find a way to remove the sleeve in this case remains to be seen. Certainly, the set cannot be made to operate unless the leads are replaced. The 2V battery clips also need replacing, as corrosion is quite evident on them. The resistors and capacitors will have to be checked in the usual manner. Usually, the resistors are relatively close to their nominal values and don’t need replacement. The capacitors are a different story – they may be close to their nominal values but the paper units are likely to have high leakage (low resistance). As a matter of routine, I would replace the audio coupling capacitors (C30 and C34) and the AGC bypass capacitors (C5, C7 & C25) without hesitation but other paper capacitors may also need replacing. Keeping up appearances If this set is to be kept in as near to original as possible, it’s essential to preserve its appearance. Although the capacitors can be replaced with the newer yellow-coloured polyesters on the HT lines and with low-voltage disc ceramic types on the AGC line, these will not look authentic. The solution to this problem is to carefully remove the “insides” of the defective capacitors and fit the newer more compact capacitors inside the casings. The ends of the old capacitor cases can then be resealed using the wax or pitch-like material that was originally used. From past experience, most (if not all) the valves will be in good order, although occasionally a new valve will be required. The alignment should also be reasonably close to the original settings but should still be checked. There are six IF adjustments on 460kHz. The front end is aligned at 600kHz and 1400kHz on the broadcast band and at 18 metres on the shortwave band. Note that the alignment on shortwave in this set, like most others of the era, is very rough and ready. The set is aligned at 18 metres which means that the local oscillator and the aerial coil are aligned for best performance at this wavelength (which equates to 16.666667MHz). However, there is no provision for tracking or accurately aligning the tuned circuits on shortwave. Instead, the way to use such sets is to mark where your favourite shortwave stations appear on the dial and not rely on the dial calibrations, as they may be up to 1MHz out. Additionally, at the 16-metre end of the dial, many stations will be heard at two spots on the dial, some 920kHz apart (ie, twice the intermediate frequency). Summary This receiver has a few design quirks that, although unusual, do not adversely affect its operation. However, when the set was being designed, I suspect that problems were encountered with the oscillator dropping out on both the shortwave and broadcast bands. To overcome this problem, padder feedback capacitors were fitted to both oscillator coils. The relevant components are C12 & C13 on the broadcast band and C16 & C17 on the shortwave band. In summary, this 50G receiver is a very worthwhile set to have in any SC collection. siliconchip.com.au