Fullscreen HTML5Med Res (??MB)HTML4

Vintage Radio The Imperial Japanese Army (IJA) Chi Receiver By Ian Batty Hats off to the fossicker who asked me to look at this unexpected treasure: a “Chi” ground receiver designed and made in Japan during the Pacific Campaign of World War II. I have previously mentioned that unique class of collectors – the ones who discover and work to preserve items most of us would pass by, or never even dream of finding. It’s thanks to them that I can document this rare find. I must also thank the founder of the Yokohama WWII Japanese Military Radio Museum, Takashi Doi, for providing the circuit diagram and background information (see www.yokohamaradiomuseum.com). Before we get to the Chi, first we must look at the landmark HRO design by the National Radio Company of Malden, MA, USA (not to be confused with National Panasonic of Japan). Collectors of communications receivers will know of it. Its seemingly-­ conservative design became the standard by which others were judged, and the standard to beat. 98 Silicon Chip It’s a design that inspired many other manufacturers: two RF stages, a converter with a separate local oscillator (LO), two intermediate frequency (IF) stages, a demodulator/AGC/first audio stage and an audio output stage. Looking at the converter stage, those of us used to multi-grid or multi-­ section converters (pentagrids or triode-­ hexodes) might wonder why the HRO used a simple pentode converter with a separate local oscillator. The HRO was first advertised in 1934, only one year after the patent was awarded for the pentagrid. While this single-tube converter worked adequately at broadcast frequencies, it was noisy, and its performance at higher frequencies was poor. Improved converters such as the triode-hexode would not be announced until 1935. Given National’s prominence as a Australia's electronics magazine supplier of top-quality receivers, and the lead time from design to release, James Millen, Herbert Hoover Jr and Howard Morgan would naturally incorporate the well-known, reliable pentode mixer into the HRO. Hoover and Morgan, designers of the electronics, opted for LO injection to the screen grid. In common with all other multiplicative mixers, this pushed the valve’s electron stream to cut off at the most negative part of the LO’s signal. This Class-B operation is vital to the superhet’s converter action. Our own Kingsley AR7 uses a similar design overall but substitutes the triode-hexode 6K8/6K8G (using an internal LO) as a converter stage. By the way, Hoover set up a lab in his garage, employing Howard Morgan from Western Electric Co and a few of his technicians to develop the siliconchip.com.au The rear view of the Chi Mark 1 chassis. From left-to-right are the first IF transformer, IF1 (#52, 6D6), second IF transformer, IF2 (#61, 6D6), demodulator/ AGC stage (#74, 6B7) and audio output valve (#104, 6C6). new receiver circuitry. It’s a tradition repeated by Bill Hewlett and David Packard in 1938, revived almost four decades later by Steve Wozniak and Steve Jobs of Apple fame. Similar designs, with two RF stages, were also used in the MN-26, AN/ ARN-6 and AN/ARN-7 aircraft radio compass receivers. The Chi (地) The Director of the Yokohama WWII Japanese Military Radio Museum kindly sent me the following description: In 1939, the Imperial Army formalised the Chi Mark 1-4 Radio Sets as the new ground-use radio equipment for the Air Force under the 4th formalisation work. The name Chi (ground) denotes ground-based anti-aircraft use. Chi Transmitter Mark 1 The Chi Mark 1 transmitter’s output power was 1kW (A1/CW). The companion receiver was a superheterodyne type, described below. The receiver was known as the Chi Mark 1 Radio Set/Receiver. The full name in (pre-WW2) Japanese was written as: 地一號受信機（に型）接續要圖 The receiver covered 140-13,350kHz using eight plug-in coil sets. An improved version was quickly introduced, covering 140-20,200kHz with nine coil sets. The receiver constituted the topof-the-line radio equipment for the Army’s field aviation units. But they were very laborious to manufacture, entirely unsuitable for mass siliconchip.com.au production, and expensive. Soon after the outbreak of the Pacific War, a large number of receivers were required for operations such as ground-to-air and base-to-base communications and intercepting enemy communications. The introduction of high-performance general-purpose receivers was requested. For this reason, the Mark 1 radio set/receiver was greatly simplified and made suitable for mass production as the Chi Mark 1 version. There was no significant difference in performance between the two receivers, and while the weight of the Chi Mark 1 radio set/receiver was 17kg, the revised Chi Mark 1 receiver was much lighter at 13kg. The set is a superheterodyne fitted with a beat frequency oscillator (BFO), automatic gain control (AGC), two stages of RF amplification, two stages of IF amplification and two stages of AF amplification. The receiving frequency covers 140~20,000kHz in nine bands, using plug-in coil sets. Depending on the frequency range, the IF is either 65kHz (receiving frequency 140~1,500kHz) or 450kHz (receiving frequency 1,500~20,000kHz), although some sources say that it should be 456kHz. The IF is changed by swapping four internally-located IF units (first IF, second IF, final IF pair, BFO). The set features a narrow-band crystal filter for the 450kHz IF, which is inoperative for the 65kHz IF. The entire design is similar to the HRO but with notable differences explained below. Tuning dials One of the HRO’s outstanding features was its patented precision dial, quoted as being the equivalent of a ‘four-foot [122cm] slide rule’. This was repeated on the AR7, but one wonders how useful it was. Ray Robinson’s AR7 review is worth reading on the matter (www.tuberadio.com/robinson/ museum/AR7/). The HRO’s calibration reportedly demanded four hours to make up the calibration charts for all four coil boxes. Calibration readings were transcribed to an individual printed scale for each coil box. Unlike the HRO and the AR7, the Chi has a simple 0-100 dial, with (like the HRO/AR7) a hand-drawn 160 × 20mm calibration chart for each coil box. The calibration chart for the Chi Mark 1 receiver. Note that the model number on this chart (40757) is different from the front panel (40780). Australia's electronics magazine October 2023  99 For the Chi, the accepted visual-­ reading accuracy of plus or minus half an intermediate division gives an accuracy of about ±15kHz in the 2.5~5MHz range. The HRO and AR7 used similar hand-drawn scales, so their precision vernier dials may not have contributed any greater indication accuracy than the Chi’s simple 0-100 dial. Circuit description The circuit (Fig.1) simply numbers components in order, similar to our Astor sets. I have kept the original numbering for consistency. The circuit supplied by the Yokohama Museum was happily clear, with all notations readable, although I have redrawn it for greater clarity. I have also redrawn the demodulator stage for ease of interpretation and description. The antenna circuit, comprising coil box sets #4a/#4b, is tuned by the first section of the four-gang tuning capacitor (#6, #17, #30 & #39). Antenna selector switch #2 connects directly to antenna socket #1a (short antenna), via matching capacitor #121 (long antenna) or to ground. There’s also a direct connection to the first RF amplifier grid via socket #1b and capacitor #120. The two RF stages are similar to those of the HRO. Both valves in the Chi are remote cutoff UX6-based UZ-6D6s, similar to the later octal 6U7. UZ is a Japanese coding; in this case, it refers to a valve with a standard longpin six-pin base. Both RF stages have AGC applied, the first (confusingly designated RF2) via 500kW resistor #23 and the second (designated RF1) via 500kW resistor #18. Bypassing is done by 10nF capacitors #7 and #19. The first RF stage operates with fixed bias derived across 300W resistor #9, bypassed by 10nF capacitor #10. The second RF stage cathode returns to ground via 300W cathode bias resistor #21 (bypassed by 10nF capacitor #22), then via the set’s 10kW RF/IF gain control potentiometer, #91. This pot also controls both IF amplifiers. The 6D6 (and 6C6) are ‘triple-grid’ amplifiers, with the suppressor grid bought out to its own pin connection on the six-pin base and wired externally to the cathode. The first RF has its own screen supply via 100kW resistor #11, bypassed by 10nF capacitor #12. The second RF shares a common screen supply with both IF amplifiers, individually bypassed by 10nF capacitor #25. That supply is derived from a voltage divider of resistors #89 (30kW) and #90 (50kW) plus RF/ IF gain pot #91 (10kW). Inductors #93 and #94 provide RF decoupling along with bypass capacitors #25, #55 and #64 (all 10nF). Making the RF/IF gain control part of a voltage divider gives more predictable gain control than the simpler cathode-circuit-only alternative. The RF amplifiers drive coil box RF transformers #15a/#15b (first RF) and #28a/#28b (second RF) with untuned primaries and tuned secondaries. Each RF amplifier is decoupled from HT by 3kW resistors and 10nF capacitors (#11/#14 and #26/#27). Unusually, the antenna and RF coil boxes only contain inductors; there are no internal trimmer capacitors. Frequency alignment for coils #4b, #15b and #28b (antenna, RF interstage and mixer grid) is by individual variable capacitors (#5, #16 and #29). These are all mounted on the front panel and allow individual adjustments of their circuits. Notice that these capacitors are drawn as variable (operator-­adjustable) and not preset (workshop-­adjustable). Given the Chi’s intended use, from military command centres to battlefield deployment, and the difficulty of guaranteeing alignment in such a wide range of environments, it made sense to give trained operators the ability to optimise front-end alignment in any situation. It can also be confusing; more on that later. The Chi’s ‘all-tuneable’ design may highlight a difference between the US military and the IJA. The US Army enlisted tens of millions, was able to train and assign many for support roles such as radio technicians, and could afford to set up local depots and repair shops close to (or on) battlefields. The IJA, by contrast, was engaged in rapid forward offensives until about late 1943, when the tide of war turned against them. Troops in forward deployments often had little in terms of advanced technical support. The military demand of ‘work first time, work all the time, work anywhere’ Fig.1: a redrawn circuit diagram of the Chi Mark 1 receiver. The scale is unfortunately a bit small but that’s necessary to get everything into the available space. There are nine valves shown here; the tenth is a rectifier in the power supply (see Fig.3). 100 Silicon Chip Australia's electronics magazine siliconchip.com.au was met by giving operators the most flexible equipment possible. The Chi uses a pentode mixer, but unlike the HRO, it uses suppressor injection. As the suppressor was designed to correct the secondary emission problem in tetrodes, it has a pretty open spiral construction. This means that it needs considerable negative bias to cut a valve off. In the case of the famous EF50, suppressor cutoff demands some -50V of bias. The mixer valve (#31) is a sharp cutoff UX6-based 6C6, identical to the later octal 6J7. Because of its sharp cutoff characteristics, it does not have AGC applied and is not affected by the RF/IF gain control. This stage works with very low supply voltages, only about 20V. This had me checking and double-­ checking my measurements. Remember that mixer action relies on cutoff for the most negative part of the LO signal. Such low voltages would ensure that the suppressor-injected LO signal does drive the valve into the cutoff region as required. The screen grid has a much greater effect on anode current; the HRO, using screen injection, could apply more normal supply voltages to its pentode mixer and still ensure the required anode current cutoff. As noted above, the AR7, coming some years later when high-performance triode-hexodes were available, solved the problem by using the 6K8/6K8G. The LO (#42) also uses a 6C6 in a cathode-coupled Hartley circuit. siliconchip.com.au Although the valve is supplied with the usual anode and screen voltages, these are both bypassed to signal ground. Feedback is from the cathode to the grid. As the circuit is a cathode follower with feedback, there is zero phase shift, and the voltage gain is less than unity. That means the circuit can use a single tuned winding with no phase inversion, and the tuned circuit gives a voltage step-up from the cathode to the grid to establish oscillation. This circuit became the preferred design in 6SA7/6BE6 pentagrid converter circuits. The selected coil box’s coil (#38a) is tuned by the LO section of the tuning gang, #39. The LO coil box does contain a workshop-adjustable trimmer (#38c), as the LO’s accuracy determines the set’s frequency calibration. There is no operator adjustment for LO calibration. Each LO coil box contains a padder (#38b) to ensure the LO tracks by the IF value above the incoming signal. Any minor tracking errors between LO and the antenna/RF circuits are corrected by the operator’s use of the three manual trimmers in the antenna/RF stages. The mixer feeds the first IF valve via first IF transformer #49b~#49e, tuned for 450kHz. The transformer’s tuned primary and secondary use fixed capacitors and inductance tuning. The first IF amplifier (#52), a remote cutoff 6D6, is biased by fixed 300W resistor #53, bypassed by 10nF capacitor #54, Australia's electronics magazine and then connects to ground via the common 10kW RF/IF gain potentiometer, #91. The second IF amplifier’s (#61) biasing and bypassing are similar. The second IF feeds the input section of the final IF’s bandpass assembly #67g~#67k. The signal is then fed to the switchable crystal filter #68a~#68c, described more fully below. The signal from the crystal filter passes to the output section of the final IF bandpass filter, #67m~#67q. Its output secondary feeds the demodulator (lower) diode in #74, the demodulator/ AGC/first audio valve, a Ut-6B7 (Ut is another Japanese prefix). An IF signal is fed, via 1nF capacitor #69, to the Ut-6B7’s (upper) AGC rectifier diode. AGC is developed across 500kW resistor #86, filtered by 500kW resistor #87 and 10nF capacitor #88, and applied to the two RF and two IF stages. For A1/CW operation, the AGC is disabled by one section of CW/ AM switch #101. A1/CW operation is described below. The 6B7’s cathode return comprises resistors #79 (1kW) and #78 (3kW). A cathode bias of around 2V is developed across resistor #79, with the grid returning to the junction of resistors #79 and #78 via 500kW resistor #75. The demodulator diode returns to the 6B7 cathode. This means it has no bias and will respond to all IF signals. Its cathode current develops another 5.7V across the bottom cathode resistor, #78. Since the AGC diode returns to ground, the drop across #78 is also October 2023  101 (see Fig.3). The supply included an AC voltmeter, allowing operators to set the correct mains voltage. For battery operation, the Chi used a motor-generator set, also known as a ‘dynamotor’ or ‘genemotor’, to convert the low DC voltage from a battery into the required ~200V DC HT voltage. It is basically a DC motor driving a generator. In this case, it is a conventional 6V DC to 200V DC unit with the usual extensive primary and secondary filtering (also shown in Fig.3). Getting it going The top view of the chassis (right-to-left), primarily showing the 1st RF amp, 2nd RF amp, tuning, gear drive, mixer local oscillator tuning and crystal filter. the AGC delay voltage. At around 6V, it seems high, but this radio was designed for weak-signal performance, so it needs such a delay to prevent gain reduction for microvolt-level signals. Demodulated audio is fed via 10nF capacitor #70 and 500kW resistor #77 to the 6B7’s pentode grid, which returns to the cathode bias point (#79/#78) via 500kW grid return resistor #75. The cathode resistors are bypassed by 10nF capacitor #73; other minor components in this part of the circuit include #71, #72 and #76. The 6B7’s screen is supplied via resistors #82 (two 100kW resistors in series) and #81, bypassed by 10nF capacitor #80. The audio signal is developed across load resistor #85 (decoupled by #84 and #83) and fed to the output stage grid via 1nF capacitor #102. Output valve #104, a 6C6, drives output transformer #112. It feeds the two headphone jacks, #114, and its screen is supplied via 100kW resistor #107, with 10nF bypass #108. BFO and crystal filter For A1/Morse code/continuous wave (CW) reception, the set uses a beat frequency oscillator (BFO), built around another 6D6 (#97). This produces a tuneable signal that can be offset from the received IF signal, making unmodulated transmissions audible – 1kHz is a common choice. It can also resolve single sideband (SSB) voice signals. 102 Silicon Chip It’s a cathode-coupled Hartley oscillator, and its output is fed to the demodulator diode. The diode acts as an additive mixer, producing a tone with a frequency that’s the difference between the IF signal frequency and the BFO frequency. The main IF channel’s bandwidth is around ±1.8kHz. This is necessary for voice reception, but a narrower bandwidth can be used for CW. Narrowing the bandwidth has the advantage of improving the signal-to-noise ratio, as a channel’s noise is proportional to the square root of its bandwidth. The crystal filter (#68a~c) exploits the very high Q of a quartz crystal (20,000+). This implies a very narrow filter bandwidth. In operation, crystal #68b is shunted by variable capacitor #68c, allowing the filter bandwidth to be adjusted. For voice reception, the filter is taken out of circuit by switch #68a. Regrettably, this set’s crystal was marked 400kHz, rather than the required value of 450kHz. While this prevented the filter from being tested, it seemed to be an original fitting – it was certainly in the correct holder. A factory error? We’ll probably never know. Power supply The Chi needs 6V (AC or DC) for the heaters and +200V DC for HT. The AC mains supply operated from 80~120V AC or 200~240V AC input, using a KX-80 in a conventional fullwave circuit with a two-section pi filter Australia's electronics magazine I took charge of this set in early 2019 but didn’t have much luck getting it going, so I returned it to the owner. He contacted fellow HRSA member Brian Goldsmith and asked him to look at it. Brian found numerous problems. Firstly, some valves were not functioning correctly. Brian resoldered all of their bases, and they came back good. Many of the 450kHz coils (IFs and BFO) were loose, so he fixed them in place using paraffin wax rather than using superglue or some kind of resin. This holds them in place but permits later disassembly if needed. The tuning system comprised two dual-gang variable capacitors linked by the central gear drive and a drive sleeve. The left-side sleeve was loose, creating backlash when tuning, and the locating bearings at each end of the two-gang sections were also loose, so all moving plates were not correctly located relative to the fixed plates. Once repaired, the tuning mechanism worked perfectly. The audio output transformer was faulty, so he replaced it with the closest match available. He then performed an alignment, only to find that some of the ferrite adjusting cores were loose. If, after doing the alignment, you turn the set upside-down and the alignment changes, something is loose inside the coil cans. The crystal in the crystal filter was confirmed as 400kHz, and Brian could not find a replacement. Finally, the BFO was inoperative. The circuit resistances and voltages appeared correct, and the fault could not be fixed. The radio came back to me a bit later. Once on the bench, I confirmed all the valves as being good. A quick check of DC voltages showed them as expected, so it was on to signal tracing siliconchip.com.au and testing. The IF and audio sections worked as expected, but the RF section was dead. After some faffing about, I discovered the three manual trimmers (Antenna/first RF/second RF). Adjusting these correctly brought the set to life. It was sensitive, but not as good as I expected. I went over the IF again and found I needed a lot of signal at the final IF grid. Checking the last IF, I adjusted the secondary core to each end of its travel without finding any peak. Removing the assembly from its can, I found the primary peaking at just on 500kHz – it was well above the correct figure of 450kHz, due to being out of the aluminium can with its capacitive and inductive effects. This indicated that the primary was OK and hinted that the secondary would have to peak around the same point, about 500kHz. I couldn’t easily get the secondary to peak with my grid dip oscillator on its 500kHz~1.5MHz range. Connecting it to a signal generator and oscilloscope showed why – it peaked at around 350kHz! 100pF tuning capacitor #67q measured high at around 120pF, so I put in a new 100pF capacitor. The coil would still not reach the 500kHz that was needed from the can. I ended up with only 47pF for #67q. Why? The protective wax may have contributed extra capacitance with age. Whatever the cause, reassembling and reinstalling the final IF, then aligning it, brought the set to life. Although noisy, it could easily respond to signals around 1μV at 5MHz. The BFO superpower As described above, the BFO is a simple cathode-coupled Hartley oscillator with electron coupling for the output to the demodulator. It wasn’t working even though the valve tested good. The DC voltages were also acceptable, and the tuning coil resistances looked fine. I disassembled the coil can and checked again. In desperation, I disconnected and measured the internal 150pF tuning capacitor, which came up at 148.5pF. While doing this, one lead on the 50kW grid leak resistor broke off close to the resistor body. The lead connecting the two capacitors to the top of the coil also parted as I worked on the assembly. The resistor itself measured 54kW. I repaired the broken leads and, after reassembly and adjustment, the BFO worked perfectly. I suspect that one of the parted leads had been minutely fractured, and that had been the problem – I’d certainly not seen any evidence of clean breakages. BFOs were originally designed to make unmodulated (CW) transmissions more detectable. With no modulation, all you hear (maybe!) is a series of clicks as the carrier cuts in and drops off. The BFO is essential to the intelligibility of the widely-used SSB communication mode, replacing the carrier that was removed by the transmitter. What’s not so obvious is the increase in sensitivity that the BFO can give. In the Chi, I could easily detect an unmodulated signal of only 200nV at 9MHz. It was usable but noisy. Such a signal would likely be below the general noise floor that bedevils all HF communication. So it’s an impressive superpower, if you can actually use it. How good is it? Its absolute sensitivity, for 1mW The underside of the chassis is neatly presented with nearly every (!) component numbered as per the circuit diagram shown in Fig.1. The chassis provides ample room for each component, making servicing a breeze. siliconchip.com.au Australia's electronics magazine October 2023  103 into headphones, ranges from 12.5μV at 4.5MHz and 9.3μV at 2.5MHz, to 0.45μV at 9MHz and 1.1μV at 5MHz. The signal-plus-noise-to-noise ratios (S+N:N) are 20dB at 4.5MHz and 2.5MHz, but only 2dB at 9MHz and 3dB at 5MHz. Dial calibration was within about 1% across the bands. Opening the 2.5~5MHz LO can showed that the calibrating trimmer had probably not been touched since decommissioning. That’s impressive for equipment that has likely been idle for over 70 years. It’s also a reminder that it is worth attempting to restore and preserve all well-built equipment, whether military or civilian. I’ve plotted the dial calibrations and signal performances in Fig.2. The well-known calculation for noise figure resolves handily for a 50W source: a noise voltage of 1nV multiplied by the square root of system bandwidth in hertz. Even a perfectly noiseless receiver with a bandwidth of 3.7kHz would have a noise floor of about 60nV. Valves such as the 6D6 have equivalent noise resistances in the kilohms range. While a full discussion is outside the scope of this article, it’s easy to see why signals much less than 10μV will necessarily have poor signal-tonoise ratios. Having a super-sensitive set is one thing, but there are two reservations. Firstly, atmospheric noise at MF/HF (300kHz to 30MHz) can easily reach the equivalent of 10μV. When exposed to such a high noise floor, the most sensitive receiver won’t be much better than any good set. Secondly, a raw figure of 1μV is pretty useless if the set’s S+N:N ratio Selectivity/ Xtal Filter On/Off BFO Tune AM/CW (A3/A1) 2nd RF/Converter Tuning 1st RF Antenna Tuning Tuning Antenna Matching RF/IF Gain Tuning Headphone Sockets Off/Standby/On Antenna Input Ground Direct Input Plug-in Coil Box (2500-4700kHz) A labelled shot of the front panel. Judging from the metallic tag, this radio was produced by the Anritsu Corporation. means that signals are unintelligible due to high internal noise. Ordinary pentodes are pretty noisy, and the noise generated in the first stage will determine any receiver’s ultimate sensitivity. The IF bandwidth is about ±1.85kHz at -3dB and ±14kHz at -60dB. Audio response from the antenna to the headphones is 500~3500Hz at -3dB, with a rapid roll-off below 500Hz. The set is intended for headphone use, so all tests were done at 1mW output. It can deliver around 60mW maximum, enough for a loudspeaker in quiet settings. AGC action is complicated by the RF/IF gain control setting. Generally, a 6dB output rise happened with only a 20dB input rise; that is certainly not as good as common domestic superhets. However, it needed over 100mV to overload at full gain. In practice, very powerful signals can be managed by a combination of RF/IF gain control and detuning one or more of the RF stages. The two RF stages give good IF and image rejection. IF rejection at 5MHz was around 93dB and image rejection around 75dB. Evaluation The set’s build quality is excellent. Despite its complex design, getting to all the test points was easy. Virtually every component is individually branded with its circuit number. This made locating components very simple, in contrast to the more common method where parts only carry their electrical values that are often either difficult to read or obscured by being mounted upside down. Under my RMA criteria, it gets a 10 for maintainability. The circuit diagram is excellent, and the parts list denotes not only most components’ electrical values but also their function in the circuit. #10, for example, is fully described as the “First high-­ frequency amplifier tube cathode capacitor (0.01μF)”. Such descriptions are valuable in the workshop – you can find out what Fig.2: a plot of the dial calibrations and signal performances at various frequencies. 104 Silicon Chip Australia's electronics magazine siliconchip.com.au 號 b c           a b c d e f         a b c d e f         a b     3    4 F 1 5  E  2   3  2     1 6 5 6 E 6    1 +6V +6V  2  –  3 4     1µF       10nF    10nF      0.1mH 1  Mechanical Genemotor 10nF 2 1µF3 4  +200V +6V     6mH +200V +200V 1  3 +6V +6V  – 4  – 2   2 x 10µF  100mA 100V 交流電源 100200V 2 5060C/S 3  1A  KX-80  +20V   1µF      1 2 3 4     +200V +6V 1 7  2   3 4  +2 +6   +20V 0 4     1 4   100V 交流電源 100200V 2 5060C/S 3  30H 30H  200V  1  200V 10µF   100mA 6mH 3.5H 3.5H 10µF  150V AC Generator 100~200V 50~60Hz/s  0.1mH  30kΩ    0 -20V 8 -20V Fig.3: the power supply section of the ‘original’ circuit diagram, courtesy of Takashi Doi (Yokohama Radio Museum; www. 名 稱 諸 元 番號 名 circuitry 稱 元 番號 名 稱 (#208) and 諸 元 番號 rectifier 名 valve 稱 yokohamaradiomuseum.com). This uses a 諸mechanical genemotor KX80 (#310). 諸 元 1μF c 欠  音量調整器側路蓄電器  電池接栓受 第一局部發振管同調直列蓄電器 番 諸 元 番號 名 稱 第一局部發振管同調並列蓄電器 b 第一局部發振管同調直列蓄電器 第一局部發振管同調蓄電器 c 第一局部發振管同調並列蓄電器 50kΩ (D-05型) 第一局部發振管格子抵抗器  第一局部發振管同調蓄電器 0.00025μF 第一局部發振管格子蓄電器  第一局部發振管格子抵抗器 UZ-6C6 第一局部發振管  第一局部發振管格子蓄電器 3kΩ (D-05型) 第一局部發振管陽極直列抵抗器 甲  第一局部發振管 30kΩ (D-05型) 第一局部發振管遮蔽格子分圧抵抗器 甲  第一局部發振管陽極直列抵抗器 甲 100kΩ(D-05型) 第一局部發振管遮蔽格子分圧抵抗器 乙  第一局部發振管遮蔽格子分圧抵抗器 0.01μF甲 第一局部發振管陽極側路蓄電器 甲  第一局部發振管遮蔽格子分圧抵抗器 0.01μF 0.01μF乙 第一局部發振管陽極側路蓄電器 乙  第一局部發振管陽極側路蓄電器0.01μF UZ-6D6 甲 第一局部發振管遮蔽格子側路蓄電器  第一局部發振管陽極側路蓄電器 乙 300Ω (D-05型) 欠 番  第一局部發振管遮蔽格子側路蓄電器 0.01μF 變周管陽極同調蓄電器 番 100kΩ (D-05型) a 欠 變周管陽極同調線輪 b 變周管陽極同調蓄電器 0.01μF 第一中間周波増幅管格子同調線輪 3kΩ (D-05型) c 變周管陽極同調線輪 第一中間周波増幅管格子同調蓄電器 d 第一中間周波増幅管格子同調線輪 0.01μF 欠 番 e 第一中間周波増幅管格子同調蓄電器 第一中間周波増幅管格子直列抵抗器 500kΩ (D-05型) f 欠 番 第一中間周波増幅管格子側路蓄電器 0.01μF  第一中間周波増幅管格子直列抵抗器 第一中間周波増幅管 UZ-6D6  第一中間周波増幅管格子側路蓄電器 第一中間周波増幅管陰極直列抵抗器 300Ω (D-05型) 500kΩ (D-05型)  第一中間周波増幅管 第一中間周波増幅管陰極側路蓄電器 0.01μF X2  第一中間周波増幅管陰極直列抵抗器 0.01μF 第一中間周波増幅管遮蔽格子側路蓄電器 0.01μF  第一中間周波増幅管陰極側路蓄電器 UZ-6D6 第一中間周波増幅管陽極直列抵抗器 3kΩ (D-05型) 300Ω (D-05型)  第一中間周波増幅管遮蔽格子側路蓄電器 第一中間周波増幅管陽極側路蓄電器 0.01μF  第一中間周波増幅管陽極直列抵抗器 0.01μF 欠 番 500kΩ (D-05型)  第一中間周波増幅管陽極側路蓄電器 第一中間周波増幅管陽極同調蓄電器 a 欠 番 0.01μF 第一中間周波増幅管陽極同調線輪 b 第一中間周波増幅管陽極同調蓄電器 0.01μF 第二中間周波増幅管格子同調線輪 3kΩ (D-05型) c 第一中間周波増幅管陽極同調線輪 第二中間周波増幅管格子同調蓄電器 d 第二中間周波増幅管格子同調線輪 0.01μF 欠 番 e 第二中間周波増幅管格子同調蓄電器 第二中間周波増幅管格子直列抵抗器 500kΩ (D-05型) f 欠 番 第二中間周波増幅管格子側路蓄電器 0.01μF  第二中間周波増幅管格子直列抵抗器 第二中間周波増幅管 UZ-6D6  第二中間周波増幅管格子側路蓄電器 第二中間周波増幅管陰極直列抵抗器 300Ω (D-05型)  第二中間周波増幅管 UZ-6C6 第二中間周波増幅管陰極側路蓄電器 0.01μF 5kΩ (D-05型)  第二中間周波増幅管陰極直列抵抗器 第二中間周波増幅管遮蔽格子側路蓄電器 0.01μF  第二中間周波増幅管陰極側路蓄電器 0.01μF 第二中間周波増幅管陽極直列抵抗器 3kΩ (D-05型) 3kΩ (D-05型)  第二中間周波増幅管遮蔽格子側路蓄電器 第二中間周波増幅管陽極側路蓄電器 0.01μF  第二中間周波増幅管陽極直列抵抗器 0.01μF 欠 番 500kΩ 番 (D-05型)  第二中間周波増幅管陽極側路蓄電器 欠 番 100kΩ (D-05型) a 欠 b 欠 番 E F G d e f g h i j k l m n o p q r a b c                        諸 番元 番號 名 稱  欠 c 欠 番  欠 番 d 欠 番 a 欠 番 e 欠 番 b 第二中間周波増幅管陽極結合線輪 (二号) 50kΩ (D-05型) f 欠 番 c 水晶濾波器入力側同調線輪 (二号) g 甲 0.00025μF 第二中間周波増幅管陽極結合線輪 (二号) d 水晶濾波器入力側同調蓄電器 (二号) h UZ-6C6 水晶濾波器入力側同調線輪 (二号) e 欠 番 3kΩ (D-05型) i 乙 水晶濾波器入力側同調蓄電器 甲 (二号)  水晶濾波器入力側同調蓄電器 (二号) 30kΩ (D-05型) j 丙 欠 番  水晶濾波器入力側同調蓄電器 (二号) 100kΩ(D-05型) k 水晶濾波器入力側同調蓄電器 乙 (二号)  水晶濾波器平衡蓄電器 (二号) l 水晶濾波器入力側同調蓄電器 0.01μF 丙 (二号)  水晶濾波器出力結合蓄電器 (二号) m 水晶濾波器平衡蓄電器 0.01μF (二号)  水晶濾波器出力結合線輪 (二号) n 水晶濾波器出力結合蓄電器 0.01μF (二号)  Ut-6B7検波陽極同調線輪 o 水晶濾波器出力結合線輪 (二号)  Ut-6B7検波陽極同調蓄電器 p Ut-6B7検波陽極同調線輪  欠 番 q Ut-6B7検波陽極同調蓄電器  水晶濾波器轉換器 r 欠 番  水晶共振子 a 水晶濾波器轉換器  選擇度調整器 b 水晶共振子  Ut-6B7検波陽極結合蓄電器 0.001μF 500kΩ (D-05型) c 選擇度調整器  Ut-6B7検波陽極低周波結合蓄電器 0.01μF  Ut-6B7検波陽極結合蓄電器 500kΩ (D-05型)  0.01μF Ut-6B7検波陽極抵抗器  Ut-6B7検波陽極低周波結合蓄電器 UZ-6D6  Ut-6B7検波陽極側路蓄電器 0.00025μF  300Ω (D-05型) Ut-6B7検波陽極抵抗器  Ut-6B7陰極側路蓄電器 0.01μF  Ut-6B7検波陽極側路蓄電器 0.01μF X2  第二検波並第一低周波増幅管 Ut-6B7  Ut-6B7陰極側路蓄電器 0.01μF Ut-6B7格子抵抗器 500kΩ (D-05型)  3kΩ (D-05型)  第二検波並第一低周波増幅管 0.00025μF  Ut-6B7格子低周波側路蓄電器  Ut-6B7格子抵抗器 0.01μF Ut-6B7格子直列抵抗器 500kΩ (D-05型)   Ut-6B7格子低周波側路蓄電器3kΩ (D-05型)  Ut-6B7陰極直列抵抗器 甲  Ut-6B7格子直列抵抗器 Ut-6B7陰極直列抵抗器 乙 1kΩ (D-05型)   Ut-6B7陰極直列抵抗器 甲  Ut-6B7遮蔽格子側路蓄電器 0.01μF  Ut-6B7陰極直列抵抗器 乙 100kΩ (D-05型)  Ut-6B7格子分圧抵抗器 甲  Ut-6B7遮蔽格子側路蓄電器 Ut-6B7格子分圧抵抗器 乙 100kΩX2(D-05型)  1μF  Ut-6B7陽極側路蓄電器  Ut-6B7格子分圧抵抗器 甲  Ut-6B7格子分圧抵抗器 乙 3kΩ (D-05型) 500kΩ (D-05型) 甲 Ut-6B7陽極直列抵抗器  Ut-6B7陽極側路蓄電器 0.01μF Ut-6B7陽極直列抵抗器 乙 100kΩ (D-05型)  Ut-6B7陽極直列抵抗器 甲 500kΩ (D-05型) UZ-6D6 Ut-6B7検波陽極自動音量調整抵抗器 300Ω (D-05型)  Ut-6B7陽極直列抵抗器 乙 500kΩ (D-05型) Ut-6B7検波陽極自動音量調整濾波抵抗器  Ut-6B7検波陽極自動音量調整抵抗器 0.01μF Ut-6B7検波陽極自動音量調整側路蓄電器 0.01μF 0.01μF 遮蔽格子分圧抵抗器 甲  Ut-6B7検波陽極自動音量調整濾波抵抗器 30kΩ (D-2型) 3kΩ (D-05型)乙  Ut-6B7検波陽極自動音量調整側路蓄電器 遮蔽格子分圧抵抗器 50kΩ (D-2型)  遮蔽格子分圧抵抗器 甲 音量調整器0.01μF 10kΩ  遮蔽格子分圧抵抗器 乙  音量調整器 I H J a component does without having to read the manual. It’s also notable for having a minimal list of component values. RF bypass capacitors are overwhelmingly 10nF in value. Most resistors are 1kW, 3kW, 50kW or 100kW. Such a design adds to the Chi’s serviceability, as technicians only need to keep a small inventory of spare components for repair. Aside from the 7-pin 6B7 demodulator/AGC/first audio valve, it would be possible to put any 6-pin pentode valve in any 6-pin socket and have a working set. Conclusion There are very few of these exceptional radios still in existence, and this D E F G 諸 元 番號 名 稱 第二高周波増幅管遮蔽格子塞流線輪  音量調整器側路蓄電器 第一中間周波増幅管遮蔽格子塞流線輪 第二局部發振管同調線輪  第二高周波増幅管遮蔽格子塞流線輪  第一中間周波増幅管遮蔽格子塞流線輪 第二局部發振管同調蓄電器 a 第二局部發振管同調線輪 欠 番 b 第二局部發振管同調蓄電器 第二局部發振管格子蓄電器 c 欠 番 第二局部發振管格子抵抗器 50kΩ (D-05型) d 第二局部發振管格子蓄電器 音色調整器 e 第二局部發振管格子抵抗器 第二局部發振管 UZ-6C6  音色調整器 第二局部發振管陽極直列抵抗器 3kΩ (D-05型)  第二局部發振管 第二局部發振管陽極分圧抵抗器 甲 50kΩ (D-05型) 第二局部發振管陽極直列抵抗器  乙 第二局部發振管陽極圧抵抗器 500kΩ (D-05型)  第二局部發振管陽極分圧抵抗器 甲 電信電話轉換器 乙  第二局部發振管陽極圧抵抗器 0.001μF 第二低周波増幅管格子結合蓄電器  電信電話轉換器 第二低周波増幅管格子抵抗器 100kΩ (D-05型)  第二低周波増幅管格子結合蓄電器 第二低周波増幅管 UZ-6C6  第二低周波増幅管格子抵抗器1kΩ (D-05型) 第二低周波増幅管陰極直列抵抗器  第二低周波増幅管 欠 番  第二低周波増幅管陰極直列抵抗器 第二低周波増幅管遮蔽格子分圧抵抗器 100kΩ (D-05型)  欠 番 第二低周波増幅管遮蔽格子側路蓄電器 0.01μF  第二低周波増幅管遮蔽格子分圧抵抗器 0.001μF 第二低周波増幅管陽極直列抵抗器 3kΩ (D-05型)  第二低周波増幅管遮蔽格子側路蓄電器 0.01μF 第二低周波増幅管陽極側路蓄電器 1μF 第二低周波増幅管陽極直列抵抗器  500kΩ (D-05型) 第二局部發振管陽極側路蓄電器 0.01μF  第二低周波増幅管陽極側路蓄電器21 0.00025μF 低周波出力變成器  第二局部發振管陽極側路蓄電器 0.01μF 低周波出力變成器並列蓄電器 0.01μF X2  低周波出力變成器 Ut-6B7 受話器ジヤツク 500kΩ (D-05型)  低周波出力變成器並列蓄電器 第二局部發振管結合蓄電器  受話器ジヤツク 0.00025μF 電源開閉器 500kΩ (D-05型)  第二局部發振管結合蓄電器 受信機接栓受 3kΩ (D-05型)  電源開閉器 電圧測定口 1kΩ (D-05型)  受信機接栓受 空中線抵抗器 500kΩ (D-05型) 0.01μF 甲  電圧測定口 空中線結合蓄電器 100kΩ (D-05型) 空中線結合蓄電器 乙  空中線抵抗器 100kΩX2(D-05型)  空中線結合蓄電器 甲 1μF  空中線結合蓄電器 乙 3kΩ (D-05型) 100kΩ (D-05型) 500kΩ (D-05型) 500kΩ (D-05型) 0.01μF 30kΩ (D-2型) 50kΩ (D-2型) 10kΩ K L M is the only one I’ve personally seen, apart from Takashi Doi’s example in the Yokohama Museum. So if you see, or even hear of, a Chi that someone wants to dispose of, snap it up! Supplementary information Unlike the Chi Mark 1 radio set/ receiver, the Chi Mark 1 receiver does not have a control that changes the amplification level of the LF stage. For telephone (A3) reception with this receiver, the manual (RF/IF) gain adjustment should set the receiver operation to maximum gain so that the AGC will operate correctly. However, setting the RF/IF gain adjuster to maximum gain is difficult, as this produces excessive sound output. In H I J K 番號 名 稱 電源開閉器諸 元 1μF  低圧側高周波側路蓄電器 甲 電池接栓受 1μF  電源開閉器 低圧側高周波塞流線輪 甲 0.1mH  低圧側高周波側路蓄電器 甲 0.1mH 低圧側高周波塞流線輪 乙 低圧側高周波側路蓄電器  乙 低圧側高周波塞流線輪 甲 0.01μF 低圧側高周波側路蓄電器  丙 低圧側高周波塞流線輪 乙 0.01μF  低圧側高周波側路蓄電器 乙 直流變圧器 高圧側側路蓄電器 甲  低圧側高周波側路蓄電器 丙 0.01μF 50kΩ (D-05型) 高圧側側路蓄電器 乙  直流變圧器 0.01μF  高圧側側路蓄電器 甲 高圧ヒユーズ 100mA UZ-6C6  高圧側側路蓄電器 乙 高圧側高周波塞流線輪 甲 6mH 3kΩ (D-05型) 乙  高圧ヒユーズ 高圧側高周波塞流線輪 6mH 50kΩ (D-05型)  高圧側低周波側路蓄電器 甲 高圧側高周波塞流線輪 甲 1μF 500kΩ (D-05型)  高圧側高周波塞流線輪 乙 高圧側低周波塞流線輪 3.5H 高圧側低周波側路蓄電器 甲  高圧側低周波側路蓄電器 乙 10μF 0.001μF  高圧側低周波塞流線輪 高圧側低周波塞流線輪 3.5H 100kΩ (D-05型)  高圧側低周波側路蓄電器 丙 高圧側低周波側路蓄電器 乙 10μF UZ-6C6  高圧側低周波塞流線輪 受信機接栓受 1kΩ (D-05型)  高圧側低周波側路蓄電器 丙  受信機接栓受 100kΩ 番 (D-05型)  欠 0.01μF  交流電源接栓受 3kΩ (D-05型)  欠 番  電源開閉器 1μF  交流電源接栓受 1A  交流電源側ヒユーズ 0.01μF  電源開閉器  欠 番  交流電源側ヒユーズ  電圧轉換器 21 乙 5V 2A 6.3V 3A 0.01μF X2  欠 番 80-200V 240V  電源變圧器 X2 60mA  電圧轉換器 乙 150V  電圧計  電源變圧器  電圧計倍率器  電圧計 KX-80  整流管 100mA  整流管直流側ヒユーズ  電圧計倍率器  整流管  欠 番 500kΩ (D-05型) 甲  整流管直流側ヒユーズ 30H  高壓電源平滑線輪 番 30H  高壓電源平滑線輪 乙  欠  高壓電源平滑蓄電器 甲  高壓電源平滑線輪 甲 1μF  高壓電源平滑蓄電器 乙  高壓電源平滑線輪 乙 10μF  高壓電源平滑蓄電器 丙  高壓電源平滑蓄電器 甲 10μF 高壓電源平滑蓄電器 乙  30kΩX2 (D-2型)  整流管直流側並列抵抗器  高壓電源平滑蓄電器 丙  受信機接栓受  整流管直流側並列抵抗器  受信機接栓受 印ハ予備品又ハ材料ヲ有スルモノヲ示ス 諸 元                   practice, a workable RF/IF gain setting does not allow the AGC function to be fully utilised. For this reason, compared to Chi Mark 1 Radio Set/Receiver, this receiver does not give optimum performance when listening to A3 signals. Thanks go to: • Takashi Doi, founder of the Yokohama WWII Japanese Military Radio Museum (see their website – www. yokohamaradiomuseum.com). • Ray Gillett of the Historical Radio Society of Australia (HRSA) for the loan of this very rare radio. • Brian Goldsmith of the HRSA. • You can find more details on the Chi receiver (in Japanese) at: http:// SC minouta17.web.fc2.com/ 印ハ予備品又ハ材料ヲ有スルモノヲ示ス O N L M P N O Radio TV & Hobbies The Complete Collection on USB Every issue from April 1939 to March 1965 If you're into anything vintage it doesn't get any better than this scanned collection of every single issue of Radio & Hobbies, and Radio TV & Hobbies magazines before they became Electronics Australia. It provides an extraordinary insight into the innovations in radio and electronics from the start of WW2 to the early transistor era! PDF Download $50 SC2950: siliconchip.com.au/Shop/3/2950 USB + Download $70 SC6142: siliconchip.com.au/Shop/3/6142 Postage is $10 within Australia for the USB. See our website for overseas & express post rates. siliconchip.com.au Australia's electronics magazine October 2023  105 9 0 1μF 0.1mH 0.1mH 0.01μF 0.01μF 0.01μF 0.01μF 100mA 6mH 6mH 1μF 3.5H 10μF 3.5H 10μF 1 1A 80-200V 5V 2A 6.3V 240V X2 60 150V 2 KX-80 100mA 30H 30H 1μF 10μF 10μF 30kΩX2 (D-2型