A few receiver reviews

From its humble beginnings back in 2004, when I started publishing short evaluations of the shortwave receivers that were crossing my way, this page has steadily grown and now includes other kinds of radios as well. This reflects in the table of contents, but the ordering on the page is currently purely chronological, and you may find the occasional inconsistency due to the varying age.

By the way, should anyone feel like sending me radios for review, I am quite unlikely to resist. ;)

Station profile

Here is some information about my 'shack' as of 2014, so you have an idea of the conditions under which I usually test receivers.

• Location type: suburban
• Interference levels MW/SW: Varying, low to moderate to high
• External antennas: difficult
• Signal levels LW: medium
• Signal levels MW: medium (can be boosted to strong with loop in the evening), crowded
• Signal levels SW: medium to high
• Signal levels FM: medium, but the usual crowded band (known spots with high levels exist elsewhere)
• Preferred power source: Rechargeables, mains if internal power supply
• Additional antennas: Sony AN-LP1, homebrew MW loop 1m²
• User experience level: SWL since 1997, electrical engineering background

Sony ICF-7601

(Sample obtained used, some time in 2003.)

This is a ca. 15 year old pocketbook-sized analog portable with MW, various shortwave bands (most of them are dual conversion) and FM. I noted the following on my 7600 series page, comparing the 7601 to my trusty ICF-SW7600G:

• Sensitivity: The ICF-7601 is about as sensitive on shortwave as the 7600G, even a bit more sensitive under certain conditions, but read on. (The 7600G was tested with the same batteries once, made no real difference vs. the rechargeables it normally uses.) Could get Radio Australia on 13620 quite listenably shortly after 2200z on 2003-10-09 when the same signal was still stuck in the mud on the 7600G (only audible in SSB), which surprised me a bit. Later tests, however, suggested that the seeming disadvantage of the 7600G on higher bands is more related to antenna tuning – with the ADDX PRE-1 (which is nothing but a simple antenna tuner), the 7600G was at least equal (on 13 meters, in this case), if not a bit better. Still, not too shabby. Later I found that the 7601 could maintain its sensitivity advantage on 19m and 75m, even when the 7600G was used with the PRE-1. (Do keep in mind that analogs usually have noticeably cleaner oscillator signals than typical PLL synth receivers, thus featuring lower background noise. [Ed. note 2008: The conclusion is nonsense, phase noise would rather show itself as reciprocal mixing, i.e. reduced selectivity.]) All in all, the 7601 provides good sensitivity off the whip on the SW bands.
  On MW, it's pretty much a toss-up – the SW7600G is more sensitive, but also more noisy, which is why sometimes the 7601 seems just a tiny bit better (though that's not easy to judge given the rather different audio frequency responses). The AGC in the 7601 apparently runs out of gain a bit earlier, making you turn up the volume quite far and leading to somewhat more distorted audio on very weak stations that as such can still be picked up front end wise (noticed this on 13m, maybe that band is not peaked up perfectly on my sample). A minor annoyance of this receiver is that an external power supply will act as an antenna, potentially introducing all kinds of line noise or even overloading the poor thing.
• Selectivity: Can't keep up with the 7600G and its excellent 7 kHz (real) filter (-60dB/-6dB: ~2), but still sufficient to separate two stations 5 kHz apart halfway well. (Such as a station on 5960 kHz in ?Japanese? from a stronger one on 5955 around 0330z on 2003-10-10. Couldn't sleep. ;) Selectivity on FM isn't great, no wonder now that I've had the opportunity to look at the service manual – just one filter isn't going to cut much. (The FM front end is also very simple, basically the signal is fed directly, with no pre-amp, to an IC that does mixing to the FM IF, has stuff filtered by the external filter and then demodulates.)
• Images: Found no images of the 49m band on 60m, which is good. (However, I was wrongly assuming images would be coming from f + 910 kHz, while in fact they are from f - 910 kHz due to the spectrum inversion during the 1st mixing. I once found DW on 6075 overlaid by some noisy utility signal, which was not present on any of my other receivers with different concepts – now guess why.)
  On the SW1 band where the receiver works with single conversion, I have noticed images from strong stations on 75m (910 kHz below), my guesstimate for the image rejection would be 20 to 30 dB.
• Spurious/mixing products: On shortwave bands that are filled with strong stations the receiver suffers from ghost signals that behave like images normally do. With a service manual at hand and some thought, it will seem that these are 3rd order IM products (being strongest on bands with strong stations) and can be attributed to the 2nd mixer (it's the only one with a variable mixing frequency). On most bands you can at least hear hets that wander as you tune around, clearly a step down from the "digital" 7600s. In addition to that, the receiver can also overload so much that you can hear the same IM mud across the whole band – not quite sure where this comes from, maybe the 2nd mixer too. On the lower 31m band, I have noticed some FM bleedthrough, which appears to be a mixing product between the 20.4 MHz crystal's 4th harmonic (*ahem* – some primitive lowpass filtering would have killed that) and a strong local station on 92.6 MHz. 92.6 MHz - 81.6 MHz = 11.0 MHz, bingo.
• Stability: Can't really get the thing to drift much on SW2-10, even by warming up manually (literally ;). If you tune to a station, it pretty much stays there. Probably this is because the really high mixing frequencies (15.6 MHz to 32.35 MHz to mix the SW bands from 60m to 13m to 10.7 MHz) are generated with crystal oscillators (and those crystals really don't drift much, a quick check with the ICF-SW7600G showed that none of them was more than about 200 Hz off), leaving only the variable second mixing frequency of around 11155 kHz as a potential drifting candidate.
• Sound: Rather good for a receiver of this size, particularly with the tone switch in MUSIC position (the trick: bass expansion). Audio is a tad hissy; in AM, quite a bit of IF level hiss seems to be getting through, which can be attenuated considerably by using the tone switch.
  Over headphones, the rx delivers very good, if rather bassy, audio on shortwave stations in the "MUSIC" position of the tone switch. The ICF-SW7600 comes next, the SW7600G ends up last with its very mid-centric "communications-quality" audio.
• Tuning dial: Not off by more than 10-15 kHz in most cases on the SW (2-10) bands. Rather coarse for SW1, MW and FM. FM scale seems least accurate here.
• Coverage: All the SW broadcasting bands from 120m to 13m including 21m (22m) usually with plenty of space around them, very good for a broadcast-only analog. Only coverage on 19 meters ends at 15.65 MHz, with the band extending to 15.8 MHz these days. (Actually, none of the SW2-10 bands cover more than 600-650 kHz. This probably is the filter bandwidth on the 1st IF.) Something similar is to be observed on the MW band, which ends shortly above 1600 kHz (at least on my model, the US model apparently covers up to 1705 kHz).
• Construction: Noticeably cheaper on 7601 (price issue), but still decent. Had to loosen 4 screws (3 outside + 1 in battery compartment) to get the back off, one more (again, in the battery compartment) to get the front off (along with some fiddling). Small stand on back feels a bit flimsy. Tiny off switch not operated easily (perhaps a contact problem on mine). Volume slider could be easier to move, too. Backlash on the tuning knob is a few kHz, not particularly bothersome in most cases. Tuning itself is quite smooth.
• Battery life: Since the 7600A already ran around 40 hours on normal batteries I assumed the 7601 would do similarly well, and as I was told this really seems to be the case. Since I usually don't needlessly run down batteries (they are expensive after all), I did not measure battery rundown time. The radio still plays just fine on a mere 2.4 V, only the dual conversion shortwave bands are useless (they fade out at around 3 volts, but the rest still receives decently) and the LEDs are rather dim. Somewhat over 2 volts, the receiver will not turn off anymore, neither do the LEDs light. Under 2 volts, the audio volume drops noticeably, along with the S/N, but increased distortion is not noticeable even at only 1.5 volts. At very low voltages, the receiver exhibits considerable drift on FM. Finally the volume gets lower and lower until you can barely hear anything – time to change the batteries.

All in all, while the ICF-7601 is a somewhat mixed bag particularly when it comes to strong signal handling, it's a decent analog travel portable that still is a lot better than the el cheapo single conversion stuff you get new for its used price these days (I paid around EUR 20 plus shipping for mine). Some were supposedly not well aligned from the factory and thus not that sensitive, this doesn't seem to be the case here. In fact, mine is a bit too sensitive for what the mixers can handle. BTW, it really shines on 60 meters.

What struck me is that in some parts of the world (such as Azerbaijan or Malaysia) this receiver still seems to be available (as of Oct-2003), in spite of production officially having ceased ten years ago.

Here are some noise / buzz spectra recorded in 2008:

While the set is lacking the extensive audio filtering on the preceding ICF-7600A, good 10 kHz suppression and so-so 5 kHz suppression are achieved even in MUSIC. Things get about bulletproof (as far as travel portables slightly below a "digital" '7600 go) in NEWS.

↑ Back to table of contents

Sony ICF-SW12

(Sample purchased used, ca. 2003/2004. S/N 53681.)

The ICF-SW12 is a combination of an alarm clock with a little analog shortwave receiver, a cute little gadget if you ask me (though a bit overpriced when new – I got mine used at a bargain price). The receiver circuitry had been suspected to be closely related to the ICF-SW20/22's (which in turn is a slightly stripped-down version of the ICF-7601's) – but as a simple oscillator frequency search showed, it is just a single conversion unit. My performance observations:

• MW/SW sensitivity: With the smaller ferrite rod, the little ICF-SW12 cannot keep up with its elder brothers (ICF-7601 and ICF-SW7600G) on MW. On SW the SW12 proved to be a bit less sensitive and more hissy than the ICF-7601, though the difference was hard to judge given plugging in headphones leads to improved reception on the latter but not the former. Generally, the SW12 suffers from a considerable hand sensitivity, so better don't touch. (Touching the antenna affects antenna tuning like on the 7601, but holding the receiver itself is not quite without effect either. The bandswitch seems to be one cause for this.)
• Mixing products etc. on SW: With slightly lower sensitivity and a shorter whip combined with the band-related antenna tuning, mixing products are no more an issue than on the 7601; 31 meters in the evening is still critical. To my dismay, I discovered that 60 meters shows very strong 49 meter images – well, this no longer surprised me once I had figured out that it's a single conversion unit... (Technically, it's about on ICF-SW10/11 level.)
• MW/SW selectivity: Channel separation proved to be noticeably worse on the SW12 when compared to the 7601 (apparently it only has one two-element filter plus IFT). A tone control might be helpful here (particularly considering sound generally is rather bright) but is not present.
• FM reception: Nothing to write home about. Next to strong stations I could find areas where the TUNE LED was lit but no signal was to be heard. FM sound via headphones (mono only) is quite clear. It's a bit disappointing that a radio of this vintage (1997) and pricing does not include FM stereo.
• Sound: You cannot expect room-filling sound from such a tiny speaker. The ICF-7601 with its much bigger speaker and bass expansion is far superior, no questions asked. On headphones, bass expansion can be a bit disturbing, the ICF-SW12 with its only slightly bassy sound (or is the 7600G a bit weak in the bass?) might hold a slight advantage here. (Note: As I found out later, the 7600G is a bit weak in the bass department indeed.)
• Dial accuracy: The SW tuning scale contains freq marks mostly in 100 or 200 kHz distance with extremes of 50 on the lower edge of the lower bands and 350 kHz on 13 meters (50 kHz like on the ICF-7601 would be better, but you can usually live with it) and seems to be rather precise. Interestingly, band spread is not the same on all bands – on 60 and 49 meters it's more like 450 kHz, while 25 meters covers some 800 kHz and 19 meters even stretches over approx. 1 MHz.
• Operation / Misc. notes: Operation is pretty straightforward, I figured out how to set the clock and alarm pretty quickly even without having the manual at hand. Tuning with the little thumbwheel is a little tedious (it could stick out a bit further and be easier to turn). The receiver appears solid and well made (in Japan, BTW), though the battery compartment lid seems a bit flimsy. The clock display light, apparently provided by two LEDs, is a bit dim. On a side note, the scales could also use some illumination.

To sum it all up, if you're a passionate SWL and need a matching clock radio or if you want a compact clock radio with acceptable shortwave performance for travelling, the ICF-SW12 fits the bill. (If you can live without an integrated clock and alarm, better choose the ICF-SW20/22 with dual conversion on shortwave, which reportedly provides noticeably better performance. These models, however, lack the 22 and 60 meter bands, which is too bad particularly in case of 22m.) It's not what you'd use for serious DXing, but OK for getting the big guys, and of course the concept is pretty unique. However, it's not exactly a winner in terms of price/performance ratio.

(If more serious shortwave performance at comparable or even smaller dimensions is needed, the ICF-SW100 may be worth a look, or even the – now discontinued – ICF-SW07. Naturally these are also much more expensive. If it needn't be quite as small but still very compact, an ICF-SW35, Sangean ATS-606 or Degen DE1102 (under whatever brand name) may fit the bill at a reasonable price point.)

↑ Back to table of contents

Grundig Signal 700

A rather unknown, all in all pretty ordinary Grundig portable from the mid '70s. Since this is not a serious shortwave rig, criteria are ordered somewhat differently now.

• FM reception: A classic strong point of Grundigs, and this one is no exception. FM sensitivity is about on par with the 7600G (as far as this can be said in an area where this band is best described as rather crowded), and selectivity isn't too bad, though no better than on the ICF-7601 with its single rather wide ceramic filter. The filter skirts are apparently not exactly steep, which drops selectivity below ICF-7601 level at times. Maybe the IF filter circuits need to be realigned, that wouldn't be too surprising. Still, not bad for a rather ordinary 30 year old radio.
• Sound: Another "classic Grundig forté", this can only be described as excellent by portable standards. The oval wide-band speaker itself isn't small, and there is quite some bass expansion used, yielding a good frequency response with only a bit of mid-bass emphasis. This is definitely a radio for speaker playback. Given that, who cares that S/N isn't so good on phones and the bass emphasis does not do much good there either? FM demodulation doesn't go up as far in the highs as usual today, but you don't notice that on speaker playback anyway. There's a bit of hum when operated off mains (110/220V), this is not noticeable at normal listening volumes. The only speaker related weakness is a slight tendency to sibilance.
• LW/MW/SW1/SW2 reception: I did not evaluate LW performance, but looked at the rest. Selectivity is not particularly good, leaving you with hets in most cases. The AGC lets the volume drop far too early. (Both points are typical for many old radios.) In terms of MW sensitivity, the ICF-7601 soundly beats the Signal; the demodulation also seems less clear on weak signals on the latter. Sensitivity on the lower SW bands seemed quite good actually (virtually on par with the SW7600G on 49 meters), while tuning the upper ones is no fun not only because of the very inaccurate scales, but also the AGC action or rather lack thereof. During a test on 22 meters, even the little ICF-SW12 – which BTW also provides better selectivity – got a lot more stations (particularly, DW on 13780 absolutely couldn't be found on the Signal but came in quite understandably on the SW12 at the time of testing). Of course you get the usual images (single conversion) and also some mixing products on shortwave.
• Construction: A classic Grundig, constructed solidly and with attention to detail (take the sliding cover for the connectors, for example). It still works fine after all these years, doesn't that say a lot already? Only some contact cleaner may not hurt. It was apparently made in Grundig's factory in Portugal, BTW.

JFTR: S/N 53681.

A fun application for a radio like this with a tape/phono input is playing back online audio streams (hey, it still beats a crappy 2.1 system with the mids sucked out). The signals will have to be downmixed to mono, but at least with Winamp that is not a problem. Availability of DIN-to-RCA adapters might be one, depending on where you are located.

There is also an entry for this radio over at radiomuseum.org. It quotes 5 AM and 8 FM circuits, which is hardly exciting for a transistor set. The Signal 500, however, featured 7 AM and 10 FM circuits, just like the normal stationary radios of the day, and thus must have been better reception wise in spite of the smaller number. (Maybe this was a case of "circuitry recycling" to allow longer production runs, given there had been some models with the same circuitry in preceding years.)

What the number of circuits says? These comprise resonant circuits on both RF and IF level (including LO circuits), so this allows only a very rough estimate of the performance level. If the distribution is known (schematic or educated guesswork), you can say something about how good IF-level selectivity and RF-level sensitivity and overload might be. (For example, a Satellit 1000 features 13 FM circuits, with 2 in the front-end, 1 in the oscillator, 10 on IF level including 2 in the ratio detector. Of interest to us are 2 in the front-end and 8 on IF level.) Finally, you need numbers for receivers with known performance for comparison. (Good transistorized FM tuners from the 60s had like 15 or 16 circuits. My guess for a 16 circuit unit would be 10 for IF filtering and 3 for front-end filtering. A tube tuner would have like 12 circuits.) Please note that numbers for tube and transistor rigs are not directly comparable (why?), also it's not trivial to compare LC circuits (#LC) to ceramic filter elements (n) – it seems to be roughly #LC = 1.5...2 x n, with things depending on Q achieved (for LC) and the ceramic filter types which are not all created equal either.

↑ Back to table of contents

Philips (Magnavox) D1835

The D1835 is a pocketbook-sized portable (about 1 cm wider and 0.5 cm deeper than an ICF-7601, with disregard to the tuning knob) covering, at least in the /02 version I have, LW, MW, FM and the shortwave broadcast bands from 49m through 11m, including 21/22m. It was apparently introduced in 1984, probably being supposed to compete with the smaller Grundig Yacht Boys of the time; manufacturing was in Hong Kong. In the late 80s, it retailed for $70US list price, costing only about half as much as an ICF-7601.

The circuitry: two gang FM tuning (1x front end, 1x osc), FM mixer with a bipolar transistor (instead of a FET like on the Sony ICF-7600A), one FM IF filter (7600A had two) plus a tuned IFT, single conversion on shortwave with tuned front end (with LC and not crystal controlled oscillators that are a bit drifty; IF is 468 kHz), SW mixer with one FET; AM filtering with one 2-element ceramic plus apparently a tuned IFT in front of the AM detector; the used ICs are a µPC1018C for AM mixing (LW/MW mixer, SW 2nd mixer) and AM/FM IF work and a BA526 as mono audio amp.

The D1835 appears to be fairly unknown in spite of its decent performance (more on this later) and thus frequently does not get any bids at all when offered for auction here in Germany (which doesn't happen too infrequently). I was able to obtain one in very good condition in its original packaging with the accompanying soft carrying case and all the docs including the manual and a tinily printed small schematic sheet for next to nothing – quite a bargain compared to the used prices of said Grundig competition (which is not likely to perform much better, also being single conversion units). In fact, the RF circuitry is extremely similar to a Yacht Boy 650's (even including the AN7218S, apparently a µPC1018C equivalent) which, however, is 2 to 3 years older. (Feature wise, a Yacht Boy 300 would be about the equal of the D1835, minus band coverage – the Grundigs only cover 49, 41, 31, 25, 19 and 16 meters, but then came out a few years earlier as well. Grundig also offered models with clock functions and later even frequency counters, while Philips had nothing comparable.)

• Looks, case and connections: Early 80s – need I say more? In those days, Philips designs were very boxy, more so than average I think. They are frequently rather bold and some still look good today, take the D2924 for example (the version without the black boxes around the buttons, just metallic buttons on light grey). If you like boxy 80s styling and sliders (no less than 4 are present on the front, plus a sliding on/off switch at the side), the D1835 is your friend – otherwise you might just find it ugly. As one of very few radios, it was available in two different case colors: grey (which seems to be most common) and blue, probably reflecting that young people were part of the target group. Even today this is not overly common, except with some Chinese manufacturers like Tecsun or Degen. The only thing I did not like about the color choices is the model number on the front, grey (case color) on red, very hard to read. Colors on the blue variant are better matched, though the paint on the speaker grille comes off rather easily.
  Connectors include a DC power socket (6 V, center negative, as usual back then) and a headphone socket (not very "stereo friendly"); a carrying pouch is present. The whip antenna slides out like those on many "7600" Sonys and is approx. 75 cm in length. Its tip features a ring for inductive coupling of some wire antenna, although none of those is supplied. (Directly attaching a wire antenna would detune the front end.) The back of the case features a small stand quite similar to that on the ICF-7601. The case itself is plastic, but rather sturdy. Only the tuning knob looks a bit cheapish, but at least there's no excessive backlash.
• FM performance: The D1835 appears plenty sensitive on FM, and pretty selective (noticeably more so than the ICF-7601) in spite of its rather simple IF design, which surprised me a bit.
• LW/MW performance: On LW, the ferrite rod antenna is very directional. Selectivity was not sufficient to fully separate weak DLR on 177 kHz from strong Europe 1 on 183. On MW, weak stations seemed more noisy and more distorted than on the ICF-7601; it seems the demodulator distorts more on weak signals and the AGC does not turn up the gain quite as fast. Upon attaching a power supply, several ghost stations from shortwave appeared on MW. There's an occasional het (image? overload?) to be noticed on LW. Apparently LW/MW front-end selectivity could be better.
• SW scales and performance: The scales for the SW bands (from 49 to 11 meters, including 22m – pretty good coverage) contain freq marks in 100 kHz or 200 kHz distance. Calibration is only average, the widely spread 49m band (about 400 kHz for the whole scale length) is off by some 30 kHz on my grey sample (but then the radio is over 15 years old and components have had enough time for aging). Looking at the schematic in the service manual, it seems that individual dial calibration for each band should be possible by adjusting inductances 5031 through 5039 (underneath the tuning dial, as usual) along with trimmer CT5.
  Sensitivity was a pleasant surprise: On 22 meters, it appeared to be quite on par with the ICF-7601, using the same set of batteries. Selectivity, as noted before, was not and made interference whistles heard more often than on the Sony (for example, when trying to separate DW on 6075 from BR on 6085 a slight background chatter remained on the Philips while it could be pretty much eliminated on the Sony); still, it's certainly usable.
  While I did notice some apparent mixing products (just like on the Sony), overload issues upon attaching a power supply as noted with the Sony were not apparent (who in their right mind would want to use the power supply cable as an antenna anyway?). 31m in the evening did overload the receiver but not as badly as the poor ICF-7601, with things being reversed on 41m.
  The noise level on empty band regions is generally very low (same goes for the audio noise level, the Sony is a bit more hissy), with available stations becoming clearly audible. I'm also under the impression that the S/N ratio on moderately strong stations is better than on the ICF-7601. (But again, very weak stations are somewhat more noisy and distorted.) Maybe the AGC is not that bad after all?
  Image rejection proved to be not perfect, but was quite decent for a single conversion set, with just a few images appearing above 49m and an occasional utility station on the higher bands. You do notice the tuned front end here. (Contemporary single conversion sets not infrequently make use of fixed band filters only, some even don't use any at all. In PLL sets you typically either find no front-end tuning or varicap tuning.)
• Sound: Another pleasant surprise. Like on the ICF-7601, speaker sound is quite full given the size, and thanks to a tone control slider (for the highs) it can be made yet more pleasant. Bass response is clearly better than on the 7601, making the D1835 my best-sounding receiver in its size class. Amp hiss during speaker playback is also low. If you want to listen via headphones, I recommend getting a mono-stereo adapter, the headphone jack is not very stereo friendly.
• Power consumption: Given the concept and achieved performance off my set of partly empty batteries I'd assume this set is about as economic on batteries as any other analog. The battery compartment lid is rather small and can be difficult to slide on, a minor annoyance.
• Age related weaknesses: Unlike the first sample I obtained, the second one shows contact problems with the shortwave band switch. Naturally it's not obvious how to get to the front of the radio in order to clean this. Ah nevermind, I finally figured out how to interpret the disassembly drawings and exploded view correctly. Turns out that the problem wasn't contact corrosion but ordinary dirt accumulated behind the front panel (i.e. not age but rather treatment related), which could be helped by means of some lungpower. The set now works flawlessly on all SW bands.

Overall I'm quite pleased with this little set. It's not the world's greatest DX machine, but a solid performer nonetheless, with an excellent price/performance ratio (given you can snag one in good condition for, say, ≤ 10 EUR). I particularly like the good sound and low noise level during speaker playback.

My own grey D1835 carries a S/N of KT 03083023095 (whatever that means). The second sample, a blue one, is # KT 0370300012.

Manuals for other Dxxx radios (not including the D1835)

The approx. 1988..1990 D1875 features a virtually identical arrangement of connectors and controls combined with more modern styling and a decently sized flip stand. I wouldn't be overly surprised if the internals turned out to be very similar or even identical to those of the D1835, at least it appears to be based on the same NEC µPC1018C IC.
On a side note, there seems to be a non-name D1875 knockoff of dubious origins that has a mirrored dial scale, i.e. the shortwave bands on the left. Now given that the original isn't too high-tech to begin with, I guess a copy needn't perform a lot worse. Strange stuff!

↑ Back to table of contents

Sony ICF-SW7600

(Sample obtained used in June/July 2004. S/N 301101.)

Intro

With some rigs you always wonder how they might perform, without actually ever getting round to trying one. Well, for me, the ICF-SW7600 was such a rig. In 2004 I was finally able to obtain one, unfortunately in need of repair as it turned out. Fortunately I was able to find someone who is a lot better in these things than I am, and after a non-trivial repair session my SW7600 was working (almost) like new. Eventually other problems cropped up, however, and it was time for a total recapping... but more on that later.

The ICF-SW7600, introduced in 1990, was Sony's replacement for the older ICF-7600DS, which had been little more than an optically updated ICF-7600D (first presented in 1983 and with some notable interior changes in 1985) to begin with and therefore was getting a bit old. It's a portable in the popular pocketbook size class, measuring an official 19.1 x 11.8 x 3.2 cm³ and weighing 615 g with batteries. The actual front is no larger than for the predecessor, but they cheated a bit and had the controls on the right-hand side stick out to allow for a larger circuit board size. The case is almost all (ABS) plastic, with a metal loudspeaker grille, as also found on a number of other "7600" models. It's held in sort of a blue-greenish dark grey. For catching the waves, a telescopic antenna (68 cm in length) with slide-out base is provided, plus an internal ferrite rod antenna (15 cm long) for MW/LW. The speaker driver installed is an 8 cm (3") job. A wrist strap is attached at the top left.

This set, which uses a PLL frequency synthesizer, tunes the AM ranges from 150 kHz to 29995 kHz continuously (coverage is limited in some areas) and the FM broadcast band from 76 to 108 MHz (starting at 87.5 MHz for a number of areas; stereo reception provided via headphone jack). Tuning steps vary from 3 kHz on longwave over 9 kHz on mediumwave to 5 kHz on shortwave, with FM being tuned in either 50 or 100 kHz steps, depending on band coverage (!). SSB reception with sideband (preference) selection is possible on the AM bands (not for Saudi Arabia model). Tuning is entirely keypad based, with up/down buttons, direct frequency entry (via a telephone-style numeric keypad) and search tuning facilities being available. On the AM ranges, analog fine tuning is available for AM and SSB modes. Clock and timer functions are provided. A total of 10 memory presets store both AM and FM frequencies. There is an LC display for showing frequency or time that can be backlit momentarily.

An external antenna input is provided on many but not all versions (my German model excludes it). There's not only a headphone out but also rec out (mic-level) and remote control jacks. There is a flip stand on the back to allow for a tilted position. A two-position tone switch (NEWS / MUSIC) is provided. This model was the first to include the sliding on/off button which in the "LOCK" position will keep the set from inadvertantly powering up.

Supplied accessories included an AC adapter (e.g. AC-240), carrying case, stereo earphones, the AN-61 reel antenna and an antenna adapter for coaxial cable (provided the set has an input), along with the usual documentation like user manual and the "Wave Handbook", which contains information on shortwave listening.

For the technical details on the set, check out its listing on the Sony 7600 series page.

Now, how does it perform when contrasted with my faithful SW7600G? I used the original Sony AC adapter for the SW7600 and the usual universal regulated one for the SW7600G, as well as rechargeables in both.

Build quality

• Good, though the case is not quite as stable as the brick-like one of the 7600G. (That may be related to the 5 screws holding the back on the later model vs. 4 on the SW7600.) Four little round feet keep the flip stand from rattling while not in use. No problems with the antenna, which at 68 cm is a good bit shorter and lighter than the successor's – to the benefit of durability, it seems. Headphone jack is a bit stiff.
• In comparison with the ICF-7601, the volume slider is much easier to move, though a smooth volume knob still is better.

Operation

• Tuning the broadcast bands is done relatively quickly for a set with keypad tuning only. Here the newer SW7600G only allows continuous manual tuning with 1 kHz steps (thankfully muting-free, but a little slow at about 10.0 kHz per second), only accepting single steps on the outer tuning buttons with the respective broadcast frequency grid (i.e. 5 kHz for shortwave) before starting a scan if you keep those pressed. The SW7600, by contrast, will chuff along with 5 kHz steps indefinitely on shortwave (it does not offer anything smaller), for a speed of about 52.3 kHz per second, with audio being muted maybe half the time. Thankfully the muting is not accompanied by clicking noises as it's on the SW30 (1 kHz steps only, about 7.77 kHz per second). Automagic scanning can be carried out in one direction only, namely upwards. (I never use this feature anyway.)
• The display backlight is a little dim on rechargables (if still noticeably brighter than the successor's, which is a joke), but OK with the AC-240. Time constant is a little short though, even after complete recapping, so I'd suggest replacing C221 (22µ, the only electrolytic on the microprocessor board) by 47µ while you're at it (same value as in SW7600G). Display contrast is a little on the low side when compared to newer models.
• Having grown up with an SW7600G, I found a number of things to be somewhat impractical or limited: The frequency entering concept with two different buttons for AM and FM (inherited from the tiny ICF-SW1) is simply silly when otherwise the user interface treats FM like "any other band" (band cycling brings you to FM after 11m), as a consequence there is no memory function for the last freq used in AM and FM separately, and the wake-up timer function can only be used with the last station tuned to.
• The clock is only shown when the set is off. The predecessor still had an independent clock display.
• A total of ten presets, the same number as on the predecessor, is hardly to be called luxurious.
• Clock and presets are retained for half an hour or more if power is removed, which leaves plenty of time for changing batteries.

Performance on the AM ranges

• Sensitivity: Here the SW7600 and SW7600G performed like brothers. (Which does speak for Sony's QC!) Using just the whips on both, I couldn't really make out a difference on 49m. On MW, there was no difference either (however, the older model switches from the ferrite rod to the whip at 1615 kHz rather than 1711 kHz, thus making a less good choice for catching pirates and such between 1611 and 1700 kHz). On 19m, 16m and 13m the SW7600 proved to be noticeably more sensitive, but when re-testing with the same AC adapter on both models (voltage seemed a bit high on the Sony adapter), they were extremely close. With both sets running off rechargeables, the SW7600 was ahead again. (I already assumed this to be a general trend since it was backed on 60m and 75m as well, plus there is a pronounced difference on 90m and 80m, but some re-testing showed that the SW7600G gave a better signal on 41m and also seemed to be a tiny bit ahead on 31m. Those hardly are the bands where you'd need maximum sensitivity though.)
  All in all, we can conclude that sensitivity wise the SW7600 is largely equal to the SW7600G and even better on the lower bands, in spite of the shorter aerial. The SW7600 holds an advantage in picking out weak(ish) stations in the clear due to a friendlier audio response with the tone switch set to MUSIC, although on very weak stations the extended audio response can also be counterproductive (when using headphones, at least – during speaker use the added presence helps in any case).
• Selectivity: Since both receivers use the same IF filter, performance here should be almost identical – but isn't. The SW7600G uses a pretty aggressive low-pass filter to reduce the AF bandwidth even in the MUSIC tone switch position, which makes for noticeably better channel separation at the cost of fidelity. (A 3 kHz lowpass is already built into the CXA1376 IC used.) Still, the SW7600 performs quite well here, though interestingly just a hair ahead of the ICF-7601 when it comes to +/- 10 kHz separation.
• Strong signal handling and image rejection:
  • Image rejection proved to be almost identical to that on the SW7600G, which was a bit of a surprise since judging from the measurements Radio Netherlands had taken back then it should have been noticeably worse. (In any case, it's not all that exciting and one of the receiver's weak points.)
  • Mixing products: Listening on 80m in the evening, the newer model already showed a het from a mixing product while the SW7600 did not – however, the difference isn't huge, attaching any length of wire also made the het heard here. Given that the front-end is almost identical in both models except for a somewhat lower supply voltage for the 1st mixer on the SW7600G (4.8 V instead of 5.2 V), this didn't come as much of a surprise.
  • Interestingly, a little MW shootout involving a fairly strong station supported by my "AM loupe" showed that its close-by cross-modulation / intermod performance (as observed on the next channel up) is decidedly better than the newer ICF-SW7600G's! On this band, at least. Like the sets with frontend tuning (SW30, 7600A, 7601, RP2000), it showed no cross-mod, while the other contestants with wideband MW frontends (ICF-SW7600G, DE1102, E100) did. Elevated distortion, just like on the other Sony models, wasn't an issue (the RP2000 at full gain showed some, and the E100 had already gone hopelessly into limiting). A retest at higher signal strength allowed further differentiation in terms of cross-mod and IMD: ICF-7600A and ICF-SW7600 tied for first (still some 3rd-order IMD directly inside the loop that's not affected much by the attenuator on the SW7600), then ICF-SW30, ICF-7601 and RP2000, then the rest. Interestingly enough even the mighty ICF-2001D only lined up with the "rest", presumably due to the low 1st mixer supply voltage.
  • 2nd-order IMD performance on MW is about on par with other portables with wideband frontends, e.g. ICF-SW7600G and E100. At one time I thought it was much better, since while I was easily able to provoke 2nd order IM on those with my "AM loupe" tuned to about half frequency (1566 kHz makes a good channel here), the SW7600 resisted. However, it seems this was third-order (3*f1-f2) and not second-order intermod.
  • Turning on the attenuator on MW does not reduce signal strength all that much, certainly less so than on shortwave. Good if there's only slight 2nd-order intermod to be fought, not so good if more drastic measures would be required. (It's an interesting construction though. in LOCAL, a parallel RC network attenuates the signal, and the preamp FET gets some source degeneration.)
• Sound: Overall, I preferred the SW7600's sound over the successor's. Not only is its speaker sound a bit more mellow and much easier to live with, also it delivers noticeably more bass to the headphones. (Which incidentally seems to be the normal amount, the SW7600G is a bit weak here.) Highs on FM are noticeably crisper on the SW7600G. (Frequency response up there measures pretty much identically though.) The ideal receiver would combine the advantages of both, I guess.
  The AM low-pass filtering is much less aggressive and makes for easier listening in many cases – what comes out as a very muffled station on the G can still be quite listenable on the older model. The SW7600 has a pretty fat, "hi-fi" AM sound in general, even a modified G can't match that in terms of bass and highs amount.
  I found that the bandwidth on the SW7600 with the tone switch in NEWS position was still somewhat wider than that of the SW7600G in MUSIC position. (My audio spectra later confirmed this.] On the 7600G, it's possible to reduce the disadvantage by using synch detection and off-tuning by 1 kHz, however that's at the expense of synch locking stability which isn't that terrific anyway. A good compromise would probably be right between the two models.
  At very low signal levels, the SW7600's AM detector seems to be working somewhat less well than that of sync detector models (e.g. ICF-2001D), whose ICs apparently use pseudosynchronous detectors.
• SSB:
  • Nice clean SSB audio, and just about as stable as it gets – even on 10 meters, where strong stations exhibit quite noticeable warble on the SW7600G. Not so on this one! Unfortunately, the detector used is not sideband selective, the LSB or USB setting only determines the frequency of the inserted carrier (+ or - 1.5 kHz from the filter center freq, respectively, to include somewhat more of the wanted sideband within the filter bandwidth, which of course is not exactly tailored to SSB use but a good bit wider still). Slight distortion on beginning of first syllable with strong SSB stations (AGC not quite quick enough). Clarity on weaker stations would benefit from less bass, the set is a little too "hi-fi" here.
  • Since both the LSB and USB center freqs are off from the filter's center frequency, you never know exactly where you are, even if a station is right on a 5 kHz step. While we're at it, 5 kHz steps are clearly inadequate for comfortable browsing of the ham bands, the 7600G's 1 kHz steps are a magnitude better. The fine tuning range is also rather huge, roughly +/- 8...9 kHz on my sample (vs. +/- 1.5 kHz on the 7600G); you may find yourself 5 kHz (or more) above or below the displayed frequency when searching for SSB stations with the fine tuning control. Surprisingly it does allow for sufficient fine tuning of ham stations when operated with a little care.
  • ECSS, however, is not possible due to the nature of the product detector used – you can get very close to the carrier but will then have to fight with the volume oscillating with a beat tone because the two sidebands at slightly different frequencies are interfering with each other. The 7600G with its sideband selective product detector (which was needed for sideband selective sync detection anyway) holds a big advantage here. It should be noted that product detectors that are not sideband selective are still the normal case with portables today – to name a few, Grundig YB-400(PE), Sangean ATS-505 and Degen DE1102/1103 including Eton E5. (Sony's "secret weapon" is the proprietary CXA1376 IC.)
  • A small quirk: When tuning to a frequency close to 455 kHz (the 2nd IF) in SSB, the set apparently receives its own BFO nice and strong. All is quiet in AM. The SW7600G remains peaceful regardless of mode (the respective oscillator runs at about 3.64 MHz here).
• Spurious signals:
  • Spurs are at least as well suppressed as on the ICF-SW7600G, which means that only those in the 27 MHz range (27012.5 and 27467.5 kHz) are moderately strong and the ones at lower frequencies are quite weak. Oh, and judging by the sound of these spurs, the 1st and 2nd LO must be pretty clean (and fairly stable at that, save for a bit of thermal and PCB tension related(!) drift; the BAND and -⇐ keys in particular seem to affect the PLL reference xtal), the SW7600G is a touch rougher sounding.
  • I did some thermal drift tests with the 27353.75 kHz spur, (m,n) = (4,6), expecting it to highlight 2nd LO drift, which then should be more pronounced in the SW7600 than the SW7600G (2nd LO fine tuning vs. BFO fine tuning) – well, turns out that the SW7600G shows a lot more initial warm-up drift! Evidently the BFO (which is the PLL VCO used for sync detection) is not too stable, while the 2nd LO in the SW7600 is properly temperature compensated and all. Eventually the newer model settles down and shows a signal with better short-time stability, though I guess that the occasional slight jumps in SW7600 frequency have more to do with the fine tuning pot (very touchy).
  • The stronger spurs are well suited for observing PLL locking – except that there is nothing much to observe on this model, which is "there" as soon as it unmutes. The ICF-SW7600G, by contrast, appears to lock quite slowly on the high frequencies, wobbling around for well over a second when tuning from a low AM freq to e.g. the 27012.5 kHz spur. (Tuning from a high AM frequency to a 49m band station gives a very short lock-in time, however. Seems like PLL loop bandwidth, which – in light of smaller reference frequency and higher dividers – I'd expect to be smaller than in the older model to begin with, becomes too small on the high frequencies. If then the signal somehow manages to modulate the VCO and throw the PLL out of lock, you get the dreaded SSB wobble. There is some inherent frequency dependence of loop bandwidth in PLL synthesizers, but I guess in this case the VCO's control characteristic may have an even bigger influence.)

FM reception

Probably in part due to the shorter aerial, sensitivity on FM seems noticeably lower than on the 7600G. (That was when the latter was still unmodified. A badly matched 150/110 kHz IF filter combo can work wonders in terms of reduced sensitivity and bad distortion on weak stations.) The stereo threshold is even more so, allowing for more consistent but also more noisy stereo reception. As mentioned before, I liked the tone quality even though highs are a tiny bit lacking (which certainly is not the fault of the LA3335 MPX chip – which is still used in the 7600GR today – but the audio circuitry that follows it). (Interestingly enough, the measured frequency responses on FM merely show the 7600G's rolled-off low end as a difference, while highs filtering is just about identical.)

Battery usage

Not evaluated properly yet. Still prior to recapping, my sample seemed to be straining its batteries somewhat, in spite of a specified quiescent current that hardly exceeds 60 mA; possibly there were some higher-frequency fluctuations (with somewhat dead electrolytics, things won't get any better either). As in the predecessors, no low battery warning is provided - however, a double pop noise when turning on points to a low voltage. The set still runs off only 3 volts, which should make for optimum battery usage but may be critical for rechargeables.

Misc.

Too bad my SW7600 (German model) does not feature an external antenna input jack, it would certainly be fun to try the AN-LP1 with the thing. (In fact, I tend to forget this once in a while, proceed to set up the AN-LP1, and then am disappointed each and every time again. Man, I hate those useless regulations, especially since the ones in question just vanished shortly after this model came out. OK, what are the poor Italians supposed to say... On the other hand, nowadays we could use some more enforcement of EMI regulations, otherwise how could things like plasma TVs with massive RF radiation have been sold?)

The screw that holds the telescopic antenna is flush with the rear and can easily be touched when holding the set, thereby degrading reception due to capacitive coupling. The succeeding model has a recessed screw.

The rec out jack really is mono only (confirmed by schematic). Probably this is because output level is very low and only suited for a mic input, as on previous sets (later models have a full-blown line out, as does the more upscale ICF-SW55). The rec out precedes the active filtering stage, so there should only be a first-order lowpass in action there. There is a REMOTE jack which can be used to turn on matching Sony tape recorders (TCM-27 and TCM-77, with PC-261M adapter; TCM-25 with optional remote unit RM-43), possibly others as well.

Conclusion

All in all, the ICF-SW7600 is a nice shortwave portable with good sensitivity, selectivity, strong signal handling, so-so image rejection and good sound that is well suited for broadcast reception and usable for SSB (reception is good, it's just not that much fun with 5 kHz steps) – as long as it works. The electrolytics used back then are the Achilles heel of this receiver, and non-working samples aren't particularly rare. See Failures and Repairs for details.

Notes

As of early 2009, my ICF-SW7600 was out of order again – this time, the left output channel no longer wanted to cooperate. Ah well, yet another dead-o-lytic™... Thanks to Helmut G. Vogel for recapping the set (for free!), who found that the old electrolytics were measurably dried out, just like on other samples he was fixing at the same time. Thankfully the board showed no corrosion damage (just some electrolytic deposit), unlike on another "patient".

It is interesting to note that in order to achieve even the relatively large tuning steps by todays's standards, a trick was necessary – see "Tuning step trickery".

Audio noise spectra

Some audio noise spectra (recorded shortly after those for the RD1220):

Looks like audio filtering consists of a 2nd-order lowpass, with additional first-order low- and highpass filters (or a bandpass) activated in NEWS. For reference: The SW7600G's response in MUSIC looks like the SW7600's in NEWS, albeit with a flatter passband and steeper filter skirts, and switching to NEWS only attenuates the highs even further! (Admittedly the changed highs response in MUSIC seems to be mainly due to a tighter filter.) No wonder I generally prefer the older model's (less "communications-grade") audio, the tone control has a greater effect as well.

A note on the AC-240

That is a pretty usable power supply. Unloaded voltage in the SW7600 is about 6.5 volts (6.6 measured directly), which drops to about 5.7 V with the SW7600 on FM and 5.6 V on AM. This is in the 240V setting on 230V mains, values are higher in the 220V setting. It's only "semi-regulated", but does a pretty good job in the RF department, which usually is more important for a receiver.

Update: I took a peek inside (easy), and it looks like it's actually an unregulated job. Rather compact construction, PCB shows about a dozen solder joints. A small ceramic capacitor was spotted at the rectifier, so it seems it uses some of those in order to quieten down the diodes. The main filter capacitor is a 2200µ/10V Rubycon. Unloaded voltage was closer to 7.0 V this time, or 7.7 V in the 220V setting. Very little mechanical transformer hum, very low idle power draw. Sensitive-ish headphones (HD590) on SW7600 reveal low-level hum, but at this point amplifier hiss is a bit annoying anyway, and resorting to a headphone amplifier (FiiO E11) drops both of these to inaudible levels.

↑ Back to table of contents

Sony ICF-M60SRDS

Parents can be difficult people to deal with, particularly when looking for a "kitchen radio" for some news in the morning and such. My favorite was the Sangean WR-1, but, being a fully analog set, its feature set does not include RDS which was desired; also, the retro look did not appeal much. (Too bad, since the thing has the reputation of giving excellent reception.) It's quite amazing how rare RDS support is among ordinary portable radios, Grundig and Sony are about the only two companies offering portables with RDS. In the end, the decision was between Grundig's Concert Boy 80 and Sony's ICF-M60SRDS. The Grundig looks ugly IMHO, so we went with the Sony.

The ICF-M60SRDS – a version ICF-M60LRDS with longwave instead of shortwave coverage (albeit only with 9 kHz steps, problematic in case of 177 kHz DLF and 183 kHz Europe 1) also exists – is a mid-sized domestic portable in a "neo-retro" design featuring MW (9 or 10 kHz steps), SW (5 kHz steps) and FM (50 kHz steps and display to .1 MHz, with RDS) coverage. It either operates off mains or off 4 'C' cells (R14). Aerials include a pretty long rotatable whip and a built-in ferrite rod, an external antenna connection is not provided.

• Operation is pretty straightforward for the most part. (Hey, it's a Sony after all.) Non-technical moms should have little difficulty with this set. The rather large display (with switchable lighting) is easy to read from a distance even by me, the decently sized and unpainted preset buttons on top like in the olden days are easy to reach (I would appreciate seeing more radios using this concept again), the jog dial "tuning knob" with two tuning speeds is halfway intuitive (though I can't help attributing a cheap plastic feel to it and would have appreciated a real opto-coupled tuning knob, but I guess that's beyond the budget; the "fast" tuning speed is really fast on SW, but a bit slow on FM). The scanning function works well on FM and really stops on nominal frequencies only. I found it interesting that Sony dug out the good ol' HOLD slider again a few years ago, someone must have looked at their mid to late '80s models and liked the concept; in this case, the slider is easy to identify and use. The 28 presets are distributed over 4 bands (with 7 presets each), which are the three reception bands with FM being present twice. The band selector button is to be found next to the preset buttons on top, one of which has a blip on it. The LCD is big with easily readable frequency or station names, the backlighting is sufficiently bright and even. The only potential shortcoming I found is that the volume and tone control knobs are about equal in size and arranged in a rather unusual order with tone control on the bottom and volume above it. Overall, I think elderly and visually challenged people should get along well with the ICF-M60SRDS. I found the standby timer function to be quite useful, although it may cause confusion when accidentally activated.
• Sound shows the 12 cm speaker, though of course this is no hi-fi set yet. Seemingly slight bass expansion is used, which you notice after plugging in headphones (the jack is mono but stereo-friendly). You can adjust the level of the highs via a tone control to make the sound more pleasant. Sound is still more tinny than on the Signal 700, but the old Grundig uses rather heavy bass expansion along with an elliptic wide-range speaker. When compared to the Philips D1835 with its smaller 8 cm speaker, the ICF-M60SRDS turns out to be better but not dramatically so, which is a bit of a disappointment.
• Reception: You can't expect much from a radio like this for approx. 75 EUR. (Or at least you couldn't in about 2004.) On FM, selectivity is better than on the Signal 700 and ICF-7601, though worse than on a stock ICF-SW7600 with its two 280 kHz filters; this is sufficient for the usual locals, and RDS performance leaves nothing to be desired (I assume the responsible chip is also used in some of Sony's car radios; supported features: PS, CT, TA+EON, AF, PTY, RT). On MW, sensitivity is pretty good and can keep up with the usual 7600s (SW7600/G, 7601), with decent, though not outstanding selectivity. On shortwave, sensitivity is sufficient for the strong stations, and selectivity, being the same as on mediumwave, is sufficient to separate stations 10 kHz but not 5 kHz apart. Despite the rather uninspiring sensitivity (little better than an ICF-SW7600 with the attenuator switched in), I did notice IM products on unused frequencies. Of course, as to be expected in a single conversion unit, plenty of images. Consider shortwave more of an add-on, as with so many radios.

Overall, this radio does what it's supposed to, namely receiving local FM stations and occasionally MW stations as well. Incidentally, I assume it's based on the internals of the ICF-M760/770.

Update: This has now been replaced by an under cabinet kitchen radio for space reasons, a 15EUR job from a local discounter. For the price that one is far from bad, but sound is even less exciting than before. I have been eyeing a Grundig Sonoclock 890 as a replacement.
Pulling out the ICF-M60SRDS again, FM selectivity seems to correspond to one 280 kHz filter plus an IFT, MW sensitivity is up there with the best (no worse than the ICF-7600A) and selectivity better than average, and sound, while showing the speaker driver size, still is too honky for my taste – the old Grundig Sonoclock 670SP clock radio with its 10 cm speaker manages to outclass the Sony in terms of tonal balance. Maybe it was intended to compete with the Sangean PR-D3(L) (/CCRadio) which also appears to be tuned for the vocal range. Whatever, not my cup of tea.
We ultimately sold this one.

↑ Back to table of contents

Degen DE105

(Sample obtained used, early 2005.)

Intro

This little Chinese made, truly pocket-sized PLL synthesized radio (only about 11.0 x 7.1 x 2.0 cm³, which is smaller than even a Sony ICF-SW20/22) with MW, SW and FM coverage hasn't been available here in .de for too long (as of early 2005 – the rebrand has been discontinued for a while now). Nonetheless I recently (2005) stumbled across a somewhat used sample which I was able to obtain for roughly half new price. Apparently someone at Technisat considered the model interesting enough to start selling a rebranded version (with English lettering, of course) as Viola WR1, together with the (presumably original) earbuds also found in the Kaito KA105 version, two AA batteries, an instruction leaflet and a little station frequency brochure. At an affordable EUR 29,99 list price, you do not expect any wonders – yet, what you get is a well-built, if not to say sturdy little radio with good fit and finish (including the Al alloy front known from the KA105 and buttons with short travel and good tactile feedback) and reasonable performance. (The cuteness factor is also pretty high.)

On shortwave, which is covered from 5950 to 15600 kHz in 5 kHz steps, the set is single conversion with an IF of 450 kHz (sometimes these images are really good for something ;). MW, with the default 9 kHz steps, ranges from 522 to 1620 kHz, while FM reaches from 87.5 to 108 MHz (unusually only in 100 kHz steps).

Whoever decided to cover 49m down to 5950 kHz and 19m up to 15600 kHz only clearly didn't have much experience with shortwave. It wouldn't have been dramatic to cover 5850 or 5800 to 15800 kHz instead. (To be fair, the DE105 was one of Degen's first shortwave portables. Maybe they'll do a little refresh once.) The rather limited overall coverage probably has its roots within a varactor tuned front-end with not infinitely variable resonant frequency (little more than a factor of 3 is usually achieved, maybe because MW requires just that?), which presumably is also responsible for the split shortwave range on some other entry-level single conversion sets. That's about the only way to establish some image rejection on a single conversion receiver, and as a welcome bonus strong signal handling (mostly 2nd order intermodulation) is improved as well.

Shortwave sensitivity

The DE105 clearly uses front-end tuning, as indicated by the decent image rejection for a single conversion set and some hand sensitivity. It can't handle big antennas like the AN-LP1 in strong signal areas, but sensitivity appears to be decent even on the whip (with little overload problems), with the usual headphone antenna trick also being present (same goes for an external power supply, which may introduce all kinds of noise that way – but hey, who would use a receiver like that on AC?).

AM selectivity

Selectivity on all bands is quite OK given the price, but still the set struggles with 9/10 kHz separation, giving the once-typical hets on many stations (when compared to the ICF-SW12 or D1835 which aren't that much better up to roughly 10 kHz distance, particularly the filter shape factor and ultimate rejection aren't that great – strong stations can still be faintly heard up to 30..40 kHz away, while my reference ICF-SW7600G with its ~7 kHz 6-element ceramic filter only makes garbled audio heard 5 kHz away from these and shows no trace of them a few kHz further; in one extreme case a very strong station which caused slight scratching on the ICF-SW7600 up to 15 kHz away could still be heard up to 50 to 55 kHz away in varying intensity, clearly a spurious/ultimate rejection issue).

AM audio

Audio on AM lacks bass and is rather bright-sounding; some lowpass filtering would have been good to reduce annoying wideband noise. The cheapo filtering is definitely the biggest drawback of this receiver, but to be fair it is certainly not alone with this problem in its class. The synthesizer seems rather noisy (background noise sounds grainy somehow, and when zero-beating the oscillator signal in SSB on another receiver, noticeable hum components are present, while an LC or crystal generated signal from another receiver sounds perfectly clean), as is the AF amp with 'phones connected.

Internally generate interference

As has been noted before, some internally generated buzzing is present on MW; apparently the µP shielding is not very good (actually, when taking a peek inside I didn't see any) and the ferrite rod catches the stray radiation. This is also noticeable on SW with the whip fully retracted, but virtually disappears once it's extended, unless you touch the antenna that is.

FM

FM shows adequate sensitivity. Selectivity is a mixed bag, as it is unsymmetric – stations disappear much more quickly when tuning below their nominal frequency than above it. So in some cases 200 kHz separation works pretty well and in others 300 kHz of distance still isn't quite enough.

Audio

Audio frequency response seems to have a slight mid-to-highs hump, but is quite good otherwise – the DE105 is certainly not bass shy and additionally I see little reason to complain about L/R separation in FM stereo. When the set is cold (and I mean cold, as in whatever temperature there is in my rucksack when temps outside are below 0°C), it exhibits considerable bass distortion, which disappears once it's warmed up (looks like crummy electrolytics in the audio section). The supplied earbuds are pretty decent actually, rather mid-highs centric but at least not with extremely crummy frequency response or high distortion (Sony MDR-101, anyone?). Ah yes, speaker audio is just what you expect at the size, i.e. nothing even resembling bass. The ICF-SW12 is no better in that regard either. At least the frequency response reduces the volume of 10 kHz hets.

Misc.

• The second clock, offset from the first one by an integer number of hours, may be helpful in keeping track of UTC.
• Memories (10 per band) and settings are lost when the batteries are removed for too long, but the buffering time of one or two minutes easily allows battery changing without any such side effects.
• While we're at it, battery life is supposed to be exceptional. With two alkalines measuring less than 1.2 V idle, certainly less still under load, the battery indicator blinks but MW/SW sensitivity seemingly remains unimpressed.
• I like the hinged battery door – why doesn't everyone include this simple but effectively loss-preventing feature?
• Operation in general is not always as straightforward and intuitive as one is used to from the big players (e.g. saving memories takes pressing the memory input key again after entering the number), but overall there should be few problems once the instruction leaflet has been read.
• The momentary display lighting with one lonely LED is pretty uneven but gives enough contrast in the dark (unlike on the ICF-SW12, where you get a weird optimum viewing angle). The first 4 digits of the frequency/clock display take up most of the available height, thus are easily readable. (Actually the numbers are bigger than on the ICF-SW7600G display.) A Chinese oddity is the last number of the frequency display on shortwave (either 5 or nothing), which is much smaller than that. The tuning indicator in the upper left corner of the display can be hard to see with light coming from above (display frame shadow).

Internals

What you find inside the receiver is not exactly SMD parts galore, at least many parts seem to be conventionally-sized (and I guess cheaper) ones, thus restricting the level of integration. It's all pretty packed, of course. (2008 update: Through-hole parts one one side of the PCB, surface mount stuff on the other, that's how it's commonly done with these sets.) As I mentioned, there wasn't any obvious shielding to be seen on the microprocessor board.

Also read: RadioIntel review
a review in Chinese with nice pics, interior included (the same in Google translated Chineseglish)

It almost looks like a bigger (and probably better) AM filter had originally been planned (CF3) but this hadn't made it to production. *grmbl* (Incidentally, the wide filter position in the '1101 and '1102 looks the same. This should also fit a 4-element filter, e.g. Murata SFP or similar.) You can also see the CD8132GP IC the receiver is based on, apparently a Chinese equivalent of a Toshiba IC (TA8132AN).

Conclusion

The bottom line? If you want a small radio with a small price tag but still halfway decent shortwave performance and good battery life for lugging around (e.g. in a shirt pocket), you may want to consider this one. If you want the best reception for the money, I would not recommend it (AM selectivity is just too much below par), but if you're looking for better performance in this kind of size, choices are few and prices non-negligible.

Back in early 2005 or thereabouts, I wrote:
"My primary use for the WR1 (DE105) is some listening (mostly FM) while on the bus, with the stock earphones. It's pretty hard to beat this combo in terms of space taken up, though I'm happy when I can comfortably listen to a decent hi-fi FM tuner like my venerable Onkyo T-4650 via the HD 590s at home again. Shortwave sees mostly some DW and BBC reception, which is OK while underway, but again at home different equipment is preferred."
Oh my, how things have changed. These days, the DE105 spends its life in the asssorted radios drawer – performance is just too much below what I like to see.

Possible alternatives

• The Eton/Lextronix E100 at 12.1 x 7.5 x 2.6 cm³ is one option. Originally almost identical to the little single conversion Tecsun PL200, it now is a "full-blown" dual conversion set that shows good sensitivity and selectivity on shortwave while not costing an arm and a leg. You may want to read my review.
• Readers in the US may want to take a look at the Kaito KA1101, a relabeled DE1101 – Degen's first serious shortwave rx (the bigger brother of the DE101 so to speak) and frequently overlooked. It's noticeably bigger (but still fairly small) at 13.7 x 8.5 x 2.5 cm³ and only comes with a plastic front panel, but is a much better receiver than the DE105 – dual conversion with wide/narrow selection and generally on par with the DE1102 (which in turn is a tiny bit larger still), just no SSB or 1 kHz tuning steps – and features a larger speaker on top of that for not much more money, at least going by Universal Radio prices.
• Another very new alternative (just came out in 2008) is the Grundig G6 "Aviator", a model with a full feature set for its very compact size (12.5 x 7.6 x 2.9 cm³): Full 150 kHz through 30 MHz coverage with dual conversion (55.845 MHz / 450 kHz) with SSB, plus FM and even air band coverage. It's powered by only two AA cells. An external antenna connector is present, the built-in telescopic whip measures about 50 cm in length. All this for US$99. It's probably made by Degen. I do hope they bring this one to ol' Europe as well (sans the Grundig badge, obviously, as Etón/Lextronix only has the naming rights for North America).
• If you do not object to ordering directly from China, check out the DE101, one of Degen's first models; it seems to be officially discontinued (apparently replaced by DE106, presumably with almost identical receiving interior) but can still be bought with English lettering. (Update: Gone since 2007 or so.) This also is a single conversion receiver no larger than the DE105 (although it does have more room inside because the telescopic antenna does not disappear inside the case – a number of people aren't very fond of this anyway), but from reviews and inside photos looks to be more promising – there's a 4-element ceramic filter inside, shortwave coverage is from 2.3 to 7.8 MHz and 9.1 to 26.1 MHz (likely with frontend tuning), and there is some shielding to be seen. It only has a plastic case, but then if in doubt, I'd go for the better receiver instead of the better looks any day of the week!
• There also is the dual conversion DE1105 at 11.8 x 7.5 x 2.35 cm³, which looks pretty, covers shortwave from 5.8 to 26.1 MHz (1/5 kHz) and features no less than 1000 memories and a temperature display, but lacks an external antenna jack. Operation seems to be a little tricky, and the set appears to be a bit antenna limited (the ferrite rod isn't huge either). This has to be ordered from China.
• A relatively new entry in the DE105's size class is the single conversion DE11, which does have an external antenna jack and extends the DE105's shortwave coverage to 5.8...18.1 MHz while measuring barely larger at 11.0 x 7.1 x 2.3 cm³. It also features 1000 memories and temperature readout. First reports would indicate, however, that the DE11 (or the rebranded KA11) isn't that much better than the DE105, with lack of shielding still being an issue. Only audio quality seems to be good, as far as a 50 mm speaker will allow that, and the specs point towards a better (probably 4-element) AM IF filter which, however, still is more on the broad side of things when it comes to shortwave use. In Europe, this set has also been sold as the Scott RXP50.
• A little larger still at 12.1 x 7.5 x 2.8 cm³, the Redsun RP300 (a.k.a. CCRadio SWP) might be another option. MW, SW in two bands (2.3 to 7.5 and 9.2 to 22 MHz, single conversion), FM, 200 presets, no antenna input. It's known as a generally decent performer, but a number of people are annoyed by the digital volume setting being very coarse, going from barely audible straight to above average volume on headphones.
• And last but not least, Sony's ICF-SW07, ICF-SW100 (the one with ribbon cable problems on earlier samples) and the long-discontinued ICF-SW1 (which typically needs to have some SMD electrolytics around the audio amp changed) are small and generally well-performing choices with some antenna assistance (ICF-SW07 came with a matching AN-LP2 active loop that has no controls and is tuned by the set, ICF-SW100 "S" version shipped with AN-100 wideband active antenna and reportedly later SW100 samples had improved sensitivity, ICF-SW1S came with AN-101), but now all discontinued and generally not cheap even on the used market. They are dual conversion concepts on the AM ranges with a 1st IF of 55.845 MHz and 2nd IF of 455 kHz throughout, with 5 kHz tuning steps on the ICF-SW1 and (minimum) 100 Hz steps on the others (which also support sideband-selective synchronous detection and can receive SSB, if not without some warble seemingly related to the 2nd LO which is shifted to generate the smallest tuning steps; the ICF-SW100 is also said to exhibit too much phase noise to be useful for a DRM mod, unlike the oldie ICF-SW1 which gives pretty decent SNRs of up to 25 dB or thereabouts). These are probably about as much shortwave receiver as you can get in a truly shirt pocket sized package (ICF-SW1: 11.8 x 7.1 x 2.4 cm³, ICF-SW100 is much the same, ICF-SW07 13.5 x 9.1 x 3.5 cm³, the latter two fold up so are larger in actual use). How much sense it makes to limit yourself to a set this small when optimum performance requires an antenna solution that is at least equally large is another matter.

Audio noise spectra

Here are some noise / buzz spectra that I took on the DE105 in 2008:

FM doesn't look half bad, deemphasis seems to be close to what it should be for 50 µs. AM shows the expected rather large bandwidth and modest filter shape factor.

↑ Back to table of contents

Philips (Magnavox) AE3405

(Sample purchased used, ca. 2005. S/N KT 039410038021.)

Intro

The AE3405 is a little pocket-sized set with analog tuning covering the shortwave bands from 49m through 11m including 22m with dual conversion (4.52 MHz (4.50 MHz on mine, aging?), 468 kHz), along with MW, FM (stereo) and, in the /23 version which I have, the tropical bands (called TB) from 120 through 60m with single conversion (which, BTW, are received via the ferrite rod, as would be LW in /00 and /20 models).
At 12 x 7.5 x 2.7 cm and a weight of 170 g, it basically is a somewhat larger version of a Sony ICF-SW20 with much better band coverage (pretty much the equal of an ICF-7601 when it comes to the number of bands) and FM stereo included.
It runs off two AA cells or an external 3V DC (center negative).
This model dates from the early '90s, around 1990...1993 (the manual available online is dated 09/1989, the service manual hints at 1990). The design is based upon the Sony CXA1238M AM/FM stereo IC.

My sample was obtained cheaply due to its condition (last two antenna segments broken off but present, case in average shape), no mention of the model number and "Philips" being misspelled (it was pure luck that I saw the auction at all). Fortunately the rather unique design makes identification by picture an easy task.

Reception, 49m through 11m:

• Sensitivity proved to be adequate, while always a notch behind the ICF-7601 even when compensating for the difference in antenna length (the AE3405's whip measures approx. 50 cm in length). The DE105 appears to be a tiny bit ahead here, plus it can take an external antenna if necessary while the Philips has no such provision and would need to use inductive coupling. Looking at the AE3405's schematic, there is an IF amplifier stage but no RF amp FET present in the frontend, only the CXA1238's frontend is being used – thus the modest sensitivity (spec: 65 µV for 26 dB S/N) isn't too surprising.
  (BTW, the IC's mixer is also being used as 1st mixer, with the signal paths for dual or single conversion selected via switching transistors. The 2nd mixer is a BF550 bipolar transistor. Overall, the concept is rather different from the ICF-SW20's, which basically is a downsized ICF-7601.)
• Selectivity was a pleasant surprise, being hard to tell apart from the ICF-7601; I used to think that a 4-element ceramic filter is used but what you find on the schematic is two 2-element SFZ468HL plus one further 1-element ceramic filter/IFT. (I mostly preferred to leave the tone switch in the highs-attenuating position, otherwise it tended to be quite a hissy experience.) The ICF-SW7600 shows no better +/- 5 kHz selectivity either, but takes the lead in +/- 10 kHz and higher suppression due to its 6-element filter; the ICF-SW7600G is noticeably better at +/- 5 kHz selectivity but has rather muffled audio. (So much for this recap.)
• Out-of-band coverage is rather limited on the AE3405 (no BBC on 9410 here), probably in order to spread out the bands as widely as possible. Not exactly a bad idea, given the tuning mechanism has quite some backlash (maybe worn out on mine, at least the tuning knob has some play) and stations can be tricky to tune. Dial calibration varies and can usually not be called outstanding; I'll try to rectify that once I've got the service docs. (In the end, I never did that. They used some special screws for the case, and I never was motivated enough to buy a matching screwdriver.)
• Overload in the form of the characteristic wandering whistles / mirror images has been very limited and only to be noticed under very good propagation conditions on crowded bands. (I do, however, still suspect this one case to have been a true mirror image from 41m on 49m.) This is certainly in part due to the chosen sensitivity, but strong signal handling itself seems to be better than on the ICF-7601 as well. The tuned frontend bandpass filters probably contribute to this.
• The AGC lets the volume drop a bit too much on weak signals, things also tend to get rather hissy then.
• Oscillator frequencies seem to be LC generated throughout, including (strangely) the fixed one for the 2nd mixer, and are thus not perfectly stable. Interestingly the 1st oscillator frequency is 4.5 MHz above the received frequency for the lower and 4.5 MHz below the received frequency for the higher bands (19m and higher), primarily to reduce drift I guess. It also has the effect of 1st IF image rejection being a touch better, since, for an example, 15.5 MHz (signal) / 6.5 MHz (image source) ~= 2.4 while 24.5 MHz (image source) / 15.5 MHz ~= 1.6. (If you've been paying attention, yes, 49m images may potentially show up in the lower 19m band. I have, however, only noticed few and weak ones even with signal levels on 49m being high.)
• I wonder why they chose the 1st IF and then went with traditional LC filtering anyway - 4.52 MHz would have permitted using TV IF filters. Like this, it's got two tuned IFTs with an amplification stage in between.

Reception, MW

Sensitivity is somewhat behind the ICF-7601, presumably due to the smaller ferrite rod. (Spec: 3.5 mV/m for 26 dB S/N.) Directionality also is a bit worse for the same reason. Still, not bad at all. The DE105 has no chance competing here.

Reception, TB

The usually rather strong 75m stations could be picked up with no trouble in the evening. Sensitivity on 60m proved to be noticeably inferior to the ICF-7601 which covers that range with its dual conversion part and the whip antenna (remember that the AE3405 uses the ferrite rod). For some strange reason, touching the whip with the tip of the plug of the Sony AN-LP1 (which wasn't even turned on but unfolded and put together) helped sensitivity on 60 and made a number of stations heard noticeably better or at all. (Maybe using a ferrite antenna coupler with some wire antenna would give good results as well.)
It be noted that tuning can be ridiculously difficult here due to the wide frequency range crammed together.

Reception, FM

• Sensitivity on FM doesn't set any new records (spec: 3 µV for 26 dB S/N), but is sufficient for getting the stronger stations in stereo. Again, no obvious RF preamp present.
• Selectivity is decent, a bit better than on the DE105. (That's what you expect with two cascaded IF filters.)
• It may happen that the stereo LED lights up but the audio remains in mono anyway – either that indicator is a bit too generous or the auto blending circuit of the CXA1238 is at work here. Provided you do get a station in stereo, channel separation seems quite good to me.
• The frequency response of the normal tone setting ("DBB", even though this is printed above the other switch position) with its bass and highs boost makes for quite enjoyable listening. Noise level in mono is pretty decent for a portable and seems mainly amp noise limited, in stereo it's average.

Audio quality

• Tiny radio, tiny loudspeaker (3 cm dia.!), tiny sound. Actually sound is quite good for such a small speaker, the DE105 (which IIRC features a "beefy" 5 cm speaker) is better but not dramatically so. Achievable speaker volume is OK with full batteries (output rating 130 mW, speaker 150 mW max) but not so great with rather empty ones.
• Things look, err, sound better with headphones, with the normal setting of the TONE/DBB switch giving boosted highs and quite heavily boosted bass ("DBB" = "dynamic bass boost") and the other one resulting in fairly normal bass response and attenuated highs (plus a slight channel shift, maybe due to parts tolerances). For AM reception the latter is recommended, otherwise you'll get lots of wideband noise. Frequency response is somewhere between the ICF-7601's two settings then, with the Sony being less hissy on AM in any case (maybe a matter of IF amp noise).
• The power amp itself (a BA5206BF) shows a decently low noise level with the HD590 connected.

Battery life

The AE3405 is reputed to be pretty thrifty on batteries, and given the circuitry it should be.

My original tests showed much worse sensitivity on FM – well, these presumably still pretty full alkalines only had 1.15 V left (i.e. 2.3 V total), and the FM section does not work exceptionally well under low voltage conditions. (Don't laugh – are you really sure this wouldn't have happened to you?) Reception on the AM ranges, however, appears little different even with fresh batteries.

Build quality

It's not a Sony, and that shows. The AE3405 indicates "Made in P.R.C.", i.e. it was produced cheaply in mainland China. The case with decent fit and finish isn't all that sturdy and tends to creak if you press on it, the front part has developed a crack above the tuning knob. Overall, not exactly awe-inspiring. Other AE3405s were apparently produced in Hong Kong. (Looks like /20 and /23 models were made in China and /37 in Hong Kong.)

Conclusion

The bottom line: A small receiver with a good feature set and great potential on shortwave (the concept is better than Sony's with the variable 1st IF around 10.7 MHz which makes the 2nd mixer prone to overload) but only so-so build quality, dial calibration and sensitivity. Overall, I like mine.

Notes

Now that I have the service docs I'm even more impressed by how much radio was squeezed into the thing (though it's also obvious where they had to cut corners – still, it's a minimalist but elegant concept). At the time this was only topped (by quite a margin) by Sony's ICF-SW1, which was a real feat of miniaturisation back in 1988.

Need a user manual? Or some service docs?

Audio noise spectra

Some audio noise spectra (recorded shortly after those for the RD1220):

My, the DBB is one heavy loudness EQ. Without that, response doesn't look too far off for 75µs deemphasis. I did not get sensible results in AM, apparently hiss levels are too high.

Philips part numbers

At least in those days, Philips did not use the common nomenclature of the type of part indicated by a letter followed by a number (e.g. C169), but a 4-digit system with the first digit indicating the kind of part. I found myself scratching my head when I first saw this, so maybe the following table is of use to someone:

Interpreting 4-digit Philips part numbers

Number	Possible parts
1xxx, other	Misc. parts (mechanical, loudspeaker etc.)
2xxx	Capacitors (incl. variable)
3xxx	Resistors (incl. pots and jumpers)
5xxx	Coils incl. ferrite bars, filters
6xxx	late 80s: diodes
7xxx	early 80s: all kinds of semiconductor devices; late 80s: transistors, ICs, no diodes

Grundig Ocean Boy 204

It'll leave my shack again soon, but maybe some notes are appreciated. A 1963/1964 classic, it covers LW, MW, SW from 2 to 20 MHz in 3 bands and FM. Single conversion, LC filtering throughout, fully solid state. Oval wideband speaker, supposedly 2-way (switching in tweeter does not make a difference though), great sound for a portable though some more highs couldn't hurt; S/N so-so. Sample not in too bad a shape, except for two shortwave bands being dead and FM dial cord stringing being wrong (tuning backwards). FM not too sensitive but sufficiently selective, LW/MW sensitivity behind the usual 7600s but nevertheless decent (selectivity is also usable, and I didn't notice any overload presumably due to the front-end tracking; good directionality with the large ferrite rod). Not at all deaf on the working shortwave band (fortunately the most interesting middle one), but very prone to overload. Hey, it was built for '60s signal levels, with '60s transistors (bipolar only, of course, and maybe still partly germanium equipped). Technology has come a LONG way since then. No idea in what way a realignment would have helped performance, but it's certainly a good idea with a set this old. Overall, nice collector's item, but as such not great for a bargain hunter.

↑ Back to table of contents

Panasonic RF-B20

The RF-B20 is an analog travel portable of the smaller kind, approx. 17 × 9 × 3 cm large, which was sold in the second half of the '80s for approx. $140US (list) and maybe $110-130US regularly (not exactly a bargain).

The version RF-B20L sold in Europe features LW, MW, SW (in 6 bands: 49m, 41m, 31m, 25m, 19m, 16m) and FM. The front says "SW Double Superheterodyne", internally this is the same concept with a 1st IF of 10.7 MHz that we already know from Sony, with the same advantages (e.g. frequency stability) and disadvantages (wide 1st IF bandwidth and thus limited dynamic range).

The front and top of the case are held in black painted brushed Al (nice) while the rest of the case is plastic; fit and finish can be called good. The set runs off 3 AA batteries (or rechargeables, for that matter). The loudspeaker apparently is a 66 mm type. Special features? A DX/Local switch (here to be found on the back) is not commonly seen on receivers like this, while the classic HOLD slider is more widespread. As usual with Panasonics, the carrying pouch is attached at the bottom of the left side of the receiver. More unusually, the whip antenna, which can be pulled out a bit like on various Sony models, finds itself attached at the top right (left is more common). What is not present (yet) is a stand on the back.

• SW reception: Side by side with the venerable D1835, there were no obvious deficiencies in SW sensitivity, while the AGC did better than on the Philips (no surprise there). Selectivity turned out to be average, with 10 kHz separation not always being perfect. Roughly comparable here. Dial accuracy is quite good on the RF-B20 with its markings in 50 kHz distance, rarely more than 10..15 kHz off. Bands are spread over approx. 550 kHz, which is plenty for some and a bit tight for others (like 31m and 19m). It's a pity that 22m is not covered. Backlash is not a problem with this set, the mechanics appear to be precise. Dynamic range test in the early evening: As with other comparable receivers (e.g. ICF-7601), the poor thing overloads on crowded bands.
• LW/MW reception: A daytime check showed that the usual suspects were coming in without much trouble, so sensitivity looks to be OK. Selectivity should be the same as on SW, not razor sharp but acceptable.
• FM reception: Sensitivity appeared to be good, selectivity at least average. Dial accuracy again is pretty good as far as that's possible with frequency marks every 2 MHz (more isn't possible at this size).
• Audio quality: While the smaller speaker inevitably limits bass response, the RF-B20 fares quite well here and shows good tonal balance which can be improved further by reducing the highs a bit via the tone control knob. The only complaint one may have is that the frequency response does not extend as far into the highs as one has seen on stuff with larger speakers. Please note that it is not recommended to lay the radio flat on a surface, as speaker sound will become noticeably thinner then. Better keep a distance of ~30 cm (1 ft) or more. Upright operation with not a lot behind the thing works well. The role of rear ventilation holes for the radiation of lower frequencies is not to be underestimated, actually my good-sounding receivers all have rather large ventilation areas on the back while the only one with subpar sound for its size, the ICF-SW7600G, has one that is covered up from inside.
  The volume control is somewhat unusual in that you have to rotate it downwards to increase volume.

Overall, while you're unlikely to be able to chase DX with it, the RF-B20 makes a decent travel companion (as long as you can live without 22m coverage). It wasn't exactly cheap in its day for what it does (but keep in mind that the number of smaller dual conversion portables was very limited at the time, actually I can only think of the Sony ICF-4900/4910 and the internally identical ICF-5100, which were a good bit smaller still; the RF-B20 looks like somewhat of a "me too" product to counter the '4900), but you can see where some of the effort went. It's nothing you'd spend big bucks on these days. Modern equivalents include Sony ICF-SW35, Sangean ATS-606A/S and Degen DE1102.

Note: This set appears to have non-negligible battery drain when off.

JFTR: S/N 8ABWB43755.

Taking a peek inside (no, I'm not curious, I just want to know :P), you find a 12 cm long ferrite rod, an AN7236S IC and some dial cord mechanics, much more isn't apparent from the back.

Receivers like this frequently suffer from scratched and/or dirty plexiglass dial covers, since not everyone treats radios that well. Display polish as used for mobile phones (essentially a very fine-grained polishing paste) together with a soft cotton cloth helps here. Be sure to tape off the fringes of the cover carefully, since you don't want to ruin any other surfaces.

↑ Back to table of contents

Panasonic RF-B11 vs. Sony ICF-SW11

(Samples obtained used, ca. 2005. ICF-SW11 S/N 1089638, RF-B11 S/N EH4GB001383.)

Intro

Well, originally I just wanted to upgrade my portable listening setup to something a bit more capable and less battery hungry (not that the DE105 would be a battery hog for a PLL set, but a plain ol' analog is hard to beat in terms of battery life), then I happened to see a few auctions for brand new RF-B11s, won one, well and then I got a used ICF-SW11 at a good price, too. The perfect opportunity for a little shootout of these quite similar radios, don't you think?

Features

Both are 12-band analogs covering LW, MW, SW (in 9 bands from 60m through 13m) and FM stereo. Both feature a flip stand on the back, but only the Sony has a battery door loss prevention band attached at the battery door. In return, the Panasonic features tone (high/low) and mono/stereo switches and a stereo indicator while the Sony does not. Both have a carrying strap attached at the bottom right, and both run off either 2 AA (R6) cells or 3V DC.

Look and feel

• The ICF-SW11 undoubtably has the more modern looks, but feels cheaper – for example, the tuning knob is rather wobbly. Fit and finish are quite good on both.
• The tiny tone and stereo switches on the Panasonic aren't the easiest to operate, and the volume control knob is a bit touchy, too. I did, however, like the sliders which enable very secure operation, including the power slider. The switch on the Sony feels like it may be more likely to be turned on accidentally. The tuning knob on the Panasonic can also be operated from the side, which may or may not be an advantage.
• Both sets are comparably high and thick, with the Panasonic being about 2 cm longer – not yet pocket sized, but a pretty handy format.
• 2008 update: It may be worth mentioning that the Panasonic's telescopic antenna, which measures about 68 cm in length (the corresponding Sony whip is about 6 cm shorter, going by what the ICF-SW30 has), is attached a little awkwardly on the right-hand side, where you might touch it when operating the tuning knob (which then throws off antenna tuning on shortwave). The control layout is similar to that of a Siemens RK721, so I wouldn't be surprised if this set had also been sourced from Sangean (like RF-B33 and RF-B55), possibly as an OEM-only model.

FM reception

• That is what I primarily bought the RF-B11 for, and it does rather well here. Sensitivity is very high, selectivity is very good, too (beats a stock ICF-7600D). Stereo separation is very good for that kind of set.
• In comparison, stereo separation on the Sony isn't overwhelming, a trait it shares with the AE3405 based on the same CXA1238 chip. Sensitivity also proved worse than on the RF-B11 (with as much intermod nonetheless). This became even more pronounced once I compared the two using the same pretty flat alkalines that still happened to be in the Sony and measured 1.1 V each with no load (less with load): With both having their whip antennas of about equal length (OK, the RF-B11's is about 5 cm longer) fully extended, the ICF-SW11 didn't even light the tune LED on a particular station, much less received in stereo, while the Panasonic not only lit its tune LED but provided stereo as well. (That gives me hope in terms of battery life.) The Sony also falls behind in terms of selectivity, which is on the level of what one can expect with a single rather wide IF filter. I'd say the RF-B11 pretty much destroys the ICF-SW11 in terms of FM reception.
• 2008 update: I assume that the RF-B11 has one 180 kHz ceramic filter plus IFT, as it does not compete with sets having 2x 230/180 kHz yet the bandwidth is narrower than with two 280 kHz filters. I wouldn't be surprised if the venerable TA7358AP frontend found use in this set, along with a stereo decoder also sourced from Toshiba (which I have found to give good performance).

Sound

• In terms of speaker sound, the Panasonic is pretty good given its size, at least in the same league as the RF-B20 I'd say – not too amazing with the 8 cm (3") speaker. (BTW, AF output is spec'd at 300 mW, which is pretty amazing out of 3 V.) The Sony with its noticeably smaller speaker sounds more tinny in comparison.
• Over headphones, things are quite different. Here the Sony pleases with a nice frequency response that's only lacking a bit in highs (but otherwise isn't too far from a hi-fi tuner) while the Panasonic sounds rather bright in "HIGH" position and a bit too muffled in "LOW". Maybe the manual saying that HIGH emphasizes the highs while LOW reduces them is right after all. Anyway, I'd have preferred a linear frequency response for HIGH instead of this somewhat braindead solution. I guess it'll work decently with cans like the Sennheiser PX100 but anything with a more linear frequency response won't sound that well. That's one of the few things that I don't like about the RF-B11.

Shortwave reception

• Both have about equal selectivity, maybe comparable to the D1835 – 5 kHz hets are plenty, 9/10 kHz ones heard less commonly. (You didn't expect any DX rigs, did you? Besides, these don't suffer from the extra sucky ultimate rejection of the DE105.)
• Overload and images? Sure – not that I had expected any wonders from simple single conversion rigs like these. The RF-B11 falls a good bit behind the ICF-SW11 (and RF-B20) in terms of sensitivity, but is not as prone to overload in return (a benefit in regions with high signal strengths on shortwave).
• The Panasonic generally has less dial markings, calibration also is not quite as good as on the Sony.
• Taking the different AF responses into account, fullness of sound is comparable on both; neither has the extremely thin AM audio of the DE105. Unlike on FM, weak batteries hardly affect the Sony's sensitivity in the AM ranges, yet another similarity to the AE3405.
• 2008 update: A little mid-day shootout on 19m to 13m between the RF-B11, RD1220, ICF-SW7600 and lastly even the RP2000 (the ICF-SW11 had since been sold) produced fairly interesting results. With headphones attached (or just loosely connected to rx ground), sensitivity on the Panasonic rises considerably, closing up to and sometimes even beating the others. Now of course, being a single conversion set, it's still more likely to catch some stuff that ought not to be there, and selectivity only becomes half-decent when switching the tone control to LOW (albeit still with not too steep skirts and somewhat muffled as well; there also seems to be a highpass present, as the audio is not extremely, but still a little thin).
  In terms of intelligibility on weak stations, usually the sets with higher bandwidths gave better results, with the ICF-SW7600 being hampered by its extensive filtering that gives it the edge when good selectivity is required (and if that doesn't do well, I don't even want to know how the '7600G fares); the RP2000 with its two IF bandwidths and useful tone controls proved to be the most flexible set here.
  Back to the RF-B11 though: I'm under the impression that the band filters are peaked slightly high, same applies to the reception frequency (it is not unusual for the readout to be 30..50 kHz low), so possibly the alignment is a little off due to component aging. One interesting detail is that the AM IF seems to be an unusual 460 kHz (or maybe 459).

LW/MW reception

Both have their share of strange whistles on LW. The whip antenna noticeably affects LW reception on the RF-B11 but has little effect on the ICF-SW11. Selectivity on MW and LW is usable but 9 kHz hets are not uncommon when the band is crowded. Daytime MW sensitivity is approximately the same on both and does not differ too much from the ICF-7601, but a weak test station was disturbed by some kind of het (mixing product maybe) on the ICF-SW11.

Misc.

It may be worth noting that the RF-B11 is made in Taiwan, possibly by Sangean (which makes the RF-B33 and RF-B55 as well). The ICF-SW11 is being made in China, IIRC.

Conclusion

The verdict? None of our contestants comes out without a few ruffled feathers, but overall I like the more full-featured and more FM-friendly Panasonic better. Only an audio mod would be in order. The Sony is somewhat better suited for shortwave due to its higher sensitivity there, but failed to impress me on FM.

Notes

BTW, the RF-B11 currently (as of 2005, ed.) constitutes my mobile listening setup together with a (slightly modded) Beyerdynamic DT 231 PRO. Both have been holding up well in over two months of use. (I keep the RF-B11 in a small plastic bag to fend off scratches, seems to work.) I usually leave the tone switch in LOW, as the cans are a bit on the bright side to begin with. (Didn't really help, now looking for something with better isolation and less brightness. AKG K26P looks promising.) The receiver does not like being operated near stronger FM transmitters and will show overload in form of very strong 3rd order IM 100...200 meters around these even with just the headphone cable used as antenna. (Unfortunately that affects the station I'm usually listening to quite badly. Fortunately I don't live there.) Admittedly these are rather extreme conditions which would cause many other receivers to stumble as well.
2008 update: Both RF-B11 and DT231Pro have long since been retired. The aforementioned overload is present, to slightly varying degree, on basically all my sets with only a band filter in front of the FM RF amp.

Audio noise spectra

Here are some RF-B11 noise / buzz spectra recorded in 2008 (the SW11 has been sold):

There is this low-frequency rolloff that is always present. FM deemphasis doesn't look too bad though. The AM filter isn't a shape factor king, as expected.

A technical comparison

When looking at the schematics of RF-B11 and ICF-SW11, they are actually quite similar. Both are based on CXA1238M (RF/IF) and CXA1522M (audio amp), neither has a protection diode against DC voltage reversal. AM selectivity is one 2-element ceramic filter + IFT for both.

The differences are in the details – the Sony has an amplifier stage after the AM IF filter while the Panasonic doesn't (explaining the difference in sensitivity), in return the RF-B11 has two FM IF filters where the ICF-SW11 has to make do with one (and apparently noone bothered to align the stereo decoder VCO properly on the sample I owned). There is an extra transistor dedicated to tuning LED drive in the Panasonic, and the lean bass area is easily explained by smallish coupling capacitors. The FM band filter is of the discrete (L/C) type in the ICF-SW11, while the RF-B11 has a monolithic ceramic filter.

↑ Back to table of contents

Sony ICF-SW30

(Sample obtained used, ca. 2006. S/N 108182.)

Intro

This receiver, the grandfather of the current ICF-SW35 and introduced in 1993, was known for good reception at its price point (little more than half of the price for an ICF-SW7600G) paired with only so-so ergonomics. Well, curiosity finally got the better part of me and I snatched one up inexpensively.

Look and feel

The black all-plastic (but nonetheless very sturdy-feeling) case of the ICF-SW30, which was produced in Taiwan, is almost as large as that of a typical "7600" (and a bit thicker) at 16.9 x 11.1 x 3.5 cm³, which forms quite a contrast to the rather minimalist front panel controls (of which the frequently needed ones are fairly large) and few elements on the sides. Hmm, don't we know that stand on the back? Yup, this is very similar to the one found on the ICF-SW7600G, but a tad larger still. The whip antenna (approx. 60 cm in length) looks exactly like the one of the ICF-SW11.

Features

• Frequency coverage includes MW (up to 1710 kHz, in 9 or 10 kHz steps) with single conversion (450 kHz), FM (10.7 MHz, as usual), and SW non-continuously in 12 bands from 60m to 13m with dual conversion using IFs of 10.7 MHz and 450 kHz (experimentally verified).
  Band coverage is plenty in most cases, only 60m could have extended further down than 4650 kHz to include 4635 kHz and 19m should have reached further up than just 15600 kHz; why on the other hand 41m would start at 6950 kHz (even below the 40m ham band!) is beyond me – 7100 kHz or 7050 kHz would have been entirely sufficient. (It's even more grotesque now that the 41m broadcast band starts at 7200 kHz.)
• You can save stations' frequencies in 5 presets each for MW, SW and FM – that's pretty tight for SW.
• Groundbreaking new functions (ahem) include a DX/Local switch for SW and FM ("Local" needed some persuasion on my sample, some toothbrush action and a squirt of contact cleaner eventually fixed that) and a tone switch. You can also have a three-element battery indicator displayed.
• There are more clock functions than I'm used to seeing – in addition to local time with switchable DST you can also have world time (UTC) displayed with an offset to local in whole hours.
• Owners of an ICF-SW7600G will already know the key protect function and the sliding on/off switch.

Ergonomics

• Functions are pretty much self-explanatory (I got the hang of the clock functions pretty quickly without having the manual at hand), and the frequently needed controls on the front panel (like those for tuning, frequency range and SW band selection and scanning) are all reasonably sized, partly even bordering on large, so if you have large hands you may appreciate this. The frequency display also is fairly large, larger than on a 7600G, but cannot be lit.
• The main gripe with the ergonomics are the manual tuning speeds, along with the accompanying clicking noise which can get on your nerves (in fact, there is short muting on every key press, and the annoying clicking is a byproduct of the muting process). FM tuning is slow already with 50 kHz steps (and scanning does not help much either, as it uses the same steps and only goes by signal strength), but at least the 5 presets should cover the most frequently received stations. On SW, manual tuning is only possible with 1 kHz steps, for about 7.77 kHz per second – slow as molasses. Automatic scanning uses 5 kHz steps, but that only runs upwards, not downwards! To make matters worse, the mere 5 presets for the whole shortwave range are not sufficient to provide relief by far. (The modern-day ICF-SW35 offers a choice of tuning step sizes, so you can do manual tuning in 5 kHz steps. Additionally, you can save up to 50 frequencies total. That's a whole lot better.) If you want to get across a shortwave band halfway quickly, scanning with the attenuator is pretty much the only option; particularly comfy this is not, of course.
• MW is the least problematic band here, with the receiver only providing 9 or 10 kHz steps which get you across the band quickly enough (on the downside, this also means that fine-tuning is not possible).

Reception

• FM is not the strongest point of this receiver, with a pretty wide IF filter (comparable to the ICF-SW11 or ICF-7601) and only average sensitivity. S/N also could be a bit better, but at least stereo separation appears to have been aligned and is pretty good, just as the frequency response (resembling that of the ICF-SW11).
• On shortwave it gets interesting – here you notice the decent selectivity pretty quickly, for which I suppose a 4-element ceramic filter (or equivalent) is responsible. The bandwidth appears to be around 8-9 kHz, and while 5 kHz hets still are to be heard not too infrequently (though 5 kHz separation works decently as long as signal strengths aren't too different), their 10 kHz relatives are usually absent. That's somewhat but not dramatically worse than on an ICF-SW7600.
• Sensitivity also appears to be just fine, only a little behind the ICF-SW7600 and even somewhat ahead on 75m. (The AE3405 cannot compete here, sorry Philips.) Given the rather low 1st IF this is not overly surprising, since the lower the IF the easier it is to get decent gain. Some reviews strangely mention only average sensitivity, this certainly does not apply to my sample. On empty frequencies you get the impression that the receiver is fairly noisy, while in fact it's just the AGC's fault – the AGC in the SW7600 lets the signal drop a good bit earlier and thus gives the impression of a noticeably quieter receiver, while actual S/N ratios on weak stations didn't differ that much. In fact, I preferred the SW30's AGC characteristic there, although it be said that the AM audio frequency response of the SW7600 includes some more highs and tends to give it the edge in "DXing".
• Yes, there are image frequencies from the 2nd IF (so 60m may not be as much fun as it could be), but no worse than on the 7600G. (But unlike with the latter, you cannot connect an AN-LP1 or similar loop with preselection. It does, however, share that problem with my SW7600 which is a West Germany model with no EXT ANT jack.) As in a number of other portables using a 10.7 MHz 1st IF, a simple ceramic IF filter is used as 1st IF filter, which typically achieves an ultimate rejection of at least 35..40 dB (over 50 dB at +/- 1 MHz is also possible), which is not worse than what a 2-element xtal filter on 55.845 MHz manages – but of course the latter has a much lower -3dB bandwidth and thus reduces 2nd mixer intermod.
• 1st IF image rejection is nonexistant, as expected (the 1st LO of the ICF-7601 on 16 meters, which runs at 28.525 MHz, gave quite a signal on 7125 kHz) – fortunately there usually isn't much going on 21.4 MHz higher. Possibly this explains why in case of heavy local interference, the SW30 seems to pick up more of that than other receivers. There seem to be some internally generated whistles sometimes (e.g. around 6005 kHz, very variable) which can also be picked up by other receivers nearby.
• A very welcome improvement coming from the ICF-7601 is the much reduced mixer overload – this is a "true" dual conversion rig with a balanced first mixer and decent filtering on the 1st IF, a big plus in my book. It is, however, not quite as overload resistant as the SW7600, its "bigger brother" at the time it came out – during excellent conditions, I found 60m covered with quite a few IM products, and if you off-tune by 1 kHz from a station on a busy band, you may find a weak het indicating 2nd mixer overload. The schematic shows that while circuit topology is very similar to the SW7600, mixer and amplifier FETs are run at lower currents and partly are different types compared to the bigger set, so one would expect less dynamic range. Still, not bad for an entry-level travel set at all.
• I had some fun chasing oscillator hets on MW and SW using the ICF-SW7600G in SSB mode. They were pretty clean sounding; however, tuning the SW30 to a strong SW station caused a weak but audible modulation of the LO het with that station's content.
• On MW, you have the same selectivity as on SW, which is pretty good here. Sensitivity is a bit behind the ICF-7601 and noticeably behind ICF-SW7600G and ICF-7600A, presumably due to the smaller ferrite rod, but still fairly good. The straightforward single conversion design (pretty much as suggested for the CX20111) includes a tuned antenna circuit, which combats intermod and images well (image rejection is, in fact, better than on the ICF-SW7600G) but makes the use of ferrite-rod based boosters (nearby tuned MW antennas) impossible – AM air loops still work. After a glimpse at the schematic, it amazes me that a single varactor tuned antenna circuit can reject images this well.
• More recently I did some "torture testing" of the MW section with the "AM loupe" and found that as indicated by the good image rejection, the tuned frontend seems to do its job well. I could not provoke 2nd order IMD or images even with the loop deliberately mistuned, only distortion on a strong station could be achieved (this, however, also affects the Redsun and the RF-1410L – IF amp or detector are driven into saturation, I guess – but can be eliminated by reducing the RF gain or enabling the attenuator, while the SW30's attenuator is inactive on MW and the rx has to be removed some distance from the loop).

Sound

Speaker sound is on the bright side with the smaller speaker (66 mm), but still fun to listen to... more so than the 7600G, which someone once described as tinny and muffled at the same time, a description hitting the nail right on the head when comparing the two. No idea what whoever did the audio engineering on the 7600G had been smoking. On the headphone jack, the audio is very warm and bassy, sounds like noticeably overdone deemphasis. This does go well with notoriously bright headphones though, and probably tames the internal speaker as well.

On a side note, I have found the Beyerdynamic DT231 to give good synergy with radios like the ICF-SW30 (and others with the same sonic signature, e.g. SW11), which nicely tame the otherwise excessive brightness of these cans.

Battery life

The ICF-SW30 apparently isn't too much of a battery hog, with a 30% longer spec'd battery life than an ICF-SW7600 on only three AA cells. (Quiescent current is given at an economic 25 mA on MW, 30 mA on FM and highest at 50 mA on shortwave. The big difference is not surprising, after all it's basically a 1-chip receiver on MW and FM and uses quite a bit of additional circuitry for shortwave. Anyway, that still is about 10 mA less than for the SW7600.)

You have to be fairly quick when changing batteries, as the set won't keep its settings much longer than a minute with them removed.

The verdict

It's hard to go wrong with this one if you find it cheap – for around 15 EUR or even less (something approaching 20 would still be OK, too) there's not much better to be found in terms of raw SW reception, which is near perfect for a travel radio. Operation, however, is the weak spot of the SW30 and may cause considerable frustration if you tune around a lot. The way the muting is implemented, it seems silly to have it used during tuning – less generous use of muting would at least have saved the users from the annoying noises accompanying this. (No idea why they didn't catch that one during testing.)

If you were content with the level of performance provided but would like to have something significantly more comfortable with better frequency coverage thrown in, look at the Sangean ATS-606A/S, the ICF-SW35 or the Degen DE1103 (rebranded as e.g. Kaito KA1103 and already playing a class higher with a dual balanced mixer, switchable IF bandwidths and SSB – the closest ICF-SW30 equivalent would be the DE1101, and the more fancy DE1102 is not too far off either).

Notes

Back in the olden days Radio Netherlands used to have a review of the ICF-SW30 in their Receiver Shopping List. It got lost during their big site revamp in 2004, but I happened to still have a very old copy downloaded back in '98.
Radio Netherlands ICF-SW30 review.

Audio noise spectra

Some audio noise spectra (recorded shortly after those for the RD1220):

These are fairly similar to the ICF-SW7600's which still was a current model at that time. Calling the SW30 a "baby SW7600" wouldn't be too far off.

ICF-SW30 Internals

With the service manual having become available online, it turns out that "baby SW7600" is spot on. The shortwave circuitry is very similar, including the CX20111 FM/AM IF IC, though the FETs used in amplifier and mixer stages are run at lower currents and some are less expensive types, and the filter on the lower first IF is a wider ceramic rather than a crystal type. No wonder the set is a pretty good performer on the short waves. The AM IF filter appears to be the same type found in its successors. A µPD1724G microprocessor and PLL IC is used, the same one also found in the SW7600G a bit later.

Examining the offending muting circuitry, it seems this was just a stupid (maybe last-minute) design. Muting momentarily pulls electronic volume on the CXA1522M audio amp down, with no declicking of any kind. Not even the capacitor used for suppressing volume pot scratchiness can do much about it, since the collector of muting transistor Q10 is connected on the wrong end of series resistor R50. Later models not only used a switching transistor with an integrated base resistor and a collector connection on the other end of said series resistor, most importantly they also added a 1 µF ceramic capacitor between base and ground. (I would have tried 100 nF or so.) It should be easy to retrofit one if you are halfway comfortable with surface mount components, as the area isn't crowded and a grounded tab for the volume pot retainer is right nearby.

This receiver has also been modified for DRM reception.

Upon taking a look inside to clean up the DX/Local switch, I sighted not only the CX20111 (later followed up by the functionally identical CXA1111M and CXA1611M in Sony's semiconductor product portfolio) but also the CXA1522M audio amp. The chips' date codes seem to suggest a 1996 production date, which would indicate a fairly late sample – looks like this model was far less popular than e.g. the ICF-SW7600G. Internal construction is reminiscent of the contemporary 7600s but a little different again; mechanically the ICF-SW40 appears to be very similar.

Related models

The direct successor to the SW30 (apparently produced in the same factory in Taiwan, but "made in Japan" samples also exist) was the 1997 SW40, which was a good bit more comfortable while still not costing a fortune. As far as I can see, reception circuitry is identical in the ICF-SW35 and ICF-SW40, which both are dual conversion throughout the AM range (thus no frontend tracking on MW and possibly worse image rejection there) and share the same kind of 1st mixer (not dual balanced, unlike the SW30). It's ironic that the old ICF-SW30, the rx with the least inviting operation of the bunch, has the most advanced shortwave reception circuitry. That's tradeoffs, I guess...

The more expensive ICF-SW33 apparently behaved pretty much the same reception wise (while using a somewhat safer but unusual 1st IF of 21.44 MHz) and had a low-level rec out and more sophisticated clock functions (who really needs those?) including two standby memories thrown in. It also remembers the last frequency on each SW band and shows the meter band received. In addition, the speaker gets a metal grille and the whole rx actually is quite a bit smaller at 16.57 x 9.3 x 3.05 cm³. Looking at a photo it seems that it was made in Japan. Like the SW30, the SW33 apparently wasn't too much of a battery hog even if specified battery life ran closer to the SW7600's – someone managed to squeeze out 50 hours of (non-continuous) operation with alkalines. This model used to be fairly expensive for the reception provided and thus never was too popular. (Ironically the other rx with a 21.44 MHz 1st IF that I know wasn't too exciting either, the Sangean ATS-505.) It seems like the somewhat newer ICF-SW30 (1993 vs. 1992) was intended to provide a downsized version with almost equally good reception at a much more attractive price.

↑ Back to table of contents

Sony ICF-7600D

Who doesn't remember it, the classic that started it all? (Well, maybe the folks over the pond who got to know this one as ICF-2002.) Back in 1983 the appearance of this compact digital dual conversion portable made quite a splash, and for good reason. Until then, the best you could get in this kind of size (184.5 × 118.5 × 32 mm³ according to the manual) were "dual conversion light" receivers like the ICF-7600A with a 4-element ceramic filter for AM and 2 filters for FM, with more modest models dominating, but here you have a true dual conversion receiver (with a high 1st IF of 55.845 MHz at that) with a dual balanced first mixer, 6-element IF filter and all of the convenience that a PLL based set offers. Even SSB and CW reception is possible with something product detector-ish, and an external antenna input is included in many models. In short, basically a full-blown portable SW receiver shrunk to pocketbook size, and generally better performing and more reliable than the preceding ICF-2001 at that. No mean feat. Unsurprisingly they sold like hotcakes and reached more than 100,000 sold units in 1983 already.

• Look and feel: The case of my sample is held in a silverish light grey as it was popular in the early 80s. I'd consider the ruggedness to be about on the level of an ICF-SW7600 but can't judge that exactly since my 7600D had once been opened in not the most professional way and is not as solid as it could be mechanically. A common appearance problem of this model are the white buttons which become yellowish over time. Do I really need to mention the carrying strap on the top left? That one is held in grey but otherwise constructed much the same way as newer ones. The back of the unit exhibits a metal name plate, some speaker grille area, a world time zone map and some SW band limits. No, there's no stand there yet. The telescopic antenna on top the base of which can be pulled out is fairly similar to its current equivalents and is pretty much exactly as long as the ICF-SW7600's.
• Features: AM coverage reaches from 153 kHz to 26100 kHz (or 29995 kHz, depending on the model) with 3 kHz steps from 153 to 519 kHz, 9 or 10 kHz steps from 522 to 1611 or 522 to 1610 kHz, respectively, and 5 kHz steps from 1615 kHz upwards. FM covers 87.5 (or 76.0) MHz to 108 MHz in 100 kHz steps. Step tuning is done via an up/down rocker switch (which is also used to toggle through the various pre-defined broadcast bands with an additional key), with direct frequency entry being possible via the numeric keypad right next to that. 10 station frequencies can be saved. The clock can display the time in either 24 or 12 hour mode, standby and sleep functions are present. An attenuator switch is provided; it doesn't attenuate as much as on newer receivers. Reception modes include AM, AM with fine tuning and SSB/CW with fine tuning. A NEWS/MUSIC tone switch is provided, as is a tape out (which, as on the ICF-SW7600, is likely to be of limited use today as its output level is far below line level and obviously intended for use with microphone inputs).
• Reception: Raw receiving performance is very similar to, say, the ICF-SW7600's on shortwave – how boring. (Do note that I cannot connect anything resembling a big antenna, as West Germany models didn't feature an EXT ANT jack.) The only real difference is the audio frequency reponse, which is somewhat mid-centric to give the speaker a more mellow sound (and in fact, the result is rather pleasing). This is also the main reason for the receiver appearing to be more noisy; you don't really notice the higher phase noise (except when trying to mod the thing for DRM, of course). SSB is yet more fussy to operate than on the SW7600, having no preference for one side band at all and being a tiny bit drifty as well. Apparently my sample already is an updated one (ca. 1985 or later) with tamed warbling. On FM, sensitivity is good but selectivity, as expected with 2 280 kHz filters, isn't anything to write home about. The RF-B11 does noticeably better than that.

Overall, I found it quite interesting how little the raw reception performance in AM had changed since the mid '80s; however, SSB reception has greatly improved since then, and current models also feature more presets and synch detection.

JFTR: S/N 378330.

↑ Back to table of contents

Toshiba RP-20

This page hasn't seen any updates in a while, but as it happens, radiophilia struck again and my collection has grown a bit. First in line is this mid-'80s Walkman-style portable FM/AM radio with FM stereo. I bought it because I found myself using the radio function of my trusty Cowon iAudio G3 MP3 player (the G3/K26P combo being my usual portable audio source with – FM radio recordings), which, not being the most sensitive, struggles with reliable stereo reception of my favorite FM station during the daily bus rides, and that annoyed me quite a bit. (Apparently this is related to the Philips all-in-one radio chip with minimal external parts count used, which is an amazing feat of integration but seems to have its quirks.)

The Toshiba RP-20 is on the larger side compared to current portable audio gadgets, the brick-like radio (pretty much all plastic) measures some 11 x 5.9 x 2.1 cm (add about another 7 mm in thickness for the belt clip). It is intended for headphone playback only; the headphone cable also serves as FM antenna, while AM is received via an internal ferrite rod. The visual appearance screams 1985 (which apparently is this model's vintage – actually the service manual is dated May 1984), the tuning dial on the front for example reminded me of the Sony ICF-4900. A stereo indicator LED is integrated within. Two sliders on the side permit choosing a reception mode from FM stereo, FM mono and AM as well as enabling a high-cut filter with the TONE switch engaged. The top holds the power switch as the third and last slider, tuning and volume adjustment are done via two thumbwheels, and last but not least there's the headphone jack. The back provides a belt clip and the battery compartment for two AAA cells. It may be interesting to know that assembly was in Korea, the price tag in the States was $20 (about $35 today). With my RP-20, I also received a matching Audio Technica carry case made of robust artificial leather.

• Tuning ranges: FM 87.5 ... 108 MHz, AM ~520...1620 kHz
• Look and feel: As I said, the edgy appearance and design just scream mid-'80s. Build quality seems good, with fairly low tolerances for the time. The plastic body is quite stable, and while one can see that the unit has seem quite some use and the occasional drop, it is pretty much in perfect working order. (Which isn't bad considering that portable audio devices like this tend to have a fairly rough life.)
• FM reception: While selectivity seems only average (feels like a single lonely 280 kHz ceramic) and precise tuning with the thumbwheel can be somewhat fussy, the practice test showed that the FM frontend not only is very sensitive (the RP-20 managed to get stereo reception for the most part in places where I remember the RF-B11 to struggle and where the G3 occasionally loses reception entirely), but also noticeably less prone to overload than other receivers previously tried (i.e. RF-B11 and iAudio G3), as evidenced by it not getting swamped with intermod in a critical location. (Near a transmitter tower, it also gives up, but the duration of interference when driving by is reduced.) Having had trouble getting clean reception at home, I was quite positively surprised by these findings. Overall, this oldie performs better underway than anything I tried previously. It, however, is not cellphone proof. The datasheet of the IC used for FM IF work states a fairly low AM rejection, which seems to show there.
• AM reception: Sensitivity seems quite decent, as is directionality, but the IF filter is on the wider and less steep side, and whistles starting to appear during late afternoon already do not inspire confidence with regard to strong signal handling. (And indeed, the band is full of whistles and hets.) Audio is a bit on the rough side but at least bass response is not artifically limited. I guess I'll stick to the ICF-7601 and the likes for AM DX.
• Sound quality: While amp hiss is clearly evident when using the low-impedance and highly sensitive K26P (32 ohms, 125 dB SPL per 1Vrms), oldschool high-impedance headphones are driven with ease. The volume pot is fairly touchy with the K26P, but there is little channel imbalance at low volumes and scratchiness on my sample was almost non-existant prior to using it on a regular basis. Pitting the RP-20 against my bedside-fi setup with a Grundig T 7000 tuner of similar vintage (since replaced by a Kenwood KT-80) and a modified BTech BT928 headphone amp using the very same Sennheiser HD420SL, the difference wasn't groundbreaking – the Toshiba may not be quite as full-sounding on the bottom (hard to tell how much credit the amp can take for this), but that's about it. Deemphasis thus appears to be more correct than on the G3 which always seemed somewhat rolled off in the highs to me on FM (presumably there only is one version with 75µs that's sold worldwide, which of course is not ideal for regions using a 50µs deemphasis time constant).
• Power consumption: The good FM frontend and stereo decoder seem to come at a price – judging by remaining charge levels in the rechargeables I use, current draw is in the order of 40 mA or so, not particularly thrifty for an analog. If you plan on using the RP-20 regularly, get some decent AAA rechargeables (I have some older 550 mAh NiMH ones here that had been gathering dust for a long time and needed a few charge/discharge cycles to fully come to life again, as I found out when a pair went out on me after only about 3 hours of playtime, and of course I had no spares at hand) or source some inexpensive alkalines.

Hey, really not bad. Might be a candidate for a filter mod (180, maybe 150 kHz), I'd just have to find out how to open the thing (the back seems clipped on, no screws there – ah, there is one hiding under the removable belt clip). If there is something to complain about, it would be that the AM section with its pitiful overload rejection could have been left out and some more effort invested into the FM section instead (another ceramic filter plus IF amp, maybe).

In comparison, the NIB Grundig City Boy 10 (late '80s, around 1990 perhaps?) that I also bought fared a good bit less well – the channel separate volume sliders are somewhat dodgy and could use a bit of contact cleaner, audio on FM seems to include a bass boost, AM audio shows limited bass response and FM selectivity appeared to be yet worse than on the Toshiba. The red version does, however, look neat, and the included earbuds featuring rather large drivers (I would have guessed 12 mm diameter) sound rather decent indeed, as far as I could judge that without matching foamies (the ones supplied have crumbled to dust due to age). The City Boy 10 was made in Hong Kong.

Being the curious nature that I am, I had to take a peek inside the RP-20 – after undoing the screw, you can lever off the back at the 4 "tabs" with a small flat screwdriver. The interior is all conventional (no SMD thankfully), even if slightly cramped. There's a small ferrite rod no more than 5 cm in length mounted horizontally (as seen in normal operating position, meaning the radio is higher than wide), which makes sense. ICs are TA7687AP and TA7688P plus some MPX chip in SIP that I unfortunately couldn't see properly (the service manual revealed this to be a TA7342P); I could spot a single ceramic filter (E10.7A, blue dot left = ordinary 280 kHz type, 10.67 MHz center freq).

Once I had the service docs, I could only be amazed once more: For its performance level, the circuitry actually is fairly simple. The FM frontend reminds me of what I've seen in Grundig Sonoclock alarm clocks (which admittedly tended to have pretty good FM parts, see below), it employs two tuned (variable) RF circuits (1x pre mixer, 1x osc) plus two fixed resonant circuits in front of the RF preamp and self-oscillating mixer (can give good performance when carefully optimized for dynamic range, as strong signals may cause detuning or even "blow out" the LO signal), for which bipolar transistors in common base topology are used.
A (tuned) IFT follows to support the ceramic filter, but before that we have a bipolar transistor in common collector topology, which I'd assume to be employed as an IF amp but which strangely is used in saturation (base-collector voltage greater than zero) – it could be, however, that base and collector voltage are reversed, at least this would make sense when looking at where the supply voltage comes from (originally a regulator output on the TA7687P). So assuming this is correct, it still looks to have a frequency-selective feedback path. Yikes – I guess I still need to learn a few things about minimalist circuit engineering.
AM filter wise, there is a single tuned IFT plus a second LC circuit, which makes it hardly surprising that selectivity isn't too great there. A standard 2-gang frontend is used, RF amp and mixer are contained within the TA7687AP. There are various capacitors used for lowpass filtering but I haven't spotted a dedicated deemphasis circuit. The output coupling caps are 100µ types, a bit on the smaller side for 32 ohm headphones and candidates for replacement by 220µ ones.
BTW, the service manual is dated May 1984.

Here's some food for the "send pix!!!" crowd: :p

JFTR: S/N 5XS3009304.

↑ Back to table of contents

ITC Julia

Wanted: Operating Instructions for Tramp WE-100 or this set! – Suche immer noch Anleitung (wegen Schaltplan) zum Tramp WE-100 oder diesem Gerät!

The large analog multiband portable once sold as ITC Julia, HGS Tramp WE-100 or Super de Luxe Globe 100 in West Germany seems to be a slightly stripped-down version of the Unitra Eltra Julia Stereo from Poland (naturally the "export version" does not bear any notes on its origin), which was made from about 1979 to 1991 as far as I was able to find out.
At 41.5 x 25 x 11.5 cm and more than 5 kg of heft, it's on the larger side of things as far as portables go, more like luggable (with the handle on top). It currently makes the largest such radio in my collection.
Noteworthy features are 5 presets for FM, bass and treble controls, a long telescopic whip the upper part of which can be tilted to utilise its directionality on FM (certainly inspired by what Grundig used), TA in/out and TB out as DIN jacks (and the input seems to be a "true" DIN input, i.e. current input with very high source impedance on the output instead of the transitionary voltage input / voltage output interpretation common in the West) and a similarly oldschool speaker output; the Polish model also featured FM stereo reception and an external antenna input.
Power input is from mains via a "figure-8" Telefunken jack or from ~12V DC via one of the usual round jacks (same as in ICF-SW7600 and other older Sony models). Something rather confusing is that the main power switch (?) on the back operates differently than one would expect – pressed means power *off*. (Turns out this is a mains/battery switch.) I already had the back open and was looking at the clean interior with neat cable routing and all (the build quality certainly is on par with that of other reputable makes, you do notice a traditional radio manufacturer there) and determining the resistance of the two fuses which of course were perfectly fine. Duh! Oh, and do unclip the plug connecting a cable at the back of the battery compartment.

• Frequency coverage: FM 87.5 .. 108 MHz (switchable to the OIRT band 66...73 MHz, which might be useful for sporadic E band openings), LW 150...285 kHz, MW 520...1605 kHz, 6 SW ranges covering the 49, 41, 31, 25, 19 and 16 meter bands.
• Look and feel: Judging by the hefty weight, I'd assume a metal frame to be hiding under the plastic shell which isn't overly thin but doesn't quite seem stable enough to support the radio on its own. That being said, weight distribution suggests that the speaker alone (a widebander that seems to be 16 x 10 cm in size) contributes quite a bit. I like the layout of the controls, which are nicely grouped near or on the sides – power, volume and tone on the left, headphone jack (domino DIN, which can be used with older headphones having a cable terminated on the corresponding plug, like many of Sennheiser's '80s home cans) under the speaker, FM presets and AFC on the right, and on the narrow sides the large tuning knob on the bottom right (quite a pleasure to use due to its size and easily-reached position) and waveband selection on top. I wouldn't know how to do it much better than that.
• Sound quality: A large set with a wideband speaker that doesn't look exactly microscopic either (as far as I could see from the back – seems it's an elliptic wideband speaker measuring 16 x 10 cm) should have some decent sound, and it does. You, however, have to tweak the bass and particularly treble controls quite a bit, and I would have expected a bit more reach into the deeper bass. (The Signal 700 doesn't do badly at all in comparison, but at higher volume its smaller speaker quickly sounds like being driven to its limits.) It still is obvious that you're listening to a portable, even if a big one. I'd love to build a radio with a *real* properly damped 2-way bass-reflex speaker (driven by two separate Class-D amps with active crossover) one day...
  Having another listen, I don't think that there is lack of bass – provided you don't place the set with its back at a wall, like I did. There is, however, some harshness in the highs, probably a peak of the loudspeaker combined with an ugly amplifier distortion spectrum, plus the lower mids are a bit on the thin side (quite the opposite of German portables with their typically "fat" sound), and there's some honkiness. The bass improves further with the back taken off, which makes sense as there is little volume to work with inside (this also causes the honk, I guess).
  The amps used (might be only one here) are UL1481P, seemingly xref TBA810S.
• FM reception: These sets seem to have trouble with FM sensitivity once in a while, and mine apparently belongs to that group. Not even the strongest locals let the S meter go over 3 of 5, and many stations barely are around 1. I suspect a blown preamp transistor/FET to be the culprit. Even as-is, the set is quite usable, just not a DXer's rig. Selectivity seems quite good. The OIRT band yields little more than faint images of local FM stations, as expected. Some sporadic E might help here. The main tuning dial isn't spot on but not as bad as my Grundig T 7000's (which later turned out to suffer from bad misalignment with the RF circuits being all over the map; better readout accuracy as well as much better sensitivity could be restored by tweaking the trimmer caps).
  There is an additional lit small tuning dial in the shape and size of the S meter (which possibly also used to be lit before the bulb went south) for the FM ranges, which apparently is driven with the tuning voltage to give a hint where the currently used preset is tuned to. Otherwise one would have to have a pretty good idea of which stations are where when adjusting the pots on the back. Incidentally, sensitivity seems slightly diminished when using the presets.
  With the schematic, I've had a look at FM alignment. Tweaking the trimmers (C127 for coverage, C118 on the input tank circuit and C123 on a 2nd tuned circuit – yes, it's a 3-gang frontend) proved to be a PITA since while these were clearly labeled and not too hard to reach, seeing them is quite difficult, so this involved some poking and cursing. Anyway, dial accuracy is good now, and sensitivity picked up noticeably (plus the worse sensitivity when using the presets appears to be gone) but still isn't world class – OIRT certainly seems a lot more sensitive, so I guess the BF195 used as RF amp is blown. But even like this, selectivity is quite good indeed for a portable, RF-1410L level perhaps. Since the signal meter never, ever moves beyond 3 (just like 4 on AM), I guess this could use some adjustment. The whole frontend is neatly shielded, matching the overall high level in terms of build quality.
  Like I said, it's a 3-gang varactor tuned frontend, and the set uses what appears to be three 3-pin 280 kHz ceramic filters. (That explains the "OK" 200 kHz selectivity and good 300 kHz selectivity.) There's an UL1211N IC for FM/AM IF grunt work (xref Sanyo LA1201), and there's a ratio detector.
• LW, MW reception:
  MW shows very good directionality with the built-in 20 cm long ferrite rod, though of course having to rotate the whole set for best reception isn't all that practical. Sensitivity doesn't quite reach the ICF-7601. Selectivity is quite good, maybe ICF-7601 level or a bit wider. There's the occasional whistle indicating unwanted mixing products, pointing towards a design using bipolar transistors as RF amp and/or mixer. With the good audio, strong stations are a pleasure to listen to. The MW dial seems the most accurate of all.
  I didn't do much testing of the longwave part when I got this set, so I recently did a 4-way shootout between the Julia, Panasonic RF-1410L, VEF 206 and my trusty Sony ICF-SW7600G. The Sony was quite a bit more noisy than the others on a weak station, so that was out pretty quickly. The Julia proved to be the least noisy, followed by VEF 206 (OK, a little tweaked at 12 volts) and then RF-1410L with a slightly unfavorable audio response. It also seems a touch more selective than the other two, but there isn't much in it overall. At night, the Julia seemed to have the edge in terms of rejecting MW images and spurious, the VEF was so-so, and the Panasonic and Sony brought up the rear.
• Shortwave reception: Sensitivity again stays behind the ICF-7601's a bit but is by no means bad. Audio on weak stations seems more distorted, a typical problem of older sets with simple AM detectors and limited AGC range. Selectivity is a bit on the wider side but again seems little worse than the ICF-7601's with rather steep skirts, certainly usable. AGC seems a little slow and tends to have trouble with quickly fading stations. Crowded bands in the early evening can overload the frontend, but the overload still is far less severe than on the poor ICF-7601 with its weakish 2nd mixer. Tuning scales are more or less off, 150 kHz is quite possible. There is some backlash, but its amount is acceptable.
  I was unable to determine the intermediate frequency when I originally wrote this review, but with the suspicion that it might be a low-1st-IF dual conversion set, a little LO het chasing was successful and yielded a 1st IF of about 1990 kHz; perhaps this was supposed to be 2.00 MHz in proper alignment. (The old Grundig Satellit models used 1.85 MHz, though frequently 2 MHz was mentioned in literature.) The 2nd IF was a little more tricky to find out, but with the help of the MW and LW bands, where the set operates with single conversion, it was determined as being 465 kHz. Stability is good for an analog set, even on 13 meters it doesn't wander around more than a few hundred Hz (and after some warmup, it tends to stay within +/- 100 Hz) – the bandswitch contacts have a far greater influence here.
  My set appears to suffer from a dirty tuning capacitor on shortwave, and upon opening it (taking off the back requires removing 3 screws on the top and another three underneath the set, access to the bandswitch is possible after removing the two large screws on top, and should taking out the chassis be necessary, the volume and tone control knobs pull off) the tuning cap turned out to be enclosed, so there wasn't much I could do for the time being.

I like this set. It's not quite as good as it could have been, but it's a solid radio that is young enough to be frequently available in good working order on the used market (apart from the FM sensitivity issue, the dirty SW tuning cap and a good bit of dirt that had to be removed, my set has no scratchy pots or other stuff like that), and with its somewhat exotic origin certainly is worth having for a collector. It also shows a few pitfalls of solid-state shortwave receiver designs as they were common up to the mid to late '70s.

In terms of innards, the Julia is definitely a child of the mid/late '70s, it just has to make do without the latest goodies like frequency counters, and seemingly semiconductor availability did not include FETs (something already noticed with GDR-made sets). Unlike some more "serious" models, it only covers a few widely spread bands on shortwave, but at least this leads to good frequency stability there. It's a pity that my sample doesn't work as well as it could. The more I look into this set, the more impressed I am.

Elsewhere on the web I found information about this model having been built by Pathé-Marconi in 1975/76 and sold for 349 DM, but neither is the IF of 460 kHz correct nor does the manufacturer seem very likely, already having been mainly in TV production at the time and apparently preferring somewhat unusual IFs of 472 or even 480 kHz. With the schematic of the Polish set matching this one very well, this theory can safely go to the circular file.

A small note found in the battery compartment of my sample states "JULIA ITC", "86 02" (production date I assume) and a serial number of 01391.

Servicing information on the Unitra Julia Stereo has recently appeared on eserviceinfo.com; the closest to a schematic of the Tramp WE-100 that I found was this (argh!) – man, I'd kill for a scan of that thing (well, maybe not, but you get the idea). Even with the different schematic style and low-resolution pictures the similarity to the Polish set is obvious.

↑ Back to table of contents

Sony ICF-7600A

(Sample obtained used, ca. 2006. Cereal number 269624.)

Intro

If you're familiar with my 7600 pages, then you'll certainly also know this classic model. As a compact shortwave portable with good reception and rather thrifty battery usage, it was quite popular in its day. So how does it fare today? I'll compare it to the later and less costly ICF-7601 with a similar concept.

On a side note, mine came with possibly the oldest alkalines I've seen in the last few years – some Sony branded ones still stating "Made in W. Germany". These must be some 15 years old at least, and they even still were at over 1.4 V per cell and hadn't leaked or anything (which I guess is directly related to the former)! The foam on the battery compartment door didn't hold up equally well, but that seems to be a common problem after that much time.

The receiver itself had apparently been gathering dust (literally) laying on its back for quite some time, there even is some on the tuning dial! It took some patience and a toothbrush to get rid of the stuff in trickier places. Thankfully, apart from a slightly scratchy volume pot, occasionally intermittent NEWS tone switch position and a whip antenna needing the occasional screw tightening (basically the same problem as on the ICF-7601), the set is in fine working order.

Build quality

It's a SONY. As to be expected from an ancestor of the ICF-SW7600G, this is a solidly built radio, right down to the controls, the pull-up telescopic whip with a ring for inductive coupling at the tip (copied with the D1835) and the off switch doubling as a hold slider. There is some creaking when pressing on the top, but otherwise the case is stable. As later models, it features a metal speaker grille. The only thing missing is a stand on the back, and the SW band indicator fields could have been larger.

SW reception

• In terms of sensitivity, results depend on whether you're using a power supply – on batteries alone and with no headphones connected, the 7600A beats the ICF-7601 by a relatively small margin (the latter, however, is quite hissy), but when a power supply is connected, the 7600A gains a lot more and leaves the 7601 in the dust, as I checked on 16m and 19m. (It still doesn't reach the SW7600G/AN-LP1 combo that way, but is very good nonetheless.) This way, it must be the most sensitive portable straight off the whip that I own. (On batteries alone it places behind the SW7600G which in turn is beaten by the SW7600.)
• At the same time, while overload on busy bands still is a big issue (it's a weakness of the receiver concept, wideband IFs always require mixers with good strong signal handling), the dynamic range appears to be noticeably greater than on the newer model which I always felt overloaded rather badly. (The 31m band in the evening is my standard testing candidate here – I could still listen to a weaker station quite well with the whip fully retracted on the '7600A, while the '7601 was unable to get rid of the IM het and give decent reception of the desired station at the same time.) Apparently the 2nd mixer (using a 2SK184 FET instead of a 2SK209-Y, with the main difference probably being the much higher Ids used) performs better.
• Selectivity seems to be a tad worse than on the 7601 or SW7600, the filter skirts appear to be somewhat less steep. It's not a big difference though.
• The difference in audio quality, however, is not subtle – the '7600A is far less hissy and maintains better audio on weak signals, also thanks to the well-working AGC. (You do notice that the '7601 basically is a hot-rodded ICF-4900 after all.) Too bad it doesn't tune 22m.
• Dial accuracy is quite good, with deviations hardly any larger than on the '7601 – compared to ordinary single conversion portables which aren't unlikely to be off by 50 or even 100 or more kHz this already makes the set a lot more workable, allowing a +/-5 kHz or at least +/-10 kHz estimate. There is some more tuning backlash than on the '7601, which with a conventional dial string (and after 20 years) isn't all that surprising.
• An empty 13m band in the late evening brought an unwanted mixing product to my attention, which I suspect to have been a strong 31m band station (which were booming in at that time) inadvertantly caught by the 2nd mixer via some PCB trace. Indeed, the 2nd LO frequency is to be found 455 kHz below the wanted signal on the 1st IF, so images from frequencies 910 kHz below 10.7 MHz +/- 350 kHz are possible. On the ICF-7601, by contrast, the 2nd LO works 455 kHz above the variable 1st IF, which neatly eliminates 49m images on 60m at the cost of slightly worse frequency stability. Incidentally, the LO het of the 7600A showed only very minimal drift (certainly no more than a few 10 Hz in a minute or so) and was very clean-sounding when checked via the 7600G in SSB, while the ICF-7601's was drifting a lot more quickly and also had some slight warble to it.

MW reception

• Much of what's been said for SW also applies here – very sensitive (directionality also is good), quite selective, my only complaint would be occasional appearance of what appears to be shortwave bleedthrough – maybe it would have been better not to connect the MW and SW 2nd mixers directly in parallel but use an IFT for each, as done in the ICF-7601.
• Anyway, the 7600A currently is the hottest AM receiver I have (followed by TRA-2350P, SW7600G, SW7600 and 7601 in this order). This might be a fun rig to use with an inductively coupled AM loop, as long as the frontend tuning isn't thrown off in the same way as when approaching another tuned ferrite antenna.
• At the opposite end of signal strength levels, the ICF-7600A also does well. The high-Q frontend tracking means that 2nd-order intermod is not an issue, and in terms of close-by 3rd-order intermod (interference by channels 9 and 18 kHz down when the big AM loop is tuned to 9 kHz down) it's right at the top of my portables, together with the ICF-SW7600, with the ICF-SW30, ICF-7601 and RP2000 following and the other sets with broadband MW frontends doing even less well.

FM reception

FM is quite (though perhaps not extremely) sensitive, and the selectivity with the two IF filters also is very usable, only the AFC seems a bit hyperactive. Dial accuracy seems fairly good, as far as that's to be judged on the somewhat cramped scale of a model starting at 76 MHz.

Sound

• Among the portables with a 3" (8 cm) speaker that I own, this may be the one with the best speaker sound, at least it beats the more mid-centric ICF-7601 and ICF-SW7600 models which aren't bad-sounding to begin with. (I won't even mention the SW7600G.) The speaker driver reportedly is glued to the front panel, and a fairly large grille allows sound to escape from the back of the unit. The effect of the latter is easily demonstrated by laying the set flat on its back – the sound will get all thin and mid-centric, so don't do that. The 7600A appears to have a fairly flat frequency response overall, with some resonance peak being detectable in the highs (presumably the one found in widebanders near the upper end of the working range) but remarkably little midrange honk. So folks at Sony apparently knew about audio engineering back then, but it got neglected more and more over time and was finally forgotten some time in the early '90s.
• The tone switch knows three positions: H, L and NEWS, with H not altering anything, L adding some lowpass filtering and NEWS employing a lowpass filter with a larger time constant and a highpass filter. There is another first-order lowpass switched on upon selecting SW. I've preferred to use H in most cases so far, but under tough conditions NEWS may be useful, and a somewhat disturbed MW station with music might benefit from L.
• It has to be said that the audio filtering is quite heavy by today's standards, which together with the mono headphone jack means that this is not the set of choice if you're into "hi-fi" AM audio. Things will sound quite muffled over headphones. However, the hiss level is low, with little high-frequency content in particular (this always bugged me about the ICF-SW7600G when using my HD590s).

AC-456C PSU

With the receiver, I also obtained a matching AC-456C power supply (even says Made in Germany, wow). While the multimeter showed an unloaded voltage of little more than the desired 6 V, indicating a regulated PSU, results with the ICF-7600A (or the 7601, for that matter) were not good. MW was quite badly disturbed with periodic ticking (buzzing) noises that I'd attribute to the rectifier. Taking a look inside, what I suspected turned out to be true: No small capacitors across the rectifier diodes to be found, so these will have to be added for the PSU to be of any use; otherwise quite solid construction. In the meantime, the AC-240 obtained with the ICF-SW7600 (and subsequently used with a number of other receivers as well) proved to be working very well.

Conclusion

The verdict? The ICF-7600A(W) may be a bit outdated (a few more bands would have been good, particularly 22m and possibly 60m), but still is a good performer in both reception and audio quality. It's anything but surprising that this model sold well in its day. Modern-day receivers with similar concepts include the Tecsun R-9700DX and Degen DE1107; if anyone would happen to own either of these and ye olde 7600A(W), I'd appreciate some information on how they compare.

Audio noise spectra

Here are some noise / buzz spectra recorded in 2008:

As you can see, audio filtering is quite heavy indeed (no wonder things are slightly muffled), but that contributes to a very good AM selectivity.

↑ Back to table of contents

Grundig Sonoclock 900

The Sonoclock 900 is a fairly fancy FM-only clock radio of 1997 vintage. The clock is radio controlled (via DCF77 in this case), you get a fancy fluorescent display that displays the clock to 1s along with the date (or volume), FM preset and some alarm-related stuff and neatly dims according to ambient brightness.

The radio itself is pretty much all analog, the presets are done like in vintage Euro tuners (you have a pot on the back to adjust the varcap tuning voltage for each preset), just switching is done electronically. The receiver is based on a CXA1019S IC and uses a 2-gang plus bandpass FM frontend, which isn't fancy, but IF filter wise you find two 150 kHz ones! A combo DCF77/FM antenna attaches to the back via some strange DIN jack.

I assume the SC 900 wasn't exactly cheap when new, but clock radios seem to be a bit out of fashion these days so one can pick up good stuff for little money on the used market.

• Operation: This is pretty straightforward, I had all the important functions figured out fairly quickly even without a manual. The snooze and alarm on/off button are not close together as they apparently are with some clock radios.
  A reader kindly informed me about some functions that are not immediately obvious:
  • Hold AL MODE to set time (but not date) manually.
  • Press SNOOZE + AL MODE to select beep or melody alarm (if internal buzzer rather than radio is used).
  • With SNOOZE + + / -, you can select display brightness (highest level not affected by automatic dimming).
  • Hold SNOOZE and PRESET for kind of a display test mode. Press one of the lower buttons then, and you can listen to the 42 different melodies (and, I guess, select a specific one along the way). Press one of the lower buttons again, and you'll return to normal operation.
• FM reception: When adjusting the pots for the presets, you'd better be a safecracker, watchmaker or similarly skilled person, as there is no more than one turn for the whole FM band and the pots don't turn very smoothly either. 300 kHz separation is flawless. I found that the limiting factor was sensitivity given that it's hard to attach a decent antenna (the combo antenna is ugly enough that it's best hidden behind the nightstand). On some stations you notice a background hum, but not on all of them – hey, it's a clock radio and not built to audiophile standards.
• Sound: For something that looks to use a 3" (8 cm) speaker (1 W 8 ohm), it gives nicely full, mellow, room-filling sound. A simple tone control exists that allows reducing the highs in one direction and reducing the bass in the other, a detent marks the center position. If you plug in headphones (which will give mono sound, but thankfully the jack is stereo and wired up such that both earpieces of a stereo headphone will get sound), you notice that there is a bass boost going on. Overall, nice speaker audio, as one is used to from a Grundig.

The Sonoclock 900 is a far cry from the screechy mess with very modest reception that el cheapo clock radios can be. Sonic and reception performance actually mattered when this was constructed, something that does not seem to apply to comparable Sony products for example. If you happen to see a Grundig clock radio of the fancier kind on the used market (and there were quite a few), don't hesitate to give it a closer look, even if the appearance particularly of 25+ year old models can be a bit funky.
Incidentally, the service manual for the Sonoclock 900 is available on the web.

Upon opening my sample to fix some nasty whine that I suspected to come from the transformer, I found that some electrolytics had apparently leaked. That doesn't speak for the quality of those and might account for the occasional hum. The set does get quite warm in operation, so that may be responsible for accelerated aging. When replacing caps, I'd use 105° types. BTW, the inside is a bit crammed, no SMD anywhere in sight but quite a few wires. When assembling and disassembling the thing you may find yourself in need of 3 hands.
The aforementioned whine only appears when the built-in buzzer has been used and persists thereafter. I've only been able to work around this by using radio alarm only.

A more ordinary mid-90s clock radio is the Sonoclock 750 – it only has two manually selected brightness steps for the display and a smaller speaker in a more conventional form-factor, but at least tuning is far more easy and you can also set the clock without having DCF reception. Not bad in reception and sound, but the small speaker does have its limits.

↑ Back to table of contents

Grundig Sonoclock 670SP

The '670SP is a clock radio of 1981 vintage, made in Grundig's factory in Portugal like most of them. Uniquely, it features speech output of the time via a UAA1003 IC. As it would have been impossible to store the large excitation / residual stream in an IC back then, only the LPC coefficients were stored and the resulting speech is metallic but perfectly understandable. (The version sold here in Germany used the -1 version of the chip, there were -2 for French and -3 for English as well.)

The FM-only receiving section is similar to many other better Sonoclocks around that time, with two frontend gangs (varicap tuned to provide 8 presets), a self-oscillating mixer (still a common sight even in upper-class portables around 1970, can give good performance when carefully optimized for dynamic range) and two ceramic IF filters. Reception performance accordingly is quite good, with decent sensitivity and good selectivity.
Setting the presets (memostat/preostat) can be done very precisely as there are many (almost too many) turns from one end of the band to the other.

Setting the clock can be a bit of a pain as adjustments only go forwards, but at least one can reset the time to 00:00:00 by pressing >>, > and SEC at the same time. Either radio or buzzer alarm is possible.
There is a baseline brightness adjustment for the clock display, additionally the brightness is adjusted dynamically according to ambient lighting as measured with a photo diode.

The form-factor of the set is somewhat clunky (as to be seen on the cover page of Grundig's Technische Informationen 5/6-'81 – I wouldn't be able to fit it on my nightstand) but it enabled the use of a 10 cm speaker which gives phenomenal sound for a clock radio, making the '900 (no slouch to begin with) sound fairly honky and tinny in comparison. It's true, they don't make 'em like they used to... (Admittedly I haven't heard a SC890.)

My set isn't quite flawless, still operating in OFF mode (but turning off in either of the alarm modes) and apparently lacking a red acrylic window in front of the clock display. The crackling noises in the audio seemingly disappeared when I cleaned out all the dust inside. At least I have the service docs, as I managed to obtain a whole folder with vintage Sonoclock service information for little money. (For this clock module, however, there seem to be some discrepancies between schematic and PCB layout.)

↑ Back to table of contents

Sony SRF-M40W

The SRF-M40W is a PLL synthesized FM/AM Walkman radio that came out in 1988 and, like many such portable audio devices of this or a greater age that weren't the cheapest in their day, is rarely seen on the used market. It enjoys a very good reputation for FM (and to a lesser extent, AM) reception performance, so does it live up to the hype?

• Build:
  This "pocket radio" measures a whopping 82 x 94.5 x 27.5 mm and thus actually doesn't fit into any but cargo pockets, so one is well advised to use the supplied belt clip. At a specified weight of 190 g with batteries (certainly approaching 200 g with alkalines or rechargeables) it feels quite substantial, which underlines the very good build quality. (My trusty Toshiba RP-20 is noticeably smaller and quite a bit lighter than this brick, and that isn't the very smallest portable radio to begin with!) This set probably wasn't cheap in its day and feels like it. Mine was assembled in Taiwan, probably in the same factory that later produced ICF-SW30s and part of the ICF-SW40s.
• Operation:
  • I didn't even need to take a look into the manual, as operation really is very straightforward. The number of indicators is very limited – the reflective LC display shows the frequency (to 1 kHz in AM, to 100 kHz in FM) and the number of the preset selected (7 for each band, with just as many big rubbery buttons with the same tactile feel) if applicable, and that's it.
  • There is one thing that one may not be too happy with: Once you slide the SRF-M40W into its neat carrying case it becomes impossible to operate the sliders on the left side, which control FM sensitivity (LOCAL probably disconnects the "antenna" = headphone wire from the RF part) and particularly FM stereo / FM mono / AM selection. (You can probably make do without the possibility to enter new presets more easily on the road.) Now if you have the set attached with the belt clip, chances are you can't remove the carrying case without unclipping the radio first, so a band change (or even switching to mono) becomes a fairly annoying procedure. (The RP-20 circumvents this problem by completely disappearing inside its case, i.e. to use the belt clip you have to take it out first – the surface shows it.)
  • It may be worth noting that every keypress is confirmed with a short, quiet beep; a longer beep is heard if you've reached one of the ends of a band, i.e. the set does not wrap around like many other radios with PLL synthesis do.
• AM reception:
  • Daytime sensitivity was found to be about the equal of the RP-20, although unwanted modulation products were not heard. Both sets thus placed behind the ICF-SW30 (still fairly noisy, but at least not distorted with its usual excellent AGC action), which in turn was surpassed by the ICF-7601 and ICF-7600A. Unsurprisingly, this is pretty much in the order of ferrite bar size, so the antenna is the bottleneck – a (tunable) loop antenna suitable for inductive coupling would probably work wonders here.
  • Selectivity is the usual 2-pole ceramic plus IFT one that makes stations faintly appear on their neighboring channels and frequently gives the typical 9 kHz whine at night.
  • The heavy overload observed with the poor RP-20, however, was not seen.
  • I'd say an AM DXer on the go should either stick to the Sangean DT-200V or just lug around a larger set unless a (very) collapsible AM loop or tunable ferrite loopstick booster is available.
• FM reception:
  • Pitting the SRF-M40W against my trusty Toshiba RP-20 again, a general trend is that the stereo switching threshold on the Sony is higher, which means that even with a less disturbed mono signal you may not be able to listen in stereo while it works on the Toshiba.
  • Sensitivity tests were ambiguous – on 107.8 MHz (~ 110 km and IIRC 50 kW, but not pointed in my direction) the Sony gave a less noisy mono signal while the Toshiba was able to switch to stereo, but on 88.7 the Toshiba generally was a little better. Perhaps a little touch-up of the FM tracking on the RP-20 would be a good idea. I tried to hold the headphone cable in the same way for both sets and looked out for optimum positioning. The next day, I compared the two again on the 107.8 station, this time with the K26P and its shorter and easier to handle cable, and now I had a hard time telling which of the two was giving the better reception provided the SRF-M40W bothered to switch into stereo; another cable orientation resulted in a somewhat weaker but less multipath disturbed signal, which the RP-20 used to give better stereo reception while the Sony remained in mono.
  • Selectivity is close to equal on both sets, with a slight nod finally going to the SRF-M40W. The RP-20 is a little hampered by its overactive AFC, but in the end problem stations were a little more disturbed overall. (Should the Sony turn out to use two 280 kHz ceramics and no IFT, this would mean that one 280 kHz ceramic and an IFT were only a little behind in terms of selectivity.) Neither set can be considered extremely selective, swapping the ceramic filter(s) for something narrower (180 and 150 kHz ones are easily available – note that if there is more than one used, they must be matched WRT center frequency to achieve low distortion and insertion loss) followed by an IF/detector realignment would improve things quite a bit.
  • Strong signal handling has yet to be evaluated on my usual test course.
• Sound quality:
  • When comparing both the SRF-M40W and RP-20 to my bedside-fi setup with a (slightly modified) Kenwood KT-80 FM tuner on a dipole with a modified BTech BT928 headphone amplifier using the same vintage Sennheiser HD420SL headphones, the Toshiba turns out to be a little sharper sounding in the highs (and a touch less full on the bottom) while the Sony seems noticeably rolled off there. Sony has had this nasty habit of supplying the whole world with portables using 75 µs deemphasis in the FM part (even though this is only needed in North America and 50 µs is correct elsewhere) to keep production costs down, and I bet this is the cause. Both units exhibit good stereo separation. I'd need to check bass response with low-impedance headphones again to see whether at least 220µF output coupling caps have been used (the Toshiba used 100µF).
  • Using my highly sensitive AKG K26P headphones (rated 125 dB SPL / 1 V or 110 dB/1mW), the Toshiba exhibits noticeable hiss that can occasionally still be noticeable on a not too noisy bus, but unfortunately the Sony is quite a bit more hissy still. The SRF-M40W is about on par with the ICF-SW30 in that regard, with which I tended to use the very insensitive Beyerdynamic DT231 (Sennheiser HD500s or HD570s would also work, but none of the three are what I'd consider outdoor headphones with at least semi-decent isolation). Both sets will drive vintage high-impedance headphones easily, but the portable stuff tends to be low in impedance and quite sensitive, for many portables particularly in the late '90s had very weak outputs. (The sensitivity test with the K26P had to be carried out at uncomfortably high levels to get the reception noise above the amp noise floor and ensure halfway acceptable channel balance on the SRF-M40W.) A possible remedy might be a homebrew adapter cable that connects a headphone channel paralleled with a 10 ohm resistor to the output in series with, say, 47 ohm, though I'm not sure whether the average DIY adapter would withstand daily portable use.
• Battery usage:
  Battery life is specified as 27 hours with carbon zinc cells, and I remember reading that someone had measured 28 mA on FM (that seems reasonable considering current consumption and battery life specs are similar on the ICF-7601 and ICF-7600A). This qualifies as fairly modest battery hunger, particularly considering that the set uses AA cells. I wouldn't be surprised if one set of alkalines lasted well over 50 hours (close to 80 maybe), though FM sensitivity may no longer be that great close to depletion.

The verdict? This is a very good little radio for FM (particularly if you're in the States) which unfortunately is plagued by rather high amp hiss levels when used with today's portable headphones (/earphones/IEMs). An attentuator such as the one suggested above may help but it may not be that easy to find one that is sufficiently rugged and does not alter the sonics significantly (such as a simple series resistor or pot). Can it claim do be the best such radio? I have yet to obtain an old SRF-40W or -30W, but considering that the SRF-M40W apparently doesn't employ the correct deemphasis time constant for this part of the world and is quite hissy, I would hesitate to award it the best.

JFTR: S/N 104185.

↑ Back to table of contents

Roadstar TRA-2350P (Redsun RP2000)

(Sample obtained new, Sept. 2006)

Intro

Redsun Electronics is a fairly new name on the field of shortwave receivers with a factory in Dongguan, China that also produces mini hi-fi systems and boomboxes. The models RP2000 and RP2100 are clearly targeted at the Tecsun BCL2000/3000 a.k.a. Eton/Grundig/Lextronix S-350(DL), the only difference between the two being that the 2100 features 10 presets for each of the 5 bands and has keypad-based tuning and scanning. The Roadstar TRA-2350P which can be bought for about 60 EUR is a rebrand of the RP2000. The RP2100 is also being sold as the Elta 3569 "GLOBESTAR" (typically 70..80 EUR), and on the other side of the pond you can buy a Kaito KA2100 (typically US$129).

The RP2000, measuring 29 x 18.5 x 7 cm (W/H/D) and weighing in at 1.8 kg, is not the smallest set you'll find, but still a nice size that allows comfortable operation and good sound via an apparently 12 cm (5") speaker. The PLL synthesizer employing a tuning knob tunes MW in 9/10 or 1 kHz steps (either 522 to 1620 kHz with 9 kHz steps or 520 to 1710 kHz with 10 kHz steps), shortwave in 3 bands (1711 – 10010 kHz, 9990 – 20010 kHz, 19990 – 29999 kHz) with either 5 or 1 kHz steps and FM (87.5 – 108 MHz) with 100 or 10 kHz (!) steps. It features external antenna inputs (male coaxial PAL jack – 75 ohm for FM, 50 ohm for SW – and 500 ohm wire posts for MW), an IF output for AM (!), a 5-step signal strength indicator, can be run off either AA or D cells in addition to mains (230V/50 Hz in this case) or DC in, and on the AM ranges is a dual conversion set (55.845 MHz / 455 kHz) with two selectable IF bandwidths and an RF gain control. Bass and treble controls are also provided, and you can have either local time or UTC displayed (even during reception if necessary). Quite an impressive feature set at this price level. Accessories supplied in this case were a mains lead and two simple adapters for connecting coaxial cable.

Note: There are reports of apparent cost-cutting in newer samples, specifically left-out shielding for the µP board and FM frontend. This seems to affect RP2000s manufactured January 2008 or later at least. Indications are microprocessor interference (whirring noises) on MW and SW plus extremely strong LO radiation on FM (10.7 MHz below tuned frequency, or above if you can't find it; it should be virtually gone 2 or 3 meters from the receiver, while for cost-cutting affected samples reception is still possible 10 meters away).

Operation

• While you are likely to need the manual for the clock functions, operating the receiver itself is pretty straightforward once you've figured out that the AM wide/narrow/FM mono/stereo switch also controls FM/AM switching.
• The keys do indeed require solid pushes in the right spot and for some more time than usual to engage, a beep (which may become unnerving as it's fairly loud, but can fortunately be disabled by holding the SNOOZE button for a few seconds – sometimes reading the manual does help ;) ) confirms any key action.
• The large frequency display is easy on the eyes even if lighting is not too even, the tuning knob also is a nice size; it partly makes up for its light weight by not turning too easily so the risk of accidentally knocking the frequency is reduced. I'd certainly have preferred a weighted flywheel job, but it's quite good as-is. (The knob is directly stuck onto the shaft of an optical encoder, and that wouldn't be too happy about any significant weight.)
• Tuning: The larger tuning steps tend to get you across a band or from one SW band to the next in acceptable time (big improvement over key-tuned sets, up to about 350 kHz per second on shortwave), and if you really want to get to a more distant band on shortwave right now, the Q.TUNE button is your friend. There is an upper limit for the tuning speed (probably imposed by the PLL which needs some time to lock, or by the microprocessor; a smarter set would engage some kind of flywheel simulation, i.e. increase tuning steps until the user considers tuning fast enough), so spinning the knob overly fast is not required.
  If the end of one of the frequency ranges is reached, the wraparound to the other end is signalled with a double beep (which can not be turned off, while reportedly on the '2100 model it works by pressing ALARM for 3 seconds when the set is off).
• Some tuning noises are present, but with a certain background noise level the ones when using 1 or 5 kHz steps disappear while you can still hear a bit with 9 kHz steps on MW. The noises present when tuning in 100 kHz steps on FM are the only ones I'd consider mildly disturbing. In any case, I prefer a bit of tuning noises to overactive muting (ICF-SW30!!!1) by a long shot.

MW (AM) reception

• Sensitivity approaches that of my ICF-7600A (my hottest MW portable to date) though the latter generally still has noticeably lower noise on weak stations (so the overall result of the Redsun is closer to the ICF-SW7600G's); selectivity in narrow is better (similar to ICF-SW7600G), and playing with the treble control you can get excellent audio on strong "locals", particularly in wide mode.
• By the way, the wide filter, while obviously not being very narrow – you can still get decent copy up to +/- 7 kHz from a strong station, in narrow that's reduced to +/- 3 kHz – does have fairly steep skirts. On shortwave, a strong station can still be heard splattering 10 kHz away, but 15 kHz away usually all traces have vanished. The filter itself appears to be a 6-element 12 kHz job (marked 455FW), which explains the good performance. The narrow filter also is a 6-element ceramic, the kind of nominal 4 kHz filter that measures more like 6 kHz in practice.
• If you're into pirates above 1.6 MHz, there is one problem: With 9 kHz steps selected, the set only tunes up to 1620 kHz, and there is no coverage of 1621 kHz to 1710 kHz at all. Not wanting to do a hard reset, I had to resort to my trusty ICF-SW7600G or even the AR7030.
• I did not notice any 2nd-order intermod, the antenna seems to be tuned as initial boosting attempts with the 7600A failed. (Image rejection is specified as >60 dB, which supports this theory.) Sensitivity is slightly reduced at the edges of the band, which might be due to additional filtering or how the set is aligned.
• Quickly fading signals aren't handled too well by the AGC, it gives a slightly "nervous" impression then. The 7600A fares better.
• Connection of the mains lead seems to influence antenna directivity, apparently acting as kind of a ground connection (this has also been noticed with other receivers).
• MW sensitivity does not differ much between AC and battery operation, save for the effect of the varying supply voltage (generally somewhat less gain on batteries). Touching the whip antenna may introduce external interference during AC or external DC operation, while the set remains entirely unimpressed on batteries.
• Upon closer inspection of pictures, the antenna appears to be no larger than what's used in the usual 7600s. Still, the 7600A yields better SNR, so better performance should be possible. Somehow the extra noise appears to be caught over the antenna – some microprocessor radiation maybe?
• For chasing weak daytime MW DX, another receiver with a mechanically tuned ferrite rod antenna (or a large MW loop, for that matter) can be used as a booster by connecting a coupling loop to the MW antenna input. The coupling loop simply is a piece of wire strung around the "booster rx" a few times, perpendicular to the ferrite rod (or in parallel to the main loop's windings). If no boosting effect is to be obtained, swap the wire ends from antenna to ground and vice versa. In case of a large MW loop, the coupling loop (usually one turn only) may need to be a good bit smaller than the main windings to avoid Q degradation. Redsun and booster should be rotated together.

FM reception

• Again, very good. Selectivity is excellent for a portable (comparing well to the KT-80 and its 180/180/230(GDT) filter combo) even though ultimate rejection isn't as good as on a full-blown FM tuner. (The schematic shows two E10.7S filters, on a photo of RP2100 innards these can be seen as having their red dot on the right side, which means a 180 kHz MS3 type. My Roadstar beats the Panasonic RF-1410L with its two early-'80s 230s.)
• Reportedly it is quite resistant to overload, beating even the (better than average for a portable) DE1103 which is totally overloaded on a roof-mounted antenna where the RP2000 remains unimpressed – in return, the Redsun is not quite as sensitive. Still, it does eventually get into trouble due to the lack of RF AGC.
• Listening with headphones reveals somewhat boosted bass and laid-back highs, now that's why speaker audio is good even with the tone controls in the middle position. Stereo separation is OK but not exceptionally large, seems like there is too much L+R; there also is a slight channel imbalance in stereo only. Since a LA3335M is being used, precise stereo alignment is hardly possible.
• The long whip antenna allows for good directionality, which increases with length. You may find, however, that maximum signal strength is achieved at a half-wavelength setting.
• I thought that the signal strength indicator was pretty useless on FM – until I found out that in fact it's a (de)tuning indicator that will show full strength on the nominal frequency and drop if you tune away.
• I thought I had found a spurious response from a strong local station – turns out that they really transmit on the frequency in another town... Phew.
• Image rejection is given as >46 dB, indicating a 3-gang set, but interestingly the schematic reveals it's a 4-gang unit – that would explain the good practical results. (The setup is tuned bandpass – TA7358 preamp – two tuned circuits – TA7358 mixer, which is the same as in more recent DE1103 samples. Early DE1103s still had to make do with a fixed bandpass in front of the RF preamp.) With a matched set of 110 and 150 kHz filters, the RP2000 would certainly make a good budget DXer. (Admittedly it is typically not hard to obtain a full-blown FM tuner with 4 gangs that takes at least 3 filters here in ol' Europe.)
• This set has its local oscillator on FM running not 10.7 MHz above, but below the reception frequency, as I found out while checking the LO phase / amplitude noise of various sets with my KT-1100. Maybe the LA1260 in this set implements the Japanese detector characteristic (which is inverted from the Western one, seemingly to accomodate just this kind of a setup).

Shortwave reception

• This is one hot portable off the whip – and certainly not only because this is yet longer than the SW7600G's at 111 cm (gives a pretty solid impression, too). If you turn up the RF gain all the way, sensitivity partly comes close to my SW7600G/AN-LP1 combo (and is quite a bit, maybe about 10 dB, better than the SW7600G's with batteries on the whip), which is very respectable. The SW7600G is about as sensitive as the Redsun with the gain knob in the 12 o'clock position when running both on batteries, with the Redsun's sensitivity apparently decreasing with frequency, requiring less gain to match the Sony on 75m and more on 19m.
• On the downside, at times of the day when strong SW signals abound here in Europe, you'll get many unwanted mixing products when running the set on AC, which is why I'd recommend not to turn up RF gain far beyond the 12 o'clock position then. (The extra RF gain does come in handy during battery operation, which means considerably lower signal levels on this and other portables when using their built-in whip antennas, seemingly due to lack of a good RF ground. At full gain, a good level of sensitivity can be maintained, not too far from 12 o'clock on AC or external DC. Sony's solution to this sensitivity problem for e.g. the ICF-SW7600G – which definitely used to be one for me – was offering antennas like the popular AN-LP1 loop.)
• The mixing products mainly appear to be of 2nd order (I am under the impression that the receiver is quite sensitive to these – RF preamp at fault?) while I have not sighted 3rd order ones yet. (Folks in areas with strong MW stations have also been complaining about 2nd harmonics of those appearing, an expression of the same issue.)
  If you manage to connect a selective tunable antenna such as my trusty Sony AN-LP1 loop (or Degen's DE31A or some homebrew shortwave loop), these mixing products all but disappear and reception flat out rocks. (The difference in sensitivity between the whip and the AN-LP1 (in favor of the latter) is not that great on some bands, but on 16 meters it's so large that it isn't even funny. The improvement still is quite considerable on 19 meters. The only bands that are noticeably attenuated are 25m and 31m.) In fact, even a simple antenna tuner / preselector like the ADDX-PRE-1 combats overload effectively even at full RF gain and tends to improve antenna tuning on the higher bands on top of that (apparently it's the high sum signal levels that drive the mixer into nonlinearity, and attenuating some of the strong bands is sufficient to get it back into the mostly linear range again).
• Now of course even the narrow bandwidth still is a hair wider than the SW7600G's (the filter specs are the same, but audio filtering is different and the IFT used appears to be untuned – in one case even the old ITC thingway and the ICF-7601 could find a spot between two stations 10 kHz apart where neither were heard while the Redsun couldn't), and there is no synchronous detection (let alone a notch filter or SSB demodulation, though the latter is possible with an extra add-on BFO now), so this is not the DXer's dream set, but for an ordinary listener it is very nice!
  Image rejection, specified as >40 dB, proved to be good enough to only let images from the very strongest stations appear; still not world-class (which would be some 20 dB better) but quite workable. In the evening, 60 meters generally had more problems with intermod on the whip, but the AN-LP1 cleaned that up nicely and made the remaining traces of images disappear. The ADDX-PRE-1 also has a very positive effect, but being more broadband it can hardly do anything about images. Listeners in parts of the world with lower signal strengths are likely to get better results straight off the whip (though even here there are times when you can have the RF fully cranked up, like at 2am).
• Incidentally, on lower bands like 60 and 75 meters touching the whip boosts signal levels, so presumably it is not matched particularly well there. On high shortwave bands I have noticed some buzzing when using the whip that went away upon unplugging the mains (or using an external antenna).
• Given that 2nd order intermod seems to be a slight problem, I would suggest to use either an improved preamp design or frontend tracking (or both, of course) in future models. The IM products on high frequencies seem to be caused by the RF preamp or first mixer, but I have noticed some at lower frequencies like ~4 MHz that could not be eliminated by enabling the attenuator, which supports my theory of the attenuator being after the RF preamp. (And indeed, a look at the schematic confirms that.)
• What I found odd is that in the lower shortwave ranges you still get quite a lot of background noise even with the RF gain turned all the way back (and the external antenna selected with nothing connected there) – but enable the attenuator, and it's reduced considerably. This background noise is strongest at around 5 MHz, dropping off towards the lower and higher frequencies. This characteristic might point to one of the transistors in the RF amp (maybe Q3) catching noise via its power supply. The noise floor with the attenuator enabled (i.e. generated by 1st mixer and "the rest") is virtually constant over the whole AM range and eventually limits sensitivity on the highest bands (e.g. 10m, where neither RF gain nor attenuator have any influence on background noise).
• DC/DC noise: The internal DC/DC converter running at about 2320..2330 kHz seems to be choked insufficiently (and in fact, its output filtering is 1st-order RC only, while a Sony ICF-SW7600G(R) adds another series choke and capacitor to ground for a 3rd-order filter), as a buzz can be picked up with the whip on this frequency and its harmonics (stronger on even-order), especially when not in vertical position; this seems to originate from the top right of the receiver.
• Spurs: You also find what seems to be a typical set of spurs for a 55.845 MHz / 455 kHz configuration - lonely carriers at 13506/7, 17856..17858, 18160, 27013 and 27465..27569 kHz. These seem to be mixing products involving both 1st and 2nd LO, generated in the 2nd mixer. Spur suppression is not as good as in a Sony ICF-SW7600(G) but better than in the little E100.

Construction

Generally solid (including the telescopic antenna, an oft-lamented issue with Sony receivers), no real complaints here. The fold-out handle flexes a little, the battery springs could be tighter and I would have liked to see a hinged battery door (I always wonder how people manage to lose these things but it happens), and the knobs feel a little cheap, but this is about it. The set has been knocked over and things have been dropped on it a few times, but so far only the whip antenna has suffered a bit. Nonetheless I'd advise to be careful with the tuning knob in particular so as to avoid damaging the encoder which it's directly attached to.

If you should find that some of the knobs are not running smoothly, pull them out a little – they merely slide on (as with other portables) and are not made with the greatest precision imaginable.

Sound

• You can still hear that it's a portable with a plastic cabinet, but by portable standards the 12 cm (5") speaker delivers good, pleasant audio easily surpassing all my sets with an 8 cm (3") speaker. (Pitted against the vintage Grundig Signal 700 of similar size with its 14 x 9 cm elliptic wideband speaker, I found that once the tone controls were tweaked to give a similar response, the Redsun actually reproduces deeper bass. It also does not distort as quickly.) If there is a weakness, it's an audible treble peak that causes some sibilance. The Grundig Concert-Boy 1100 obtained later is definitely better in both smoothness of frequency response and bass, but it also is a larger set with a larger, higher-quality speaker.
• For FM I usually turn up the bass and treble controls a bit only (via headphones, the neutral positions appear to be about 11 o'clock for bass and 1:30 for treble), on AM I tend to dial in more treble (esp. in narrow mode, where I frequently have it turned all the way up), which does not worsen selectivity a whole lot but gives a more natural response. Who says you have to have an AR7030 to enjoy good AM audio? ;) (OK, the RP2000 does not have a very low-distortion product detector as the AOR uses for sync mode and still can't match its highs response, but it's a definite step upward from e.g. my trusty ICF-SW7600G, besides the plain AM detector on the '7030 severely lacks in lowpass filtering and is almost unusable. What I did find irritating about the AOR is its lack of bass, which is definitely being filtered away somewhere.)
• If you have some hi-fi gear (preferably with bass/treble controls) at hand, do try the line-out for AM. The signal seems a little more rough-sounding via the built-in amp, and it's been shown that even for AM a low-distortion audio section is beneficial. (Presumably this has to do with different distortion spectra of AM demodulators vs. amps.)
• If you take a look at the schematic, you'll see that the speaker is run with the TDA2822 audio amp in bridge mode – talk about efficient use of resources. During mains operation the amp runs off about 9 V, and a look into the datasheet confirms the 2.5 W output power spec (or 2.2 W if you want to stay under 1% THD). One can't complain about lack of power buffering either, there's a battery of four 2200µ electrolytics plus 470µ||100n at the amp itself. Originally I experienced quite heavy clipping at high volumes, that seems to be noticeably reduced now, apparently driver (surround) break-in took care of this issue.
• Something that might be annoying is that the output level into headphones is much too high – I am using my least efficient set of 600 ohm cans with this set (vintage Sennheiser HD420SL) to get rid of amp hiss and volume pot imbalance. While the combo works well for me, this issue may cause grief with people having a less comprehensive array of headphones (I happen to be a headphone geek). I would suggest swapping the 22R resistors for 100R and connecting a 22R resistor from each channel to ground, directly in parallel with the headphone – this should drop the output level quite a bit (at least 7.4 dB) while about maintaining output impedance and not degrading bass frequency response much (in fact this would improve for loads of about 80 ohm or less; the -3dB cutoff freq f3 would range between 6.0 Hz and 6.6 Hz for loads of 16 through 600 ohms, which is plenty good enough for a mid-fi device).

Quirks, quarks and quibbles

• The backlight may behave strangely upon changing the power source (it tends to reengage in "permanently on" mode when disconnecting from mains).
• Channel balance at low volumes isn't perfect on my sample, a common problem of inexpensive volume pots.
• Audible transformer hum shouldn't be there, please read up on snubber circuits.
• And of course, no presets – but then if you grew up with an ICF-SW7600G like me (with a total of 10 presets for the whole of LW/MW/SW), you probably won't miss them a lot except on FM. At least the set remembers the last frequency per band.
• You may not be able to connect some RCA plugs to the line out jacks if they employ a thick outer rim, as it's the case with many (older) stock cables.
• Line out channel designation (L/R) is reversed – trust the jacks, not the writing.
• Unplugging the set from mains but leaving the cable in may drain the batteries, so always remove the cable as well.
• On high shortwave frequencies around 20 MHz, interference from the built-in clock may be noticed when using the whip; a fix (parallel C114 with 10n) has been suggested (might work, for a cleaner mod that provides more consistently low impedance I'd suggest swapping C114 for a matching 22n and soldering maybe 2.2µ or 4.7µ in parallel). I don't know whether this is fixed in current production sets.
• See note.

Notes on DRM (Digital Radio Mondiale) reception

• The Roadstar has been successfully used for DRM demodulation using the 455 kHz IF out, with reported SNRs of up to 25 dB (which is decent and perfectly usable, if not extremely exciting), but do note that performance may vary:
• Distortion in the IF section at higher levels seems to be an issue, so be sure to adjust the RF gain as needed. Judging by ear on AM, I would say that one should not turn up much further than necessary for all 5 signal strength bars to appear; in experiments using a Morphy Richards DRM receiver behind an upmixer (to circumvent the lousy MW reception of said rx) it was even necessary to turn down to 4 bars, but it's unclear whether it wasn't actually the upmixer that was acting up first, given the pretty high IF output level (in the hundreds of mV).
• If you are using a common DRM downmixer based on the NE602 or SA612 you must make sure that input signal levels are low enough. The 1 dB compression point for this low-noise mixer originally intended for VHF use (not exactly the right tool for this job) is at about -25 dBm, and saturation is reached at about -18 dBm. This means that you'll want to restrict input levels to about 20 mV, i.e. attenuation for the IF output signal should be some 20..26 dB.
• This set is not likely to be a phase noise king, especially on the higher bands. Because of the high 1st IF, the FM LO input on the synthesizer IC is being used (same as in DE1102 and DE1103).
• If the filter shape should appear to be asymmetric, try shifting the reception frequency a bit. If the impression can be confirmed by ear with the narrow filter, it is likely that either the 2nd IF isn't spot on or doesn't match the filter perfectly. In this case one could try adjusting trimmer capacitor TC5 in the 2nd LO (see service docs below) with suitable non-metallic alignment tools.
• A slight filter asymmetry seems to be normal in this set, presumably due to the non-adjustable tuned IFTs on 1st IF level.
• Trying to tune in the IF output signal with a standalone DRM receiver is not likely to be successful, since at this level the signal will be "upside down" in frequency domain. (This is because the first mixer is an upmixer with f_LO > f_s and thus inverting, and the second mixer is a downmixer with f_LO < f_IF1 and thus non-inverting, in sum inverting.) DREAM can accomodate for this if you tell it to ("flip spectrum" or something like that), but a standalone receiver will not expect this scenario.

Conclusion

Hats off to the engineers at Redsun – this is a really nice radio. One doesn't see a set with consistently high performance regardless of band so frequently, plus it's knobtwiddler friendly (I'll admit right away that I'm not a big fan of nested menu systems and prefer one-control-per-function sets these days), has good audio (the biggest complaint I have about my trusty ICF-SW7600G is that its audio is honky and bass-shy even on headphones, I should really bridge those nasty superfluous undersized coupling caps one day), is generally well made and is very versatile to boot.

Upon closer inspection, it does show a number of shortcomings (partly reminiscent of '70s sets) and compromises made to keep costs down, along with some plain quirks, but taking into account what it does do in its price range I'm not complaining. This set really makes you wonder why receivers of this size are not more common - it certainly improves sound, operation and versatility. Would I recommend the RP2000? For the most part, yes. (Not so sure with one of the newer cost-cutting affected samples though.)

Particularly over here, this set is a worthwile alternative even to normal FM/AM radios – a good performer on both bands it certainly is, particularly regarding selectivity.

Notes

Particularly for shortwave use in areas with crowded shortwave bands filled with strong signals (e.g. Europe, US east coast), I strongly recommend a tunable antenna. I have obtained excellent results with Sony's somewhat hard to obtain AN-LP1, but basically any decent tunable loop of a certain size (why not try building this one?) should give good reception. Even the ADDX-PRE-1, a simple whip antenna tuner / preselector, improves reception noticeably (reduction of overload and improved antenna tuning on the higher bands).

There's a Yahoo! Group for pretty much everything and anything, so it should come as no surprise that there is one for Redsun radios as well. (For inside shots no longer available on the web, check out the files section.)

You can also buy an SSB add-on for the RP2100 (and related sets, of course) named TG37, cost is about $26 shipped to the US. It seems to operate like an oldfashioned BFO, i.e. it generates a carrier with adjustable offset that allows the radio's internal AM detector to make sense of SSB/CW transmissions. Not exactly high-tech but first reports are quite positive.

What's more, a "service manual" (actually a schematic and PCB collection) for the Roadstar unit is available on the interweb. Oh yeah! 8-) Looks like a piece of classic oldschool radio engineering, lotsa standard components but put to good use. They didn't skimp on power buffering either. And of course, there's the usual assortment of not immediately obvious tricks. ;) (Anyone care to explain why they use the AM mixer on the LA1260 in that way? Makeshift IF amp?)
The schematic admittedly isn't the easiest to read, as there are no function blocks marked, usual prefixes on semiconductor components tend to be omitted (e.g. K192 instead of 2SK192, or 4148 instead of 1N4148), smaller capacitors are given in <nne> notation (nn*10^e pF) and there are no DC voltages given. There also is the occasional tpyo (like VRAD being mistyped as VARD), and the AM IF IN pin on the LA1260 being connected to the AGC pin makes little sense, probably the connection was supposed to go to the nearby mixer output either directly or via IFT T17.

I have investigated the intermod problem using some circuit simulation in LTspice (needs files for dual-gate MOSFETs from the LTspice Yahoo! group), and found that:

1. Maximum gain is quite high, approaching 40 dB in places (with, however, a wide adjustment range as noticed in practical use), and
2. Gain is reasonably constant across 75m through 11m, with a slight peak around 6..9 MHz – the area with some of the strongest signals in the evening.

Other sets have far less voltage gain, and the input stages of Sony models ICF-SW7600 and ICF-SW7600G also show reduced gain at lower shortwave frequencies. No wonder the poor buffer JFET and/or mixer in the RP2x00 would break into sweat. The Redsun definitely isn't 40 dB more sensitive than those – maybe 20. I guess the rest has to make up for less gain in the following stages. A bit of experimentation with the amp circuitry showed that an RLC network in parallel with RF choke L6 would achieve a nice V-shaped EQ with minimum sensitivity in the critical areas. Might be worth a shot if one can fit it.

Incidentally, a more full-featured model RP3000 with sideband selectable synchronous detection, SSB, keypad input, 1000 presets and air band coverage (!), along with a simpler "export model" RP3100 sans the sync detection and air band, was expected to come out in 2007 but never materialised. Some RP3100 prototypes appeared and one or two were even sold on eBay, but the project was put on hold with the factory being busy enough with OEM production of foreign designs, and as of late 2008 we might never see mass production. There are, however, rumors about a shrunk-down version of the RP2100 design.

Audio noise spectra

Some audio noise spectra (recorded shortly after those for the RD1220):

AM filtering appears to be quite heavy by default in order to provide good selectivity (the low cut, btw, is not present on FM, IIRC there's a smallish coupling capacitor for that), but as you can see the tone controls (which appear to be of the shelving type) can be tweaked to obtain a flatter passband that is easier to listen to, at the cost of some selectivity.

↑ Back to table of contents

Toshiba RP-2059

An FM/AM pocket radio almost equal in size to the RP-20 (i.e. about 11 x 5.9 x 2.1 cm), the 1989 model RP-2059 is a PLL synthesized receiver (FM 50 kHz, AM 9/10 kHz, 5+5 presets) with clock and alarm functions. It runs off two AAA cells and in terms of fancy features can offer "DBSS", something I refer to as a "cheesy midbass booster". It's not entirely clear to me when Toshiba stopped making radios, but since I've never seen any post-'80s ones from them I assume this must have been one of their last models.

• Ergonomics: While operation is generally fairly straightforward (there even is the luxury of beeps upon tuning, with the pitch depending on the direction), there are two annoying flaws: The power button sticks out right on the front and thus is easily pressed when the radio is floating around among other stuff in a bag (even when inside the RP-20's case) - the RP-20s slider on top (the Sony SRF-M40W also uses one there) makes a lot more sense. Additionally, when putting in batteries after they have been removed for a while, the radio is not unlikely to need a reset, something requiring a pointy device (e.g. a pencil) to push the small button – pretty annoying. Unlike on the RP-20, the belt clip is not made from a flexible plastic but rather from the same used for the case, and that apparently gets brittle with time – the retention nose on the clip broke when removing the clip. (I have probably removed and reattached the RP-20's belt clip more than a dozen times, which uses different fastening and seems better thought-out.) The case itself isn't quite the brick the RP-20's was but still feels quite solid.
• Reception: FM is pretty good, showing noticeably better AM rejection and at least equal sensitivity and overload rejection compared to the RP-20. Selectivity, however, looks to be somewhat worse – maybe they skimped on the filtering, leaving out the IFT to reduce costs. Thankfully, cellphone sensitivity is much reduced. AM doesn't show the overload issues of the RP-20, but as usual selectivity is only average and sensitivity is limited by the small ferrite rod inside.
• Sound: If the RP-20 is somewhat brighter than neutral, the RP-2059 is noticeably rolled off in the highs (not quite as much as the RP-20 with the tone switch to LOW, I think, but more than the SRF-M40W) and thus sounds like your average portable radio (though bass is not an issue). You can enable a cheesy midbass boost, which is not hi-fi but likely to improve the listening experience with notoriously thin in-ears like my new Etymotic ER-6. Background hiss is just as high as with the RP-20 (thus beating the notoriously hissy SRF-M40W), and maximum volume is high enough to drive ancient HD420SLs (600 ohm, 94 dB/mW) to loud levels. Again, the volume pot shows little issues with imbalance.

While the RP-2059 improves on the RP-20 in several respects (AM rejection and EMI sensitivity on FM), there are some annoying ergonomic flaws as well, and you lose some selectivity along with the ability to achieve a hi-fi(-ish) frequency response. Mind you, it would still be on par with or better than a number of current pocket radios of the better kind, but overall it just doesn't knock me off my socks. At that time, pocket radios like this had become common fare (maybe not with alarm, but honestly, that was just late-'80s fashion anyway), and as it happens with mass-market consumer goods, they just don't get as much TLC as when the market is still new.

Ironically, the RP-2059 now is my portable radio of choice to get lugged around along with the trusty iAudio G3. Its sonics fit the Etys quite well, and the improved AM rejection makes the set less fussy on average.

To fix a rattling noise inside, I finally opened up my RP-2059 and took some photos along the way. After removing the screw in the battery compartment, the front part along with the attached PCB can be levered upwards, it's only fixed with several plastic clamps. The microprocessor and main PCB are connected via a ribbon cable on the left side. The set can thus be opened like a book. The µP PCB is largely populated with SMD parts, save for some electrolytics, while the main PCB uses all conventional parts. Main PCB area seems to be a bit less than on the RP-20, and IC and air coil count are down. ICs are KIA8132N and KIA7688P, both obviously equivalents to Toshiba ICs and sourced locally. Judging by IC date codes, my RP-2059 might have been a 1990 sample. In terms of FM IF filters, I sighted a 280 kHz type (E10.7A). The ferrite rod is small as ever. Oh, and the little noisemaker was nothing but the piezo buzzer for the alarm function which had come loose – a little solvent-free super glue fixed that. Opening the case expectedly made it a bit more creaky, but it still feels more solid than the Aiwa CR-D6's (below).

↑ Back to table of contents

Aiwa CR-D6

The Aiwa CR-D6 is a ca. 1991/92 (the ZZF no. on this German sample tells, older sets would have an FTZ no. and newer ones a BZT no.) PLL tuned FM/AM portable radio much the size of my trusty Toshiba RP20 (about 10.9 x 6.0 x 2.0 cm) that runs off two AAA cells. FM is tuned in 100 kHz steps, MW in 9 or 10 kHz steps (selectable via a switch in the battery compartment). 5+5 presets are provided that can used for either FM or AM stations. A mono/stereo switch is provided, you can also enable a bass booster termed "SUPER BASS" via a 3rd position of the power switch. It looks like a belt clip can be attached but my sample didn't come with one.

• Ergonomics:
  Good. Operation is quite straightforward really. The on/off/super bass switch is on top, where it belongs. Presets can be stored either manually or automatically, the latter saves all the stations strong enough to halt the search; they are, however, quickly lost if batteries are removed. Tuning wraps around after it has reached the end of a band. The stereo switch is on the side. The buttons have a rubbery feel to them, the small round ones can be a little hard to push as they don't stick out much.
• Reception:
  Sensitivity on FM seems quite good. Selectivity on the "how far can a very strong station be received" test beat out the RP-2059 by a hair but the SRF-M40W remains better. MW reception is about what can be expected from a set this size, sensitivity is mainly limited by the smallish ferrite rod mounted horizontally and selectivity also is as usual, with not much audio filtering.
• Sound:
  • Sound on FM is similar to the RP-20, good frequency response, maybe a bit on the bright side even (it's a touch brighter than the Toshiba).
  • Stereo channel separation seems good, the stereo threshold also is low.
  • There is quite some wideband amp hiss, clearly more than on either of the Toshibas. Testing with medium-impedance headphones revealed that the noise even on the SRF-M40W is less bothering as it has much less high frequency content. On the K26P the Sony still is a bit ahead but no longer as much (indicating a non-negligible output impedance on the Aiwa). When turning down the volume all the way when an FM station can be received reasonably well, a high-pitched whine can also be heard.
  • Output power is sufficient to drive HD420SLs.
  • Even though my sample looks quite beat-up, no scratchy volume pot. Imbalance also isn't much of an issue.
  • On very sensitive present-day 'phones (e.g. K26P) the pop noise upon powering up can be very loud and annoying.
• Construction:
  While the set has apparently passed the time test well, the 3-layer plastic case as popular back them is a little creaky. The CR-D10 I used to have had a 2-part case but wasn't much better in that regard either. Production of the CR-D6 took place in Korea.

The CR-D6 is a bit of a mixed bag. While FM reception and sound aren't bad, it's quite hissy and the pop noise can be annoying.

While my search for the perfect portable FM/AM radio continues, any suggestions as to which ones to look at are highly welcome. ;) (Sorry folks, no DT-200V(X) to be had here in ol' Europe. :( )

↑ Back to table of contents

Panasonic RF-3700

This model, also referred to as GX700, has been Panasonic's PLL tuned AM/FM/SW offering in terms of ordinary portable radios since the mid-90s. If my radios had nicknames, this one would be called "The Handbag" – it measures 27.4 x 14.9 x 8.8 cm³, weighs 1.1 kg sans batteries (it can be powered by either AC or 4 D cells), and the shape and particularly the somewhat oddly shaped carrying handle remind me of a women's handbag quite a bit. (Admittedly the current Grundig portables can top this by quite a margin.)
The set tunes FM from 87.5 to 108 MHz (in 2 ranges to allow for more stations to be saved) in 50 kHz steps, MW from 522 to 1611 kHz in 9 kHz steps and SW from 5.9 to 18 MHz in 5 kHz steps in my European "EG" version. 8 presets per band are available, for a total of 32. Tuning can be done via either a clicky tuning wheel (a nice touch) or up/down buttons; the latter is generally faster, on SW an additional speed boost is activated after a few seconds (which is needed since there is no meter band choice).
A high/low tone switch is provided, and there's a mono headphone jack, but that's about it for fancy features. Not even the display is lit. There is a mechanical volume level indication, the volume control itself is backwards (i.e. downwards = louder), typically Panasonic.

• Build:
  This doesn't feel like the most expensive set, but not cheap either. The handle may be interestingly shaped, but stable it is. Production was in China. The telescopic antenna measures about 80 cm when fully extended. My sample suffers from a loose AC jack (who knows what happened to that), and while I was inside to investigate the feasibility of a repair (tricky), I also tightened the screws of the speaker a bit (only so much as to eliminate potential rattling, no more). Repair of the AC jack was eventually accomplished by resoldering all of its connections following the removal of control board and power supply board.
• Reception:
  • On FM selectivity isn't very exciting and more on the level of what you could easily fit inside a non-radio handbag, i.e. pocket radio level. A DXer this set is not.
  • On SW sensitivity is quite good and about on par with the ICF-M60SRDS (which isn't quite as deaf as I'd found earlier on), it still compares well to the RP2000 even if the AGC range is not quite large enough. Now of course, the performance of a good radio like the Redsun can be improved further by using a whip antenna tuner (which would throw off the tuned input circuit on a single conversion rig like the Panasonic) or even attaching a better antenna like the AN-LP1, with which it definitely outclassed the others. Selectivity is the usual 2-element plus IFT one, which isn't exciting on mediumwave and not great at all for shortwave. It's about on ICF-M60SRDS level, but the filter shape is not as symmetric (bad impedance matching?).
  • I had hoped that MW sensitivity would be great, but apparently there isn't a big fat ferrite rod in there so it's left behind by my references, including the ICF-M60SRDS. Selectivity is the same as on shortwave, ordinary. Be sure to retract the telescopic antenna, else strange mixing products from shortwave may creep in.
• Sound:
  So if it receives like a handbag, does it sound good at least? Yup. The 10 cm (4") speaker seems to provide a good compromise, as it delivers plenty of midbass punch yet retains good clarity. It looks moderately beefy and seems to be rated at 2.7 ohms, my multimeter claims 3.3 ohms. I can crank up the volume to about 6 until distortion appears; the 1.2 W RMS of output make for quite a racket. The ICF-M60SRDS is a complete and utter joke compared to this, being much more mid-centric in spite of the larger speaker (possibly to provide good volume in spite of the 400 mW maximum output). The last Panasonic I've reviewed here, the RF-B20, also featured a nice mellow tone for its size, so audio quality seems to be valued more highly at Matsushita. (Anyone who can name a portable radio with great speaker sound made by Sony since the mid-90s wins a virtual fortune cookie.)

The verdict? I'm not so sure whether this set is worth the 60..70 EUR it's being sold for, since it receives little better than a handbag, but for its size it does sound very good.

In about 2007, some new Panasonic portables appeared. The RF-3700's successor apparently is the RF-U350. Oh, and "Matsushita Electric Industrial Co. Ltd." is history now, the company is called "Panasonic Corporation" since Oct 2008.

↑ Back to table of contents

Philips AE6775

The AE6775 is an unusually styled pocket AM/FM stereo radio that was available from about 2001 to 2003/2004 and costing some 25..30 EUR. The AE6780 was the same with a small speaker built-in. At 7.9 x 7.9 x 1.4 cm³, it deviates from the usual form-factor, which is not to say that the rest of the design wouldn't: The buttons on top and display are surrounded by turquoise transparent plastic, the rest of the front is covered by silver painted metal (a thin metal sheet fixed with 4 tiny screws, covering a plastic front that has a taped-over hole for an electrolytic capacitor in it). The buttons are a great example of why painting those in silver is a bad idea – if the paint wears off, it looks like poo. The front is not the most scratch resistant either.
The feature list of the PLL synthesized AE6775/00 is relatively short: AM in 9 kHz steps, FM in 50 kHz, stereo headphone out only, bass booster, 5 presets per band, locking function, time display. The power button (thankfully on top of the radio, along with the preset and tuning buttons) cycles between AM, FM and off. It runs off two AAA cells. The battery door is hinged (great), and when removing the batteries, the LCD is turned off and the clock is stopped, probably to retain memories long enough even with a relatively small goldcap (the difference in microprocessor power consumption with and without LCD can be a factor of 20 or more, several hundred µA vs. something in the nA range). There is no belt clip, but the formfactor is quite pocket-friendly (on the downside there's the scratching problem).

• Reception:
  • MW is not particularly sensitive, being left behind by the RP-2059 – quite a bit more noise. Selectivity isn't better than usual.
  • FM is quite sensitive with a stereo threshold that's only a bit higher than on the Toshibas and the CR-D6, selectivity again isn't out of the ordinary.
• Sound quality:
  Frequency response on FM is good, but hiss is on the high side, somewhere between SRF-M40W and CR-D6 when using the K26P and even a touch louder than on the CR-D6 via medium-impedance cans (HD535, 150 ohms). The set will go about as loud as any of my other portables (achieving unconfortably high volumes even with 600 ohm cans at home), so I don't get why it's been commented on as having a fairly low maximum output volume. There isn't much muting, you can clearly hear the tuning frequency change during AM/FM or preset switching. Thankfully with the electronic switching, loud popping noises are not present.

Overall, not a bad pocket radio for FM, but somewhat on the hissy side.

↑ Back to table of contents

Grundig Music Boy 160

The Music Boy 160 is a ca. 1983 domestic portable with the usual coverage of LW, MW, SW and FM, a fairly ordinary model even in its day. Accordingly, bells and whistles are few – there's a tone control, and you can connect a headphone, that's about it. The set takes 5 C cells and can be supplied from mains, it measures about 28 x 15.5 x 6.5 cm when disregarding the carrying handle. At that time, production of these sets had been moved to the Far East, and quality no longer was what it had been before 1980, when portables were still made in Portugal (or Northern Ireland). I do like the brown/silver metallic paint though, doesn't look half bad.

• LW/MW/SW performance:
  In a word, unexciting. Neither sensitivity nor selectivity really knock you off your socks.
• FM performance:
  Not half bad. Selectivity is better than the RF-3700's at least, and sensitivity is not an issue either.
• Sound:
  Decent for a set with a lightweight plastic cabinet and relatively large speaker. A little honky though. Unlike in the Concert Boy 220, there's no foam damping behind the speaker, as a glimpse inside shows.

Not a top performer anywhere, but the sad thing is that today's AM/FM portables aren't really better.

↑ Back to table of contents

Panasonic RF-1410L

The RF-1410L, also called GX10II (my sample says RF-1410LBS on the back), was the top model of the about 1980..1984 (or thereabouts) series of AM/FM portables that also comprised the RF-1405L / GX5II and RF-1403L / GX3II. (Presumably other regions got non-L versions.) The preceding series was RF-1110/05/03LBS (GX10/5/3 – a 1978 catalog for Germany only lists the GX5, however), from 1985 onwards the replacing models were RF-1680/50/30L (GX80/50/30). The RF-1680L came out in an era when everything had to be digital and pushbutton controlled, so even the volume control is! (Tuning still is analog; it nominally uses a 2-way speaker, so it certainly is not a bad set.)
The RF-1350BA appears to have been a more upscale mid-70s relative with MW, FM and 3-band shortwave coverage from 1.6 to 30 MHz and a 12 cm speaker that never made it to Europe.

Now coming back to the RF-1410L, this is an AM/FM set of the more fancy kind which covers LW, MW, SW (49m .. 16 m in one band) and FM. It measures about 29 x 18 x 8 cm (officially 28.9 x 17.7 x 8.0 cm), the whip antenna is an impressive 103 cm in length when extended; nominal weight is given as 2.0 kg. Semiconductors include 3 ICs, 1 FET and 8 transistors.
The speaker is a (fake) 2-way system, there's a tone balance (highs) control, fine tuning is available for the AM ranges (though it makes the most sense on shortwave), there's a switchable AFC (the same switch also toggles an attenuator for the AM ranges, but DX and Local should have been reversed), loudness, funny 5-LED tuning indicator (nothing = all 5 lit, then the middle 3, up to a well-tuned station which lights only the middle LED), and if you've already turned off all the other power-consuming stuff, the "battery saver" makes sound more mid-centric to get more mileage out of the 5 D cells in case you're far from a wall outlet (there's a "figure-8" Telefunken jack, and operation from either 220..240 V or 110..125 V is possible). A line in/out is provided via a DIN jack, and there's a mono 3.5 mm jack for speakers and earphones that says 8 ohms only (perhaps they also meant at least 8 ohms?). Sadly there is no dial light.
The whole set feels solid and well-made. (A look under the hood, however, reveals that the ABS plastic isn't extremely thick.)

My sample wasn't in the very best condition and needed some cleaning badly (the dirt that came off smelled like it had been in use in a fries stand, ugh); while it did play, switches and pots were scratchy and intermittent, with the bandswitch only giving good contact on FM, apparently the only band used for quite some time. A good excuse to finally get some tuner spray, which did help (the push switches proved to be tough though).

• FM reception:
  Good sensitivity and selectivity. It's not quite as good as the RP2000, but not very far off either – certainly respectable for a set of this age. Dial accuracy seems pretty good, it's only about 0.5 MHz low near the low end of the band and pretty much spot on near the high end.
• MW reception:
  • With the attenuator disabled, this set is quite hot on MW. On a side by side comparison with the ICF-7600A, I had to resort to using headphones and adjust the tone control, and then I found that sensitivity was just about equal (save for some more cracks from mains on the Panasonic). The AGC range is smaller, however.
  • You can decrease the level of nearby RFI by grounding the case, e.g. by touching it (the side parts are made from Al and can take a beating) – funnily enough plugging in a DIN cable and laying that on the floor (with the set being on same, obviously) also helps.
  • Selectivity is somewhere between the RP2000's two bandwidths, something around 9 kHz maybe. The filter is, however, not as steep (not more than a 4-element I guess, and there's not too much audio filtering), so 9 kHz hets do appear and 9 kHz separation can be difficult in some cases. Again, dial accuracy is quite good, with readout becoming less precise due to markings being more compressed towards the upper end of the band.
  • Fine tuning does hardly anything at all near the low end of the band and covers about +/- 1 channel near the high end (obviously there's a little variable capacitor in parallel with the main one, actually it seems to be a varicap diode as the fine tuning control is a pot). It's not really needed here.
• LW reception:
  No complaints here either. Separation of the fairly weak 177 kHz (DLF) from the strong 183 kHz (Europe 1) works quite well. No complaints about readout accuracy here either.
• SW reception:
  • This was the most stubborn band to get to work, bandswitch wise. Only applying some contact cleaner restored repeatable operation (and after I found out how to get the stuff inside efficiently, the bandswitch started working perfectly again).
  • General notes: Reception wise, I found pretty much what was to be expected – it does pull in stations decently with frequent whine (5 kHz hets) and a bit of whistles, CW (images, party mixing products) and ghost stations (images and spurious from LO's 2nd harmonic) here and there. The higher you get, the more fussy tuning is, with the fine tuning control not being all that helpful due to the smallish knob (provided one has fastened the main tuning knob so that it sits firmly), even with this being free of backlash (the main tuning knob has what I'd call a "slingshot effect", sometimes frequency even runs in opposite tuning direction first).
  • Selectivity actually is good enough to get decent 5 kHz separation, only the filter skirts aren't overly steep by today's standards.
  • Daytime sensitivity appears to be pretty good, it probably uses an RF preamp and that shows. Sensitivity appears to be about equal to the RP2000's at little more than the 12 o'clock RF gain setting. A little anecdote: On the ICF-SW7600G used with batteries, a weak 19m test station (that was even caught, if with noticeably less satisfying AGC action, by the RF-3700) was inaudible, which left me scratching my head and wondering whether it might be defective, but the SW7600 proved to be no better – until I connected the power supply. With that kind of RF counterweight, the station came in quite well. The Panasonics and the Redsun had been mains operated to begin with, which gave them a distinct advantage (on the ICF-SW7600, close to 20 dB or so).
  • There is quite some wobbling in the LO, it likes to drift to and fro by a few 100 Hz quickly – I found this to be strongly reduced after disabling the LED S-meter, so it appears to be correlated to supply voltage changes. Longer-term drift can be 2 kHz or more in a few minutes but is strongly reduced once the set has warmed up. This generally is just fine for AM use, and by the way not even the big RF-4900 was free of drift on shortwave, so one can't expect any miracles from this far more basic set.
  • Image rejection is unexciting and approaches zero dB near the upper end of the SW range, but that was to be expected in a simple single conversion set. The IF used appears to be 455 kHz.
  • At daytime, you can notice ghost stations on 41m which come straight from 19m, apparently due to the LO's 2nd harmonic – for example, 7400 kHz would show interference from (7400+455)*2-455 kHz = 15255 kHz. This is also possible on 49m (from 25m and 22m). These may generate interference with real stations, and if they don't, they can be identified by being much more fussy to tune than real stations, as of course they move twice as fast. It's issues like this that required careful selection of IFs in low-1st IF dual conversion sets. (But don't think that this would be a problem of the past – the 1990s Grundig YB400 was known to suffer from FM broadcast band interference on the higher shortwave bands, doubtlessly caused by 1st LO harmonics, and the current Sangean PT80 / Grundig YB80 was found to exhibit a voice utility signal in the background across the whole shortwave range, presumably straight from the VHF air band – AM, remember – via the 2nd LO's 2nd harmonic, which doesn't speak for shielding on the 1st IF. This must have been near some airport.)
  • Strong signal handling actually is quite decent, aided by the tuned frontend. It certainly doesn't overload like crazy. When IF-level stages saturate, this is clearly audible and can be fixed by switching to LOCAL.
  • Dial accuracy isn't half-bad but with 5.9 to 18 MHz all crammed into a single nonlinear scale, you can say little more than which band you're in towards the upper end. The coarse logging scale is of no real use, the old trick of attaching a stripe of millimeter paper next to the dial would probably work better but is not easy to do.
  • Overall, SW reception works about as well as can be expected from a set like this; it's certainly good enough for a getting a glimpse of what there is to be had. It shows several typical shortcomings of basic shortwave sets back then – lousy frequency readout, images, some instability – but neither is it deaf nor is the IF filter barn door wide. A number of modern-day cheapie shortwave sets do noticeably less well in terms of raw reception performance!
• Sound quality:
  Smooth and balanced. There's still some "portable radio sound" (upper bass hump and dropoff below) but the mids are quite lovely indeed; a sweep shows a pretty smooth frequency response. (I don't actually like the effect of the loudness all that much, but it's quite useful at very low volumes.) It does not have as much bass punch as the RP2000, but does not suffer from the latter's audible treble peak either. And it goes really loud – the specified 4 watts are certainly believable. As expected with a 12 cm widerange speaker (the tweeter does pretty much nothing at all), the highs are quite directional and thus correct positioning is important. One could also connect an external 8 ohm speaker (preferably of the more sensitive kind, i.e. rather some funky horn speaker than the venerable LS3/5a) – or use the tape out, of course.

That was the kind of AM/FM portable I was looking for. It's true, they don't make them like they used to. Ironically, it would be easier to obtain good selectivity and sound today, but nobody does it. Panasonic's AM/FM portables haven't seen a refresh in years, and even the RF-3700 hardly reaches RF-1405L level. One can buy table radios with good sound these days, but I can't think of one with a tweeter offhand. Well, if a proper crossover was too expensive back then, why should this be any better now...

Something I consider interesting is that even after many hours of being turned off, the set still has some voltage left for a brief emission of noise (on FM) when turned on with no mains connected, sometimes it's even sufficient to get some audio of the tuned station. Even after a few days there still is a bit of life left. I guess the big 2200µ buffer electrolytic must have seriously low leakage current and thus be in excellent shape for its age, a quick estimate assuming exponential discharge gives something around 1µA, probably less. It is not unusual to find new ones with specified leakage currents in the single-digit µA range.

Taking a look inside, we find a main board that is smaller than the RF-1405L's, but more crowded, and there's an extra power supply PCB (filled with a transformer, rectifier diodes plus bypass Cs and a fat 2200µ 10V capacitor). The speakers are made by Mitsubishi (Diatone), as it's apparently still the case with the RF-3700; the low/mid driver is a 12 cm type (EAS_12P83GKG, 8 ohm) and the tweeter a 3 cm (EAS 3PH022A-H, 4 ohm, probably cone). (The lesser RF-1405L had to make do with a single 10 cm driver.) The tweeter is dated "80.5.9". The crossover is simple, a 2.2µ capacitor in series with the tweeter. That makes a -3 dB frequency of... 18 kHz?! On to some listening tests. Nice resonance at 150 Hz, I think the case could use some soundproofing and damping and stuff. A sweep shows a fairly even frequency response. 14 kHz still seems to come from the main speaker, same for 15.5 kHz (which is about the maximum for me these days – I used to be able to hear up to 17.5 not so long ago). I would have sworn that directionality in the highs was reduced vs. e.g. the Redsun RP2000 – I love this placebo effect... *grmbl* Anyway, what did they include the tweeter for (and even make sure the baffle is tight around it) if it basically doesn't do anything? Bragging rights? I guess so. The low/mid (or actually wideband) driver seems to perform excellently, but honestly I would have expected a crossover frequency of more like 6 kHz... Perhaps a more fancy crossover was initially planned but then fell victim to cost-cutting – however, I tried 4.7µ in parallel to the other capacitor and the tweeter still didn't say much, if anything.
The audio amp is a Sanyo LA4125, with the supply voltage of apparently 9 V off mains it can achieve an output power of up to 7.7 W into a 4 ohm load in bridge mode; into 8 ohms that'll still be at least the 2.4 watts that can be put out in stereo mode into 4 ohms, but most likely more. (Found a catalog scan, this says 4 watts.)
I also sighted one ceramic IF filter, E10.7S, red dot on the left, i.e. an MS2 type with 230 kHz. Assuming that there's another hiding underneath a nearby shielded section, this would explain the selectivity results – the Redsun uses 180 kHz (MS3) types, which are a good compromise between selectivity and price these days.

This is a fairly service-friendly set. Many connections are merely plugged in, and after removing five screws (the reddish ones) you can pull out the receiver chassis. While the bandswitch can be accessed with the chassis in, cleaning the pots pretty much requires removing it. The push switches are tough to clean, as any available openings are very small.

When opening the set for the nth time to apply some contact cleaner, I noticed that one of the standoffs that hold the screws had broken off. Oops – a case for super glue. Don't mix up the screws and do not tighten them too much...

I've modified my RF-1410L with some bitumen damping mat here and there (not too much space inside), and some foam was added to the speaker on 3 sides for damping (thankfully adhesive tape exists). I was, however, unable to get rid of a buzzing resonance in the bass that grows with volume and lower frequencies in a nonlinear manner – either that's the speaker driver hitting its excursion limit, the nearby dial string or the amplifier running out of juice due to unstable supply voltage. Anyway, total weight seems to have increased a bit, and when tapping against the case, it usually doesn't sound hollow. Sonic improvements? I feel like this brought out the highs more and cleaned up some hollowness in the mids. Even before, sound was more detailed than average. Digging out the unmodified RF-3700 and comparing the two, it's no contest.

Inside the battery compartment (which apparently never saw any batteries), there's a hardly readable serial number sticker saying something like 1BCPB08859.

Useful cleaning utilities are a normal and microfiber cloth, some mild detergent and classic H2O, and something that I always tend to forget until I read Jay Allen's RF-2200 restoration article again, an old toothbrush. The venerable dental cleaning tool is highly useful for removing dirt in more obscure locations. In this case, there was some crud in the speaker grille (a wire mesh grille over cloth apparently) that I hadn't been able to remove even when most of the rest was already cleaned up – a wettened nominally soft toothbrush did the trick, for the most part at least, restoring a much nicer look. It was also helpful when cleaning the knobs with their many grooves. The scratchy dial window could be improved with some display polish but still is far from perfect.

↑ Back to table of contents

Homebrew AM (MW) loop

OK, so it's not a receiver, being completely passive and all – but for the MW DXer, this is one of the most useful tools available. I have experimented with inductive coupling to tuned ferrite rods inside other radios in the past, with certainly noteworthy results, but this thing should be called "AM loupe". Technically, it's a square box loop with a side length that came out as 94 cm (37"), with 9 windings of fairly thick braided wire (conductor dia. ~ 3.0 mm) in about 1 cm (0.4") distance; it uses one gang of a 2x 500pF + 50 kOhm air variable capacitor for tuning. Yes, that is not exactly tiny and I wouldn't want it to be any larger, but it's hard to beat raw aperture. There's still a bit much paper tape in use, but hey, the looks aren't that important here. The woodwork and glueing took about a day, the whole loop was pretty much finished a day later. Admittedly my dad did most of the work, he's a tad more handy...

The loop tunes about the whole MW band, perhaps one winding could even be removed, but in any case Q doesn't seem to be extremely high (the wire was only chosen because we still had close to 40m of it gathering dust) so it gives a good boost beyond 1.6 MHz and even still on 160 meters. When correctly tuned, the increase in signal strength over using just the built-in ferrite rod of test receivers next to or inside the loop is quite dramatic, I would guess 20 dB or so. (Man-made noise from further away rises equally, of course.) Even notoriously antenna-limited pocket radios grow really big ears at daytime. I had not seen (or heard) something like this before. At the same time, the bandwidth is low enough to improve intermodulation considerably as well. In the evening, the level of a correctly tuned stronger station can be so high that e.g. the Redsun RP2000 is overloaded and distorts. Of course, it is also possible to provoke intermod by mistuning.

The best catch I've had with this antenna was RNE1 from Madrid, Spain on 585 kHz (500 kW) here in southern Germany in the middle of the day in April of 2007. Yes, the signal was very weak and occasionally disappearing into noise, and it is to be assumed that atmospheric conditions were favoring propagation, but I would never have thought that something like this would be possible in the afternoon when one would expect maximum D layer absorption.

↑ Back to table of contents

Sony ICF-SW7600G

(Sample obtained new, approx. Mar 1997; S/N 157792)

Intro

This is not just any receiver, it was my first "real" shortwave rig 10 years ago and has been my reference since then.

The ICF-SW7600G is a pocketbook-size portable almost exactly the size of its predecessor – the actual front has become a touch wider so that the controls on the right don't stick out as much, and the speaker grille is a few millimeters less wide, but that's about it. Changes on the inside are more significant: 1 kHz tuning steps are available on the AM ranges, we get synchronous detection for AM (with the additional bonus of sideband-selective SSB demodulation) as a first in this class, and the German version, now a regular international model, finally has official FM coverage down to 76 MHz and an external antenna jack. Reception ranges thus are 150 .. 29999 kHz AM and 76 .. 108 MHz FM.

There also was a change in price tag – this model sold quite a bit cheaper than the previous one, as a number of previously included accessories were left out. The buyer had to make do with only a carrying case and AN-71 windup antenna, plus the usual manual and "Wave Handbook".

For more details on ther ICF-SW7600G, I'll refer you to the Sony 7600 page. So what do I think of its performance these days?

FM reception

This has never been a strength of CXA1376 based sets, and this one is no exception. While sensitivity of a stock unit isn't bad, it overloads easily in high RF environments (use in cities is likely to require use of the attenuator), and with the stock 280 kHz barn doors pretending to be IF filters, selectivity is not outstanding either (though at least the set should work well with low distortion even in countries with an empty FM band and highly modulated stations). I would recommend an IF filter modification with something like two matched 180 kHz filters (much better already), or perhaps 180 + 150; I have a 110/150 combo and stereo separation is largely gone (plus they do not seem to be matched very well, as weak signals come out pretty distorted).

MW reception

• It's a pretty good performer on this band, if not up to the very best in terms of sensitivity (and I mean the very best – for its size and ferrite rod dimensions, it does a perfectly fine job, with only oldschool analogs like the ICF-7600A rendering weak stations with somewhat less background hiss, though how much of that is related to audio frequency response is up for debate).
• Selectivity is very good, but the tight filter means muffled audio (actually the filter itself is not that narrow, but there is fairly steep audio filtering). Not entirely free from images from 910 kHz higher, as the MW/LW input circuit is broadband. For the same reason, it is possible to observe intermod when using a large air loop (2nd as well as 3rd order, the latter can be disturbing when looking for pirates in the 1620..1710 kHz range with an AM loop that only tunes up to slightly under 1.6 MHz with a rather broad peak, like mine); at least a strong station will not distort easily.
• As far as intermod is concerned, 2nd-order is about as strong as on other sets with wideband MW frontends (e.g. ICF-SW7600 and E100), and close-by 3rd-order (as tested using the big AM loop) places about on par with the E100 and behind sets with RF tuning like ICF-7601 and ICF-SW30 which in turn are surpassed by ICF-7600A and, interestingly, ICF-SW7600. Here the attenuator helps only a little at best, so the intermod is probably generated by the 2nd mixer or the preceding IF amp. However, I was apparently also able to generate far-off 3rd-order (or higher-order) intermod, maybe 3f1-f2, which did not appear on the SW7600 (but did on the E100). Now the only major difference in the broadband parts between the SW7600 and SW7600G is the AGC, so possibly the respective attenuator circuitry is at fault.

SW reception

• Sensitivity is only average to good (depending on band), particularly when operated on batteries (the old SW7600 fares somewhat better in the latter case, apparently its antenna matching works better especially on the lower bands like 75m and 90m, but also on the higher ones like 16m und 19m). Presumably this was done to keep 2nd-order intermod down even when using an external power supply / ground. The AN-LP1 antenna does a good job compensating for this; such an external antenna also tends to do away with hum when using an external power supply (an effect that presumably is caused by resonance in the transformer's secondary which should have been eliminated with a snubber circuit).
  Sensitivity drops off quite a bit towards the low end of shortwave – don't expect 160m to be of much use without some special tricks. (This should keep off interference from strong MW stations pretty well, however.) Even around 3 MHz (including 80m), the set still is noticeably less sensitive than its predecessor (or ICF-2001D, let alone the DE1102) on the whip. (A modification mostly fixes that, then yielding about equal sensitivity in SSB.)
• Selectivity with the nominally 4 kHz (more like 6 kHz real) 6-element Murata SFR455I and matching tuned IFT plus steep audio filtering is good. It's a good compromise for the short waves but for the stronger undisturbed stations one wouldn't mind something a little wider.
• If adjacent-channel interference should be annoying, enabling synchronous detection and selecting the less disturbed sideband may help. It tilts sonic balance towards the highs and does tend to increase background noise noticeably (the PLL used for carrier sync is quite simple and I guess phase noise isn't very low), so weak stations are better listened to in AM or even SSB using ECSS.
• Strong signal handling is not outstanding, but decent; the Redsun RP2000 with its 2nd order intermod problem still has more trouble at the same sensitivity level. With a wideband frontend, one cannot expect any miracles – more upscale portables like ICF-SW55 had some input filters (usually octave), the venerable Grundig Satellit 700 even used a tracking preselector circuit (the realignment of which seems to be a good idea), with the ICF-SW77 having even more tuned band filters than the Grundig (they had a lot of trouble getting this to work properly though).
• The AGC does a pretty good job – its action sets in at reasonably low signal levels, and it keeps the IF stages in the linear region and thus distortion low even with rather strong stations (something you take for granted until you come across receivers that stumble here).
• Image rejection with its 2-element crystal filter on the 1st IF is only average, at most 40 dB or so. Without a selective antenna, 60m is not that much fun, and 20m is yet more critical.
• Tuning speed: While direct entry and memories enable quick access to frequencies, a great broadcast bandscanner this set is not. You can step in 1 kHz steps at about 18 steps per second (a bit slow), but continuous tuning in the larger 5 kHz intervals is not possible without activating the automatic search (who actually uses this?). A kingdom for a tuning knob...
• In terms of SSB reception, this is one of the best portables of this class – no other model (save for the 7600GR, of course) allows sideband selective SSB, and audio is of good quality, without the distortion associated with overly long AGC attack times. There is some temperature related drift as the LO used is the PLL VCO that also is in use in sync mode (basically only the PLL action is stopped in SSB mode), but one can live with that. Searching in 1 kHz can be carried out at an acceptable speed and is muting free. With some care, analog fine tuning is possible to ≤10 Hz. On the higher bands (particularly 15m and up) warbling can be noticed with stronger stations, apparently some LO pulling going on there.
• There are few birdies to be noted. You find what seems to be a standard set of spurs for a 55.845 MHz / 455 kHz config, but several of them are quite weak, more so than on other sets. I found 13506 (weak), 17856/7 (weak), 18159 (weak), 27013 (strong), 27467 (strong). No problems with DC/DC converter radiation as found on a few of my other PLL tuned sets. In SSB, you can tune in the VCO (when using the whip) when runs at 4 times the nominal BFO frequency, i.e. zero-beat at 3640 kHz means a nominal 455 kHz BFO.
• The stronger spurs are well suited for observing PLL locking. While the preceding model showed negligible locking times, the ICF-SW7600G appears to lock quite slowly on the high frequencies, wobbling around for well over a second when tuning from a low AM freq to e.g. the 27012.5 kHz spur. (Tuning from a high AM frequency to a 49m band station gives a very short lock-in time, however. Seems like PLL loop bandwidth, which – in light of smaller reference frequency and higher dividers – I'd expect to be smaller than in the older model to begin with, becomes too small on the high frequencies. If then the signal somehow manages to modulate the VCO and throw the PLL out of lock, you get the dreaded SSB wobble. There is some inherent frequency dependence of loop bandwidth in PLL synthesizers, but I guess in this case the VCO's control characteristic may have an even bigger influence. Or maybe it's the DC/DC converter voltage which is not sufficiently stable, which would get the more critical the higher the tuning voltage becomes – there isn't a whole lot of filtering on the +3V line feeding the DC/DC, unlike on the predecessor.)

Sound quality

A big weakness of this receiver.

• First off, there are two 68 nF SMD coupling capacitors (C449, C458) that seem to serve no particular purpose other than needlessly restricting bass response (the other coupling caps are 1µF types). I presumed that originally they were intended for a tone control network and later "forgotten", and maybe that wasn't too far off – somewhat different routing would have enabled flat response in MUSIC and a bass and highs cut in NEWS. Computer geeks may still have some conductive silver paint and a very fine brush, one could try bridging the offending capacitors this way (otherwise some zero ohm resistors in 0805 also do the job). As-is, sound is decidedly lean over headphones.
• Even without this, however, speaker audio still is tinny and honky; a speaker driver this size could do better (and in fact does in any earlier model of the line). Technically speaking, the speaker has just about zero volume to work with, and I have my doubts about the baffle being airtight. And then the openings in the rear cabinet – which tend to boost lower frequencies – are pretty much covered up. The result is lousy low-frequency performance. The old 7600A and 7600DA models had the speaker glued to the front (not that service friendly, but solves the baffle issue), which could also "breathe" towards the back; add some boost for the lower freqs, and sound is much better. The preceding SW7600, not too different from the SW7600G mechanically, had a larger coupling capacitor (470 µF vs. 220 µF) and a 100 µH choke in series with the speaker at least.
  Further improvements pretty much require a larger cabinet though – higher volume, larger baffle area to reduce low-frequency cancellation (the trick with the openings in the rear seems to work well when actual case volume is inadequate but effectively turns such a radio into something not too far from an open-baffle dipole speaker). The Tivoli Model One also uses a 3" speaker driver, but with a correctly tuned bass reflex cabinet it goes down to about 100 Hz where the 7600GR starts to drop off below some 800 (as measured by Matti Nisula who developed the speaker audio mod for the GR). External speakers highly recommended.
• There is some hiss over the headphone out, not extremely much but to bother me on the HD590 (which with its 100 ohm impedance unfortunately gets the maximum output power given the equally large output impedance). This is about the standard hiss level for a portable shortwave radio.
• While you can reduce the highs further, I wish there was some means to elevate them to compensate for muddy modulation and the low filter bandwidth. The tone control is something that I like very much about the Redsun. The preceding SW7600 also had a more elaborate tone control network that allowed nice and full sound in MUSIC and restricted both ends of the audio spectrum in NEWS. This apparently was stripped down and the set made a "DXer" by default. They could, however, have wired things as suggested in the extended tone mod at zero extra cost.
• The headphone output on my set has developed an imbalance of about 2 dB, which is a bit annoying.

Build/Misc.

• The case itself is rock solid. Weaknesses are the telescopic antennas which tend to develop problems with their rotary joints (seemingly not as sturdy as necessary for an antenna as long and heavy as this one), and the power jack which is only soldered onto the main PCB and when used regularly may need some resoldering once in a while (any force stresses the solder joints directly). Purely electronic failures appear to be extremely rare so far, a good sign. The occasional scratchy volume pot and worn-off lettering is about it.
• Most of the keys have a nice clicky tactile feel and short key travel.
• Presets are sufficient for FM, but too few for the AM ranges (only 10 for LW/MW/SW, plus possibly 2 timer memories). As a result, I've always remained preset-phobic...
• The numerals for the frequency display are fairly small, and unlike in the newer ICF-SW7600GR one cannot display time when the receiver is operating. Moreover, the dim backlight (especially on batteries) doesn't make this the ideal nighttime listening set, given that it doesn't really ensure good readability and can only be activated momentarily.
• Oh, and I still like the looks very much, more so than the successor's in fact.

Bacteria, err, battery life

The 7600G is fairly thrifty on batteries for a PLL set, with current consumption being somewhat less than average (at about 65 mA on shortwave) and not too high a switchoff threshold. You may experience some instability when using synchronous detection at low voltages.

Conclusion

The verdict? Still a nice, well-made travel portable with some unique strengths but also a number of weaknesses. It's nice for listening to some hams, but not quite ideal for broadcast listening, particularly if you're a "band cruiser". I don't use mine a whole lot these days, save for LO het chasing.

Notes

This sample's serial number is 157792, apparently the very last ones in 2001 got a bit past the 300000 mark (I've heard of a 326xxx sample).

The special trick for 160m (tested with a stereo mini-to-mini cable, but a stereo or possibly even mono 3.5mm / 1/8" plug should do):
Tune to a 160m frequency with some activity, e.g. the CW areas. Insert plug into antenna jack and try to find the intermediate (not entirely plugged in) position that results in much higher signal and noise levels – apparently this engages the ferrite rod.
That should already work quite well, but ideally you now have an AM loop of the larger kind that still provides a boost up there (like mine). This should improve reception a good bit further.
It wouldn't have been too bad an idea to use the ferrite rod up to, say, 2.5 MHz instead of switching to the whip at 1711 kHz (which does limit MW intermod up there, but like stated is equally bad for signal levels).
The only downside I could find was a fairly strong internally generated het at about 1822 kHz that apparently couples in via the ferrite rod. Not sure how it's created. Operating the set inside a compact AM loop, I also noticed that noise around this frequency seems to be picked up with the loop over the right-hand side of the receiver. Middle or left is OK.

Audio noise spectra

Some audio noise spectra (recorded shortly after those for the RD1220):

The SW7600G's response in MUSIC looks like the SW7600's in NEWS, albeit with a flatter passband and steeper filter skirts still, and switching to NEWS only attenuates the highs even further! (Admittedly the changed highs response in MUSIC seems to be mainly due to a tighter filter.) No wonder I generally prefer the older model's (less "communications-grade") audio, the tone control has a greater effect as well. Since the new (cheaper) tone control design did not permit switching in a highpass for NEWS, they just slapped in these additional coupling capacitors which are always in the signal path. Or maybe the routing would have become a mess otherwise.

Here's the effect of the tone mod, which removes the additional coupling capacitors:

You can see that audio response is now pretty much flat down to 50 Hz. Also included is a spectrum of SSB with a 2 kHz BFO offset, which would be a typical ham band listening setup for me and, while not quite as selective, is noticeably easier on the ears. Some of these recordings were done over the headphone output, which shows the channel imbalance there (no idea where that comes from, maybe a bad solder joint at the audio amp).

↑ Back to table of contents

Siemens RK709 G4

(Sample obtained used, Aug. 2007)

Early/mid '90s analog travel portable with clock/alarm function measuring about 14.5 x 9 x 3.5 cm. Single conversion throughout, receives LW, MW, FM (mono/stereo selectable, AFC) plus 75, 49, 41, 31, 25, 22, 19, 16, 13 meters (with generally decent coverage, even if "The Beeb" on 9410 is just outside the 31m range). Runs on 3 AAs, one of which is for clock operation. Clock display can be lit for a few seconds if radio batteries are installed. Telescopic antenna can be pulled out a bit like with Sony sets to enable operation on back (no stand present). Minimum number of connectors, headphone and DC. LEDs for power, tuning indicator and stereo. Looks and feels quite well-made, just the bandswitch is quite small and could be easier to move. Most probably made by Sangean (model=?), matching the absence of a carrying strap. Price unknown, estimate 100 .. 150 DM.
Note: Do not open up unless absolutely necessary (PITA, 2 clamps on top and bottom each), and when doing so loosen center screw only.

1. FM performance: Good sensitivity, decent selectivity but seemingly hampered by AFC. Good channel separation, sound a bit loudness-ish but really not bad. Forced mono is possible.
2. AM performance: Common to all ranges is only mediocre selectivity, with frequent 9/10 kHz hets. Mediumwave sensitivity proved to be disappointing, with some whistles and stuff thrown in, while longwave is surprisingly good in that regard (maybe my set has trouble with MW, contact problems on the bandswitch perhaps); the ferrite rod is 12 cm long.
   Very good shortwave sensitivity (particularly when used with headphones), up to par with the ICF-7601, with dynamic range typical for a single conversion set but still better than the Sony's. Only the ICF-7600A beats them both. AM audio is quite pleasant if there is no interference.
3. Sound quality: No more than average hiss, fairly good sound over headphones. Speaker seems to be a smallish 50 mm type, and the rear vents are covered by the main PCB, so it's not exactly a bass monster...
4. Misc: Best used on batteries – LO gets hum (AM) with external DC (and that's a regulated supply, too), probably power filtering is minimal. Lettering on case almost completely worn off on my set, can't be too durable.

This wouldn't be a bad set if AM selectivity were a little better. Too bad.

Some audio noise spectra (recorded shortly after those for the RD1220):

Don't ask me why frequency response differs in the left and right channel. It is, however, perfectly clear why AM selectivity isn't too exciting – the filter looks like a common 2 or 3 element (+ IFT) job with about 6 kHz but unexciting shape factor, and audio filtering is minimal (which effectively means elevated highs in FM).

↑ Back to table of contents

Opera PLL 205 (Leetac KA-LT203)

(Sample obtained Aug. 2007)

Cheapie Chinese-made PLL travel portable (apparently a Leetac KA-LT203) that used to be sold by discount chains for about 20 EUR in 2006. Covers LW (1 kHz), MW (9/10 kHz), SW (5.95 .. 15.6 MHz, 5 kHz) and FM (50 kHz steps). About 14.5 x 9 x 3 cm, almost equal in size to the RK709, but with a cheaper feel, of course. Runs on 3 AAs or DC. Headphone jack, antenna jack (certainly not standard in this class and a welcome sight) and attenuator provided. 10 presets per band, direct frequency entry plus extra button for selecting meter bands. Gimmicky LED-based stereo bar volume indicator. Momentary display lighting as long as button pushed, flip-out stand on back, battery lid not hinged. This set got my attention because I spotted a 4-element AM filter (LT455HTU) on an internal photo of what seems to be an internally identical set in a different color scheme sold as "GPX DA8". Another variant is called "JGC RWE-56". Based on the venerable CD8132GP IC. A variation with nearly identical looks but MW / FM / TV / weather (NOAA) coverage (no shortwave or longwave) is sold as Kaito KA2031 in the US.

1. FM performance: Very wide filter, so-so sensitivity. Sound over headphones is pretty decent.
2. AM performance: The filter still is on the wide side of things, with an estimated 12 kHz bandwidth, but thankfully it keeps off the dreaded 10 kHz hets pretty well (which was to be hoped!). Sensitivity for the most part is fairly unexciting, bordering on semi-deaf. It picks up a lot on shortwave if you attach an external power supply (or touch the whip during battery operation) but then overloads easily. Connecting the AN-LP1 (mounted at window) results in overloading, maybe a passive tuned loop would work better. Seems like this set could have done with a little more IF gain, say, 10 or even 20 dB. Internal interference only noted on longwave. On mediumwave, my "AM loupe" helps a lot, which was to be expected.
3. Sound quality: Not bad over headphones, with average hiss levels, the speaker is a 50 mm job and sounds just like that (almost identical to RK709).
4. Misc: Mutes during tuning. Volume pot turns backwards. Indicator thingway only shows anything at ear-blasting volumes. Lettering under numeric keys and keypad layout are awkward and take some time getting used to. Up/down keys must be pushed pretty hard. Attenuator doesn't do all that much. Uneven display lighting with a single LED, gives audible *thump* in audio and LO signal when activated.

Well, one could buy far worse junk in the bargain basement category. That doesn't mean, however, that this set would have knocked me off my socks. It is sufficiently selective to listen to the big, powerful stations on shortwave (which is about anything it'll get without some help in the antenna department, something that with an antenna input is easier to accomplish than with other cheapies not having one) without annoying 10 kHz hets. With a big tuned AM loop, it might be a decent start into mediumwave DX, and if you're a very skilled person you might be able to squeeze in an IF amp, plus the LO signal seems to be clean enough for decent DRM SNRs on MW (though not too much on an absolute basis when listening to it), but generally I can't really recommend this model. Ultimately it's the lack of IF gain and thus sensitivity and dynamic range that spoils it for me.

↑ Back to table of contents

VEF 206

The VEF 206 is a portable transistor radio with longwave, mediumwave and shortwave coverage made by Tento in Latvia (then part of the USSR) from about 1973 to 1989; the 206 was an export-only model. This single conversion receiver (IF 465 kHz) uses analog tuning with the scales going from right to left, and band switching is done with a turret/drum tuner. Shortwave coverage includes 120...75m, 60...41m, 31...25m, 19m, 16m und 13m, longwave stretches to beyond 400 kHz and MW to about 1650. Frequency marks are fairly sparse everywhere.

The set measures 30.5 x 24 x 10.5 cm and weighs 2.7 kg. The case is plastic with a "fake leather" type surface at the sides and top. That also is the weak point of an otherwise very rugged set, as it gets brittle with age, and a drop may cause quite some damage. (A moderately tricky glue job was necessary with my sample.) Other than that, the mechanics are very rugged, even the knobs are made of metal (not so good in case of the volume control, since touching it when using an external power supply will cause a hum).

In terms of semiconductors, the whopping number of 10 transistors and 2 diodes is found inside (but remember it's AM only). Funnily enough, it runs off MINUS 9 V DC, + is grounded. (This seems to be common in other old transistor portables as well, like early-'70s Grundigs.) A DC input jack (center positive) is provided, along with a battery compartment for six D cells (R20). The battery compartment door pretty much cannot get lost, being fixed with screws.

The receiver concept: One tuned RF tank circuit, RF preamp, mixer, 7 tuned IF filter circuits with two IF gain stages, 3 AF gain stages.

• General notes on reception:
  • Like with any drum tuner set, the tuner contacts must be clean or it may not say a thing. Cleaning them is pretty much inevitable with a set that has never seen service, and mine belongs to that group. The contacts do not seem to be gold plated as they were in the Grundig Satellits (and those still give trouble!). Some tuner cleaner and a lot of scrubbing are well invested here.
  • Regardless of band, the scales are neither too detailed nor extremely accurate. On the shortwave bands, deviations of 200 kHz are not unusual, and on longwave the readout still is about a channel off. I guess capacitor aging is responsible for this. Realignment is not too much fun as the antenna tank circuit and LO circuit influence each other (and LO tuning also depends on operating voltage), besides I don't have the correct tools for the coil cores.
  • Compared to more modern sets, AGC range is not great (in spite of AGC affecting multiple gain stages), indicating too little overall gain. There is only so much you can do with 10 transistors, I guess. (In particular, it seems like there is one IF gain stage missing. Grundig never used 2 or at most 3 tuned IF circuits between two gain stages in portables of similar technological level, while here you find no less than 4 in one case.)
  • Sensitivity wise, the VEF 206 typically is in a class with the RF-1410L, which is not too shabby. The AM detector seems to do well at low input levels, but it also needs to.
  • Frequency stability is so-so, given that pretty much everything influences the LO (operating voltage, antenna tank circuit, drum tuner).
  • Selectivity with the 6 IF filter circuits is about in a class with the RF-1410L, although 5 kHz separation proved to be noticeably better. As usual with these older sets, filter shape factor is not too great.
• Longwave reception:
  The 20 cm long ferrite rod antenna makes itself felt here – sensitivity is in a class with RF-1410L and ITC (Unitra) Julia, noticeably better than on the Sony ICF-SW7600G (which is not known as the very hottest longwave set). Basically, if there is anything to be heard on a channel, it'll probably be audible. Sensitivity to MW images at night is moderate, neither great nor all that bad.
• Mediumwave reception:
  Sensitivity is currently power supply buzz limited, so the RF-1410L or Roadstar TRA-2350P fare better (the latter also has noticeably better selectivity in narrow). At night, there's the odd whistle, must be spurious signals (lack of shielding, y'see).
• Shortwave reception:
  • The shortwave ranges were a little hit and miss prior to tuner cleaning round number two but tracking and thus sensitivity improved afterwards. Sensitivity compares quite well with other portables of the better kind like the RF-1410L, but the limited AGC range means listening to weak and fading stations does not exactly qualify as fun.
  • Due to the more widely spread bands, tuning is far less fussy than e.g. on the RF-1410L.
  • Being a single conversion set with one antenna tank circuit in terms of frontend selectivity, this is not exactly the most image-proof model ever. Given that, however, it does quite well.
  • One advantage over sets like the Panasonic with a reception range covering 49m through 16m only is that you can actually tune into the tropical bands as well as 75 meters. In fact, I listened to some 60m stations below 5 MHz which came in decently. The dial marks are no big help in finding out where you are, of course.
  • Something that strikes me as remarkable for a set of this type is that I did not encounter any signs of overload. It's not free from images and the odd whistle and other spurious signals, sure, but heavy mixer overload, no.
  • Sensitivity ultimately drops off on the higher ranges, 13m in particular is not particularly hot. That is not unusual for a design of this vintage, showing the transistors' age.
• Sound quality:
  The audio power amplifier uses two identical transistors in a symmetric push-pull configuration with input and output transformers. This still was fairly common in the 1960s (transformerless power amps first showed up in 1961/62 or thereabouts, seemingly a Grundig invention and accompanied by a patent lawsuit in the 1960s), but certainly seems quite oldschool for a set made until the 1980s! The speaker is an elliptic wideband type. An effective tone control of variable-frequency lowpass type is provided that allows adjusting audio response according to reception quality and one's taste. Sound definitely has a "vintage transistor radio" touch, being somewhat midrange-centric (maybe there is some transformer distortion as well). It, however, also makes for good intelligibility, certainly better in critical situations than on the RF-1410L with its fairly muffled AM audio.
• Misc:
  • An external power supply needs to have very clean output to avoid interference on the lower reception ranges. An RF bypass cap directly after the DC input jack would have avoided this. Even stronger buzzing is heard if you touch the band selector knob.
  • A little "tweak" is using the set on 12V DC, which gives a little better sensitivity and has not shown any negative side effects so far.

Overall, this may not be the DXer's dream set, but for what it is (a relatively humble concept by today's standards), it performs remarkably well. Today's radio designers could learn a lesson about parts selection here. (It, however, also illustrates why resonant circuits should be kept reasonably compact, shielding can't hurt and clean PC board layout comes in handy at times. The limited AGC range is very oldschool as well.)

Some docs including a schematic can be found on this page detailing a DRM mod (certainly a brave endeavour given the frequency stability).

↑ Back to table of contents

Redsun RD1220

(Sample obtained slightly used, early 2008, originally bought late 2007)

Intro

I recently (early 2008) managed to obtain an almost new sample of this Chinese shortwave travel portable (the export version with English labels, only the back still shows Chinese text). This is an analog-tuned set with digital frequency readout and clock / alarm functions that measures 16.9 x 10.8 x 3.2 cm and runs on 4 AA cells. MW, FM and shortwave in 9 bands ranging from 75m to 13m are covered. On shortwave, it employs dual conversion, more on that later. Readout accuracy is 1 kHz on the AM bands and 100 kHz on FM.

The RD1220 is positioned against sets like Degen's DE1104 and sits a notch above dual-conversion analogs like DE1107 or Tecsun R9700DX, so it's not quite a "serious" shortwave set yet. However, they didn't skimp on useful and partially unique features – you can disable the AFC for FM, there's an RF gain control for shortwave, an external antenna input for SW and FM and a DX/Local attenuator switch affecting said reception ranges. A 3-step tone control is provided, along with the usual headphone and DC jacks, and a slow charging function for rechargeables (now basically a standard feature in Chinese portables) is also provided.

The set ships in a box with some shaped eggshell carton holding the things inside. Supplied accessories are a power supply (for the Chinese market, so one still needs a mechanical adapter for outlets here in Germany), a nice leatherette carrying pouch (even has a hole for the headphone plug in it), a windup reel antenna (fairly light but solid feeling) and some earduds, err, -buds that I didn't bother to unpack. Apparently some of the usual cheapo 1200 mAh rechargeables are also included, I did not obtain these with my set. I was a bit stumped when I found the carrying strap as a separate item – maybe this had been removed by the previous owner, even though the screws looked untouched.
tquchina (who this set was originally bought from) also includes an 8 cm CD-R with some user manuals on it; the RD1220's is in moderately understandable Chineseglish, and I would recommend trying out the set without looking at it first if you normally get by without one quite well (Real Men[tm] don't read manuals anyway :P).

The current sales price is US$34.99 plus US$23 for shipping. US$24.99 plus shipping will do if you can live without the earbuds and rechargeables, higher-quality counterparts for which you might already own.

Look and feel / Operation

• While closer observation reveals that the plastic used for the case is a bit on the soft side, the set feels very solid in your hands, like a small brick. (The trick is using a lot of screws, as you find out when disassembling the set. The back comes off after removing five including the one for the telescopic aerial and pulling off the knobs at the left after which you can loosen the clips on the right side and slide the back off to the left, then another six screws hold the chassis frame with the main board in place. This is as far as you have to go if you want to attach or remove the carrying strap.) Only the bottom of the case flexes a little where the speaker is, and the stand is a little wobbly when extended. There is a strip of cloth glued to the inside of the stand to keep it from rattling around when not in use. (Only upright and tilted operation seem to be intended, btw, as the set does not lay flat on its back.) The battery compartment door is solid but not hinged.
• The buttons are decently sized and have a good tactile feel, but in operation suffer from the same gremlins known from the RP2000/2100, i.e. they can be quite unresponsive. Setting the time is not that much fun this way, particularly if you've got a microswitch for the minute adjustment that needs quite some persuasion (pressure) to operate, like it's the case on my sample.
• The rotary controls for tuning, SW RF gain and volume all turn smoothly, but are somewhat on the smaller side of things. I wouldn't mind them being a little larger and sticking out further.
• The telescopic whip antenna measures about 81 cm in length. Its base cannot be pulled out, but other than that it gives the impression of being quite robust, just like the RP2000's antenna.
• The display is similar to the RP2000's, with large numerals and a nice soft orange lighting (either momentarily for 8 seconds or permanently) that is bright enough to ensure good readability yet still far from blinding in the dark (I really dislike blue in that regard). Vertical viewing angles are optimized for viewing from slightly below, as when using the set with the flip stand (this is different in the larger set, which is intended for upright operation). While only one LED is used for illumination, the display is far more evenly lit overall than on the RP2000 which uses two (on one side) – that might be sample variation, hard to tell.
  In comparison, my trusty old ICF-SW7600G stinks – small numerals and fairly dim green lighting (momentary only) don't exactly make nighttime use a pleasure – not to mention those sets that don't even have display or dial lighting (ICF-SW30, ICF-7600A, ICF-7601, RK709, ...).
• During radio operation, pressing TIME SET or WORLD/HOME will briefly (~3s) bring up the local or world time display, respectively.

FM reception

• The RD1220 is every bit as selective as the RP2000 and very sensitive, too – a strong showing. (Looks like my RP2000 has 230 kHz filters then, but those still appear to be narrower – and/or implemented more effectively – than the ones used in the Panasonic RF-1410L around 1981.) I guess the overload handling won't quite match the larger set, given that I only spotted one unaligned and two aligned air coils around the frontend IC (making the frontend similar to what a DE1102 has – fixed FM bandpass, then one gang each for LO and tracking, respectively, and I guess the IC is the same TA7358P), but nonetheless, FM performance is impressive for such a set. There seems to be plenty of IF gain since the output level does not drop on weak stations (same as with the RP2000 – maybe it also uses the same LA1260 IF IC, at least the 16 pins would match).
• AFC capture range seems well-chosen, the LO is not pulled much further than +/- 100 kHz. So even when enabled, the AFC is not too much in the way but does its job of keeping the set on frequency when you just want to listen. The LO frequency is about 10 kHz high with AFC on my sample, I guess that's discriminator sample variation or alignment. As to be expected from the workings of such circuitry (which extracts the DC from the discriminator by heavy lowpass filtering of the output signal), the LO signal gets some FM component and wobbles around a bit, as to be seen on a modern spectrum analyzer.
• Frequency coverage is generous, ranging from about 86 to 109 MHz on my sample.
• As expected, the frontend gets into trouble in strong-signal environments, but no more so than the Toshiba RP-2059, which isn't exactly bad to begin with.
• The attenuator seems to act a little too radically on this band. At least this should keep overload to a minimum even on large external antennas.
• You cannot switch to mono manually when using headphones, which might be disturbing for some. The stereo threshold is very low.
• Stereo separation is good, as is sound quality over headphones. The stereo decoder appears to be some SIP-9 IC, maybe something from Toshiba or a Chinese equivalent (the CD7343GS/TA7343AP comes to mind, there also is a nearby trim pot that could serve for VCO adjustment).
• The signal strength indicator actually isn't one on this band – it's a detune indicator that goes down as you tune away from a station's nominal frequency (the minimum is about 2 bars). With no input signal, it shows 4 of 5 bars.
• As typical for Degen and Redsun sets, the LO runs below the reception frequency.
• Gremlins? Well, there is slight interference from the clock on some frequencies.
• Having a R&S FSUP phase noise analyzer at hand, I ran some measurements on the RD1220 just for fun. I had to use some amplification to get the signal to the level required, and if you look closely you may see things at 100, 200, ... kHz which correspond to broadcast stations, but anyway, here are the results. Significant filtering of the AFC signal doesn't seem to set in until 2 or 3 kHz, but as you can see the modulation is quite low to begin with and wouldn't have much influence on frequency response (as already confirmed by ear).

Shortwave reception

• The filter bandwidth makes for good audio quality but is a little on the wide side, if not as wide as the RP2000's "wide" filter choice. A look inside shows this to be an LTWM455H, a nominally 6 kHz 6-element filter that is likely to be about 8..9 kHz wide (-6 dB). (My guess would have been a 6-element 9 kHz or 4-element 6 kHz job, so I wasn't too far off.) 5 kHz separation is mostly possible but with very noticeable hets, 10 kHz separation works well. When pitted against the old Sony ICF-7600A, the filter itself does not seem all that different but the Sony has far more drastic audio filtering where the Redsun's choices are more like slight tweaks. On the other hand, this makes for very nice audio on undisturbed stations, even the RP2000 with tweaked tone controls is not as balanced sounding.
• Sensitivity on batteries is very good at maximum gain, even if the ICF-7600A seems a touch more sensitive still, at least on some bands (I had forgotten how good this thing actually is – but the band coverage sure is tight). Levels drop towards the edges of the reception bands, I would estimate the range with good sensitivity to be about 400..450 kHz wide.
• Absolute band coverage is either about 800 kHz or 680 kHz, depending on which band you're tuned to (see below for the reason). On my set, this means (+/- 5 kHz or so, depending on temperature):
  • 75m: 3615 .. 4295 kHz
  • 60m: 4525 .. 5325 kHz
  • 49m: 5760 .. 6440 kHz
  • 41m: 6820 .. 7620 kHz
  • 31m: 9320 .. 10120 kHz
  • 25m: 11460 .. 12140 kHz
  • 22m: 13320 .. 14120 kHz
  • 19m: 14940 .. 15740 kHz
  • 16m: 17485 .. 18165 kHz
  • 13m: 21310 .. 21990 kHz
• The S-meter is quite conservative and may show no bars or just one with a signal that subjectively is perfectly fine.
• I'm under the impression that the AGC is possibly doing an even better job than in the RP2000 – in any case, it beats the ICF-7600A's which lets weak signals slide away noticeably earlier.
• Tuning is backwards when compared to the other bands, which is a little awkward and takes getting used to. Since they used two separate tuning capacitors (this way one doesn't need to switch anything, which may have side effects or just be unreliable), it shouldn't have been that difficult to reverse rotation for one, at least Sony always did that. However, other such Chinese sets are like this, too.
• Concept wise, the shortwave part is similar to many other dual conversion analogs – fixed antenna tank circuits (one per band), the same wideband 1st IF around 10.7 MHz (with a 280 kHz SFE10.7MA filter), a bunch of crystals for 1st LO signal generation (e.g. 22.5 MHz for 25m), then a 2nd IF of 455 kHz with a conventional LC oscillator as 2nd LO.
  However, someone put in a little extra thought – presumably to optimize image rejection, the 2nd LO runs below the 1st IF on some bands (75m, 49m, 25m, 16m, 13m) and above on others, and while the external antenna input is nothing new in this class of set, the RF gain control to combat the dreaded 2nd mixer overload is a very good idea. (Should even the RF preamp be overloaded, one can still use the attenuator which seems to be located at the input, judging by things getting very noisy with it on – the attenuation is its noise figure, which dominates what comes afterwards.) I tried my AN-LP1, with good results.
• As opposed to the old Sony sets (ICF-7600A and even more so ICF-7601), the dreaded wandering hets on more crowded bands do not occur. When there is overload, it is of the broadband variety that covers the noise level and wanted signals with hectic intermod "mud". The dynamic range itself does not seem to be much greater than on the ICF-7600A, however, so overloading the RD1220 isn't too difficult. (A single very strong station may already be sufficient for reception to be affected across the band when the set is used with an external antenna; the strong 75m DW DRM signal is a good example.) Expect to be using the RF gain control frequently.
• This is my first analog set with dual conversion coverage of 75m, and that shows. Sensitivity on the whip is at least as good as on my other sets (a shootout with the ICF-SW7600, with both on batteries, didn't yield a clear winner), and with my Sony AN-LP1 antenna I've had some nice catches of both weak broadcast and pirate stations at night. 60m generally also works well, but selectivity isn't ideal for a tropical band with quite some interference like this.
• Not too surprisingly, I've encountered very few signs of images. Today's ceramic filters (like the SFE10.7MA on the 1st IF) have a rejection in excess of 40 dB or even 50 dB at +/- 910 kHz (+/- 400 kHz) from the center frequency, typically better below f0, so if as a designer you can make sure that the frequency range with the higher interference potential gets mixed there, you've won. In addition, the band-specific antenna tank circuit would also provide some attenuation of images, particularly on the lower bands.
• There is the usual small amount of backlash in the tuning mechanism (not as much as the ICF-7600A, which I guess shows its age here), and the frequency counter has some lag, so precise tuning can be a little fussy at times and still is best done by ear. Short-term frequency stability is quite good, and there is basically no hand sensitivity, but there can be quite some long-term drift, seemingly due to incorrect or imprecise temperature compensation on the LO circuits (warming up the set by hand accelerates it) – the frequency may wander by as much as 5 kHz over time; I observed a speed of about 4 Hz/s. This usually is not much of a problem (as the frequency display is accurate, so drift is easy to spot) but means that occasional retuning may be necessary.
• The clock-related interference already noticed on FM can also be found on various spots on shortwave in the form of periodic noise, for example at exactly 7 MHz.

MW reception

• This frequency range is not exactly the RD1220's strength. While the IF filter gives good selectivity with reliable 9 kHz het suppression here and the audio is just as nice as on the short waves, daytime sensitivity falls behind the big guns like the ICF-7600A, RP2000 or ICF-SW7600G and is more in line with e.g. the ICF-SW30 (or even a touch behind that) – unsurprisingly so, given that the ferrite rod is only 10 cm (4") long and correspondingly thin; I guess the limited thickness possible ruled out something a little longer. (DE1101 and DE1102 appear to have an equally sized rod antenna, while the RK704 has a 5" one and DE1103, ICF-7600A, RP2000, ICF-SW7600G all have 6" ones. About the largest you'll find in radios is 8", while AM/FM pocket sets may have a cute 2" ferrite rod or even something yet smaller.)
• Best sensitivity, btw, is obtained with the telescopic whip fully retracted and in the rest position. Maybe it shifts ferrite antenna tuning a little, and factory alignment took this into account.
• Another problem are the audible wandering hets that you get on just about every station at daytime – turns out the interfering signals are shortwave stations, caught via a non-negligible 3rd LO harmonic (it could also be that the mixer is basically a switching one). I found this out when finally one of the interfering signals became strong enough for an ID, a German-language station that came in a little below 1420. Sure enough, a cross-check on 6075 identified this as Deutsche Welle, and some pocket calculator exercise revealed that (((1420+455)*3)+455) kHz = 6080 kHz. Higher signal levels in the evening tend to cover these up, but still it doesn't speak for the radio's design if it picks up this much stray shortwave signals. The ICF-7600A may not have been perfect in this regard, but compared to this set it's entirely harmless. In spite of a stronger radiated 3rd LO harmonic, it didn't pick up the DW signal in question at all.
• The S-meter reads even lower than on shortwave, or possibly that's just because of the lower sensitivity.
• Band coverage is about 512 to 1650 kHz on my sample, so only part of the extended AM band (used in some areas) is available.

Audio

• Speaker audio is fine for a set of this size, maybe comparable to the ICF-7601. Sound is a little mid-centric but quite pleasant to listen to.
• With amplification circuitry that is quite likely to be similar to the RP2000's (a TDA2822M is being used, which is a smaller 8-pin version of the 16-pin TDA2822), it is not too surprising that the headphone output is quite hissy even by shortwave portable standards. I had hoped for this set to replace the venerable RP-2059 as my portable FM radio, but without a voltage divider network (like I suggested for the RP2000), one can forget about that. As with the RP2000, my headphones of choice for stationary use are vintage Sennheiser HD420SLs (600 ohms, specified 94 dB / mW butseemingly closer to 90 in reality) which show barely any hiss.
• At least on my sample, the volume pot tracks fairly well, i.e. has good L/R balance even at low volume.
• The tone control is not too exciting, at least it's not very drastic. NORMAL seems to give an approximately flat response (judging with headphones), in MUSIC bass and highs are boosted (goes well with the '420SL), while in NEWS an additional lowpass, seemingly 1st order, is switched in. There is no real voice-centric setting with heavy filtering, which would be helpful in tougher reception situations. This is more of an "audiophile" set, which also has its merits.
• The RD1220 goes plenty loud, just about as loud as the ICF-7601; the ICF-7600A can't keep up by quite a margin, maybe it could use some fresh electrolytics. Unlike in the RP2000, maximum volume is obtained near the upper end of the volume control.
• Unlike the Sony sets or the RP2000 but as in other smaller Chinese portables, a speaker driver with weak drive (small magnet and voice coil) is used, which gives higher efficiency at the expense of higher resonance (generally means less good bass response) and is generally more suited for open-baffle designs. (It is also less expensive to make and takes up less space.)
  There are no markings on the driver, but I would assume that it has an impedance of at least 8 ohms – more voice coil inductance (which means longer wire, preferably thinner to keep moving mass down, thus higher DC resistance) is a common way of compensating for weak magnets and wide air gaps (think production tolerances). Of course a higher-impedance driver also means that an amplifier limited by operating voltage – the usual case in devices like this – will not deliver as much power. That I guess is why the Chinese (at least Degen and Redsun) like to use the bridged (BTL) mode, which about doubles the maximum voltage swing.
• Bummer: Left and right channel are reversed! Now that is dumb – especially if swapping channels on your cans isn't as easily done as with my old Senns. Speaker phase also seems to be inverted when compared to other portables, which indicates a routing problem after the amplifier.
• Also see the audio spectra and accompanying discussion below.

Power

• As already seen on the RP2000, the battery meter goes from three to two bars quickly.
• Apparently there is no super capacitor for microprocessor backup in the RD1220, which means that after removing the batteries, the set will go dead within just a few minutes.
• Power supply rejection seems insufficient. Operation on my usual regulated power supply that rarely gives a set trouble (one exception is the RK709) introduced quite some buzz and hum on the AM ranges, a bit of hum even got into the audio. I might dig up an adapter for the stock wall wart, but the transformer rattling around inside is not exactly confidence inspiring. For now, rechargeables will be the power source of choice.
• The RD1220's power hunger seems to be average. This Chinese review mentions idle currents of 43 mA for the AM ranges and 54 mA for FM, so it's not quite as thrifty on batteries as I thought, but still quite economical on the AM ranges when compared to "full-grown" PLL synthesized dual conversion sets which may have 65 .. 75 mA. (FM is somewhat above average but then so is reception performance – TANSTAAFL.) One set of 1850 mAh Sanyo rechargeables which typically clock in at about 1600 mAh (bought them as NOS, made in 2003) will give me about 20..25 hours of operation, mostly FM, occasionally shortwave with display lighting on, all with headphones.

Misc.

The main board is dated 2006-08-12, so the set was developed at about the same time as the RP2000/2100. Thus it's not too surprising that is has some of the same oddities. As seen on other sets like Degens, the dual-layer board is equipped with surface mount components on one side and through-hole stuff on the other.

Conclusion

So do I like the RD1220? Absolutely. It takes an old concept and breathes new life into it with high flexibility, a superb FM part and a digital frequency display. (It's too bad that MW performance is only average.) As a travel portable for listening to relatively undisturbed shortwave stations in good quality, it gets the job done easily. For mobile usage with headphones, some soldering action will be required.

This reft and light channel mixup on the headphone jack is unfortunate. Things like that do happen once in a while, particularly with a tight schedule as it is frequently the case these days (and the Chinese tend to invest only as much as needed to get things working halfway well, see RP2000/2100 firmware quirks), but something as basic as this should really have been caught in the prototyping stage.

Turns out the RD1220 actually is a keeper. Together with the HD420SL, it is just small enough to be dragged around every day, and that's just what I do did, before actually getting the DE1102. Can't beat having some good music when doing work at the computer. (Note: Matching radio station required. Not provided with set. :P) And when I feel like getting informed, I can tune into DW on 31m or one of the MW locals, with reception quality mainly being limited by an interference-ridden office environment. At home, it meets the criteria for nighttime listening pretty well (reasonably compact, permanent display lighting, stereo headphone out, antenna jack).

These days, the RD1220 has largely been replaced by the DE1102 (mobile FM usage) and E100 (nighttime listening) here and spends most of its time in the assorted radios drawer. That being said, while the RD1220 "barefoot" places behind the E100 (mostly due to its large AM bandwidth), the AN-LP1 / RD1220 combo remains a hot choice for catching those weak 75m/76m stations in good audio quality. (Provided the DW DRM flamethrower is not on air, that is – the Redsun's dynamic range is somewhat limited after all.)

Audio noise spectra

Some audio spectra when receiving noise only:

What we can see is:

1. Deemphasis is off, about +2.7 dB at 8 kHz and about +3.5 dB at 10 kHz (referring to 1 kHz), rising further as frequency increases. At the same time, 1 kHz is already down 1.5 dB from 100 Hz, effectively giving a boost of the lower frequencies. So there already is a slight loudness built-in.
2. There is no dedicated internal highpass filtering present, as frequency response remains flat right down to 20 Hz (even below, actually), even in AM. Neat!
3. There even is a built-in pilot tone filter. It suffers from drawing board syndrome, as the actual center frequency comes out as about 16 kHz instead of 19 kHz (a frequency response measurement in prototype stage would have shown this, but apparently nobody bothered to carry out any), and the Q is not terribly high (plus it starts to influence frequency response at about 10 kHz), but hey, at least they bothered with including one at all.
4. Do not trust your ears when levels may vary. While MUSIC turns on a loudness-type EQ indeed, NEWS seems to switch in an additional 1st-order lowpass to reduce the higher frequencies when compared to NORM – however, it also is louder than NORM (presumably it uses the same EQ circuit as MUSIC in a different configuration), so I was fooled into thinking that in fact a highs boost is enabled!
   Switching from NORM to NEWS attenuates 5 kHz by about 7 dB and 10 kHz by about 13 dB. This is just audible and not groundbreaking by any means.
5. The AM filter is fairly wide indeed. Closer inspection of the plots in RMAA yields -6 dB at close to 5 kHz (subtracting the effect of the audio response), meaning a filter bandwidth of about 10 kHz. At 8 kHz, we've gained another 10 dB of suppression, for a total of 16 dB (off-tuning by 3 kHz already gives a fairly decent 5 kHz het suppression). At 10 kHz, filter response only is 33.5 dB down. (And this is with the effect of the AM detector's filtering, which however does not seem to be significant at audio frequencies.) The plateau between 10 and 15 kHz could be some IF-level noise. Further up, the noise floor of a 16-bit recording is limiting things (I had already cranked the volume all the way up but didn't want to change the carefully adjusted recording level).

↑ Back to table of contents

Degen DE1102

(Sample obtained new, May 2008. S/N AE044834)

Intro

The Degen DE1102 is a radio I had been intrigued by for a while but never gotten around to actually buying. Well, now in 2008 I have one. It's a very compact portable measuring about 14.3 x 8.8 x 2.85 cm³ (yet smaller than the RK709 G4 or PLL 205) that covers mediumwave (522 to 1620 kHz in 9/1 kHz or 10/1 kHz steps, I have not tested whether my version will receive up to 1710 kHz with 10 kHz), shortwave (2998 to 29999 kHz on my sample, 5/1 kHz steps) and FM (70 to 108 MHz in this case with 50/10 kHz steps). On the AM ranges, it works with dual conversion (10.7 MHz / 450 kHz). The set can be powered by three AA cells or via a DC input (6 V, center negative), with a charging function for rechargeables.

My DE1102 is a Thieking & Koch sample, obtained for the princely sum of EUR 69 + 5 shipping after the Scott RXP80 variant appears to be no longer available (which even sold for less than 60 but had adapted firmware with e.g. FM coverage starting at 87.5 MHz). That kind of price is OK (remembering that a directly ordered sample cost about 60 when it came out), unlike the 119 EUR that officially imported samples sold for originally.
My readers in the US might know this model as Kaito KA1102, which sells for $79.95 at Universal Radio (as of 05/2008). The Degen "original" can also be ordered from one of the usual well-reputed Chinese dealers.
November 2009 update: It appears that this model has been phased out or at least pushed back by the manufacturer some time this year – offers have thinned out quite a bit. So if you still want one, better hurry up.

The DE1102 may be small in size, but it's not light on the feature side: Antenna input, switched attenuator (SW and FM only), two AM filter bandwidths, SSB with fine tuning (previously unheard of at this kind of receiver size and price) and a tone switch are present. A total of 190 memories on 10 pages are provided (an auto-memorizing function = ATS is also available), and a total of four LEDs serve as a signal strength display, with one of them being relegated to function as a stereo indicator on FM. Speaking of FM, you can choose between mono and stereo reception (when using headphones, the 2.5" speaker is mono) and enable a bass boost there.

Supplied accessories include a power supply (with the right kind of mains connector in this case), stereo earbuds with pads (which enjoy a good reputation regarding sound quality), and, underneath the plastic inlay holding the radio and the parts just mentioned, a velvety drawstring carrying bag, a wire antenna with a mono 3.5 mm plug, three NiMH rechargeables with an official 1300 mAh and, of course, the manual. The latter is in German only in this case, quite well written and even includes a few tricks not found in the original English language manual; it was authored by MicroConsult GmbH (my German readers may know eBay seller "wittsfeld"). This cooperation explains the "Thieking & Koch" brand, I guess.

Look and feel

• Cute! It's a fairly small radio indeed, with a bit of a high-tech look to it. They only sell the black version here, which I always found to be nicer looking anyway. (Silver is so yesteryear, y'know.)
• As on the DE105, the battery compartment has a hinged lid to prevent loss. The inside is, however, fairly tight for Eneloops, and it doesn't get any better with regular 2500 mAh Sanyo NiMHs.
• The plastic case feels solid and stable, all the parts fit together well with small gaps only. (Degen must have access to some pretty good, modern molding facilities.) Only the fold-out stand seems a little on the thin and flimsy side but gets the job done, and squeezing the front and back of the cabinet where the speaker is yields a bit of flexing (as in other modern-day Chinese sets, the plastic is less heavy and flexes more than the stuff I'm used to from e.g. Sony). Two little round buffers on the back keep the stand from rattling when not in use.
• The DE1102, like a Sony ICF-SW7600(G), can be operated in 3 positions: upright, tilted and flat on its back (though the latter will not allow a vertical position of the telescopic antenna, which does not have a pull-out base).
• No complaints about the whip antenna, which for a set of this size is quite long (~72 cm) and extends and retracts with little effort. It does not have a pull-out base but given the reliability of that on Sony sets (the antenna on my 7600G always tips over and has been doing so for close to 10 years), I'll happily do without.
• Buttons have a positive, clicky feel to them. I wouldn't mind if they required a little less force though.
• The LC display has high contrast and a good viewing angle (better than RD1220) under daylight. The numerals for the hour/minute or first four frequency digits have a nice size, not RD1220 large but noticeably larger than those on the ICF-SW7600G and still a touch larger compared to the ICF-SW30 (the very last frequency digit is smaller in return). Lighting is apparently done from the right side with a blue LED (I'd prefer orange but hey, blue was modern when this set came out); the result could be a more even light distribution. When lit, optimum viewing of the display is from slightly below; it goes blank when viewed from above (similar to RD1220).
• While we're at lighting, both the display and the buttons on the front panel are lit (the latter is a nice touch if you're in the dark; the button labels are molded in as transparent plastic). Light is enabled for about 8 seconds on every button pressed by default. It can be disabled entirely, permanent lighting is possible during mains operation only.
• The DE1102 has a pretty lengthy carrying strap attached to the top left – it's as long as on the ICF-7601 and ICF-7600A and longer than what my other Sony sets have, but all of these are a good bit larger of course! Actually the set is small enough for a strap to make some sense (it's not of much use on larger and heavier ones). It may also be practical for removing the set from the carrying bag into which it goes sideways.

Operation

• If you normally do not need a manual for a radio, this is a set where you won't get around taking a look into it. Operation is partly menu-driven, and while there are some clever solutions, other things just take getting used to. Without a manual, you'll never even figure out how to set the clock (once you know, however, you'll appreciate the direct time entry via the numeric keypad, though as with multiple setup procedures, you have to be pretty quick). While we're at it, the clock can be shown during radio operation by pressing EXIT, and time readout also gives the seconds.
• A typical stumbling stone is the role of pages. The currently selected memory page influences a number of operating parameters. For example, tuning with the larger available steps is performed on pages 0 through 6, while on pages 7 (FM only), 8 (MW only) and 9 (SW only), the smaller steps are being used. Page 9 also allows SSB operation (which can only be activated momentarily when tuned to a shortwave frequency on another page), and ATS is available on page 0. The set also seems to remember the last FM, MW and SW station tuned per page, but that is not too transparent to me – there's about a 50/50 chance that it goes to where I thought it would. Apparently frequencies from pages 7-9 are "carried over" when changing to 0-6, but not vice versa. So if you are listening to a broadcaster in AM with the page set to 0-6 and would like to check a previously visited ham band frequency in SSB on page 9, this works, but assuming you want to fine-tune the reception frequency on said AM station in 1 kHz steps, you won't get around re-entering the frequency (after saving the ham frequency). At least direct frequency input is quite comfortable, with EXIT serving as backspace to correct typos, no further confirmation after the maximum number of digits is reached and extensive auto-completion guesswork.
• An oddity also found on the DE105 is the headphone jack on the right-hand side. It's not dramatic or anything, just unusual if all your other sets have it on the left side. Another oddity is the SSB fine tuning and the attenuator being backwards, i.e. move upwards = frequency or signal strength decreased.
• You have to be pretty quick indeed when changing the batteries – in fact, expect it to take longer than the backup time (which is only in the order of seconds). If you're not too keen on setting the time again (the memories are non-volatile anyway), connect the power supply. (Keep in mind that the set is designed to run off rechargeables that are charged internally. That doesn't help you, however, if you have a good external charger that you'd like to use.) If the set does strange things after a battery change, it probably needs a reset – take them back out for a minute or two, that should do the trick.
• There are no key gremlins (à la Redsun) to be noted. Buttons react like they should, keystrokes are rarely lost.
• While I'd have my doubts about entirely blind operation, basic unsighted operation (e.g. when the set is in a pocket) is entirely possible, with the different button shapes (no less than five) along with a blip on the case next to the number 5 button allowing fairly good orientation. If the lock function is activated, accidental bumping of the volume is not possible either. If, however, you want to temporarily disengage the lock (e.g. in order to change the volume), this is not so easy – it's one of many buttons on the front panel that one has to feel for first, a separate control would be easier to find (but cost more, of course).

Audio quality

• The 2.5" (66 mm) speaker expectedly is not a bass monster and a little mid-centric in general (cheapo Chinese speaker), but it's OK. The bass boost function (FM only – I wouldn't mind if it were available on AM as well) allows a decently full sound at the expense of maximum volume, beating sets with a 50 mm speaker by quite a margin. Maximum volume is not breathtaking but adequate.
• The DE1102 is one of the few portable radios with a virtually noise-free headphone output. This is because it uses a combination of LM4811 (digital volume control, headphone amp) and LM4862 (speaker power amp). (Seems like this set was an attempt to bring in some fresh ideas, which also shows in the choice of these relatively new ICs. The DE1103 used a far more conventional CXA1622 based audio stage again, with a LM358 for line-out driving.) I still wouldn't connect any hyper-sensitive IEMs since the lowest volume is final and obvious distortion sets in once volume display drops from two bars to one, but something in the 115 dB SPL / 1 V_rms range should work fine. My old AKG K26P would be a touch too sensitive, but Etymotic ER6 or Sennheiser HD590 work fine, except that you'll want to engage the BASS boost and possibly slide the tone switch to NEWS to get balanced sound out the notoriously thin Etys. Hiss levels are still low with Shure SE420s, but as expected minimum volume (at two bars) is too high, so an Ultimate Ears attenuator finds use.
• Output impedance is fairly low at little more than 10 ohms, and the 220 µF coupling capacitors should result in negligible additional bass rolloff. (The dropping frequency response in the bass – -3 dB @ ~80 Hz – is thus mainly to be attributed to the two 47 nF film caps (C105/C108) which could be swapped for something a little larger, like 220 nF film, should anyone feel disturbed by this.)
• Maximum volume on the headphone out is entirely sufficient to drive my venerable Sennheiser HD420SLs adequately. Overall, a wide range of headphones should work well on this set.
• Volume steps are small enough (3 dB nominal), but I wouldn't want them to be any larger either. Unsurprisingly for a digital volume control, channel tracking is good.

FM reception

• There is no muting during manual tuning, which is a plus. Only search tuning mutes the output.
• Search tuning stops on nominal frequencies only. That's a definite plus – a number of my other sets only evaluate signal strength and stop multiple times on a single station. The PLL synthesizer IC provides an IF counter input which is used here.
• Performance is similar to the RD1220 here, as expected – good sensitivity and selectivity. The set does, however, ultimately run out of limiting at low fieldstrengths (even if only a bit), which does not happen with either of the Redsuns (also expectedly, as the LA1260 used in these shows better than average performance in this regard). In return, selectivity is a little better than on the Redsun, which was expected with 180 kHz vs. 230 kHz filters, though I had to look for a while to find a station that showed a difference.
• Some microprocessor/display related interference may be heard on weak signals when the light is on. Usually no such things are heard except maybe when the telescopic antenna is positioned close to the display.
• Like on the RD1220 and other sets with a similar frontend, overload will be encountered in strong-signal areas. The attenuator tends to help. One would need to resort to a DE1103 or RD2000/2100 for better performance in this regard.
• As typical for Toshiba ICs, stereo threshold is low.
• Stereo separation gives no reason for complaints.
• FM audio is not as bright as on the RD1220. In fact, the measured response (below) indicates that deemphasis is quite accurate for 75 µs. Good fidelity for a portable like this.
• Mobile use has yielded fairly good results so far. The DE1102 is not entirely pocket sized, of course, so tends to be a bit of a tight fit, and when in a pocket, touching the retracted telescopic antenna may be helpful to achieve optimum sensitivity (in areas with strongly varying fieldstrengths, the human body simply makes a larger antenna...). I'd say that it beats the RP2059 in both selectivity (which was to be expected) and sensitivity, with no audio hiss to be heard. Results on my dynamic range test parcours (read: way from home to university and vice versa) were good, in particular the wanted station was not swamped by intermod in a critical spot with very high signal strengths that gave previously used sets trouble – and even if it had, there'd still be the attenuator.
• Since the set does not seem to disable stereo decoding when using the speaker, I'd advise to do it manually to reduce current consumption.

Shortwave reception

• Tuning is always accompanied by momentary muting. On page 9, continuous tuning will speed up after a while, to maybe 15 steps (kHz) a second or somesuch. This slows down in the presence of strong signals. When 5 kHz steps are in action on pages 0-6, holding up/down pressed continuously starts scan tuning, which mutes until a strong signal is found. Then the search stops. When the scan reaches the end of one broadcast band, it continues on the next one in the respective direction (most radios wrap around to the other end of the same band, which one may or may not prefer).
• Sensitivity on the higher bands like 16m is similar to ICF-SW7600 (and G) when using just the whip (and lower on 10m), "some more antenna" may help here. However, towards the low shortwave regions it increases quite a bit, to the point where the DE1102 with the attenuator in LOCAL is just as sensitive as the SW7600G in DX. (If this set doesn't cut it in terms of sensitivity down there, chances are no other affordable, compact setup will do much better. A tuned loop will still do a better job than the whip if nearby interference sources are present, of course.)
  That the DE1102 has a good bit of RF gain (as expected from the schematic) is easily shown by connecting my trusty Sony AN-LP1 – then sensitivity picks up dramatically even on 16m, so much in fact that switching the attenuator to LOCAL may be a good idea to avoid overload (but more on that later); at this point it's limited by man-made noise anyway. Apparently the Miller effect is at fault here, which effectively multiplies the input capacitance of the common source RF amp circuit and thus reduces amplification at higher frequencies – big effect with a high-impedance load like the whip (even in parallel with 10 kOhms), small effect with the AN-LP1's amplifier's small output impedance (or a tuned dipole or loop). The Sony sets employ a source follower, which gives no voltage gain (impedance converter only) but does not suffer from this effect. Apparently RF gain even gets smaller than one for the DE1102 on the high shortwave ranges. That used to be a big problem for sets in the olden days. On the ICF-SW55 which has a whip preamp, Sony worked around this issue by deliberately using a small parallel resistor (1 kOhm) and employing a two-stage amplifier.
  Towards the low end of the shortwave range, like on 80m, the AN-LP1 tends to give less signal strength than the whip – expectedly so, since the Sony antenna just gives about as much signal as the ICF-SW7600G's whip on 3600 kHz, its 4 MHz range (the lowest one) appears to be peaked at slightly above 4 MHz.
• Audio response seems to be restricted at both ends, so this is not a "HiFi AM" set like the RD1220 and more like a "classic" shortwave set. That does help selectivity, but the restricted lows don't exectly benefit music.
• Selectivity: The wide filter is a relatively humble 2-element type, which shows in not even ±10 kHz selectivity being all that exciting. At least there are no issues with ultimate rejection like the DE105 had, thanks to the preceding 1st IF filtering. (Given that space for a larger filter had originally been reserved, one should check whether fitting a 4-element (e.g. LPP450H or LT455HTU) would be possible. Apparently it is, since pre-modified sets are being offered.) The narrow filter, apparently a 6-element type with nominally 3 or 4 kHz of bandwidth, is quite a different story. When comparing the DE1102 in narrow to my ICF-SW7600G, the two sets gave virtually identical selectivity, except that at 10 kHz distance from a very strong station the DE1102 showed a bit of 10 kHz het, maybe ultimate rejection is not as good (unlike in the Sony, the preceding IFT is not tuned). Still, not too shabby. Expect pretty decent 5 kHz and mostly bulletproof 10 kHz separation.
• The attenuator is not too drastic, increasing the chances of it being actually used. (The difference between the two settings should be around 14..15 dB in theory.)
• AGC range is not quite as good as on more upscale sets, e.g. ICF-SW7600. Weak stations drop noticeably more in level, and average audio output is slightly below FM level (as evident in the specified performance curves for the TA8132, output level keeps rising slightly when input is above the AGC "knee", about 4 dB per 70 dB input level change). AGC loop gain thus isn't extremely high.
• IF-level hiss: Weaker signals show a bit of broadband IF-level hiss, indicating that the post-filter IF amp has fairly high maximum gain (and/or is not that low-noise) as-is. This is not uncommon when all the gain has to be squeezed into few amplification stages, and it becomes apparent when (a) the filter has a relatively small bandwidth and steep skirts (like the narrow one here) and (b) audio filtering is not too drastic.
• Image rejection (1): 2nd IF images (from 900 kHz higher) are suppressed well, noticeably better so than on e.g. the ICF-SW7600G (which admittedly is not exactly outstanding in this regard, just like a number of other portables with a high 1st IF). When the Sony will already receive an image loud and clear, there is nary a trace of the carrier in general noise on the DE1102 in SSB. Excellent! (This was to be expected given the lower 1st IF, less leakage around the filter and all, but still is nice to see.)
• Image rejection (2): 1st IF images (from 21.4 MHz higher) are pretty well suppressed as long as the reception frequency is below approx. 10 MHz; not a whole lot of suppression is noticeable above this point. This is because just like in the DE1101, additional lowpass filtering kicks in for the lower ranges. This is disabled for higher frequencies to avoid degrading sensitivity.
• Strong signal handling (1): While I have not managed to provoke 2nd order intermod, indicating good 1st mixer symmetry (the RP2000 had its share of problems there), things look somewhat less good if 3rd order intermod is concerned. An intentionally mistuned AN-LP1 (set to 6 or 7 MHz in the evening to simulate a "big antenna") lured out plenty of intermod around 20 MHz – as well as on 49m and 41m themselves – that could not be eliminated completely even when using the attenuator (which affects the signal after the RF preamp, so if this were in trouble it wouldn't help). The ICF-SW7600G used for comparison endured this test stoically and barely showed weak carriers only detectable in SSB, if that. Now to set the record straight, I was unable to provoke overload on the whip with some wire attached, at least when running on batteries (using the mains adapter made some mixing products appear on 80m, an antenna tuner helped). At comparable sensitivity levels (on the high test frequencies), the RP2000 already showed intermod. As expected, the RD1220 is more sensitive to in-band 3rd order intermod than the DE1102.
• Strong signal handling (2): Given strong input signals, the second mixer can get into trouble as well. Using the AN-LP1, a strong 16m signal may cross-modulate onto weaker stations up to ±25..30 kHz away, but rarely further than that, indicating that the 1st IF crystal filter is doing its job. Use the attenuator if needed. This is a limitation that one should be aware of when cruising busy bands with many strong and tightly spaced signals, like 49m through 31m in the evening here in Central Europe. If antenna signal levels are high enough (remember that RF gain is quite high), the attenuator alone may not suffice yet. A passive tuned loop might do well in these cases if you don't want to resort to the whip alone.
• SSB: The DE1102 is the least expensive receiver with SSB reception available, and one of the smallest ones at that (only Sony ICF-SW100 and ICF-SW07 are yet smaller, but those were much more expensive as well), which in itself makes it fairly attractive. In practice, this function works pretty decently although I would recommend tuning the BFO to the edges of the filter passband for better audio fidelity and opposite-sideband suppression; ±2 kHz is easily accessible. The fine tuning control, although backwards (turning upwards makes BFO freq go down), is far less fussy to operate than e.g. on the venerable Sony ICF-SW7600 (a fav of mine, albeit not exactly for SSB), not only because of a smaller tuning range but also due to the control not moving too easily.
  The frequently noted warble, yep, it's there – a clean carrier alone will change pitch depending on signal strength (indication, that is – as has been noted, it's related to the tuning LED current draw and can be eliminated by disconnecting the respective supply). This is less disturbing on SSB signals than one might think, but does tend to make the audio sound "rough". Things like that happen when the DC/DC output voltage is unstabilized yet subsequently used to generate the voltage for the varactor diode tuning the BFO via a voltage divider network (Sony uses a zener diode and capacitor for feedback to get stable DC/DC out). This actually is a double construction blunder – since the native tuning range of the BFO is far too large (LC circuit with varactor diode as the only capacitance), they decided to do fine tuning by restricting the tuning voltage range instead of just using some parallel capacitance and coupling the varactor diode loosely. It would have been smarter to use a ceramic resonator and pull that a little.
  On higher freqs, the 1st LO seems to add noticeable ticking/buzzing (15.6 MHz is still quite OK, it's noticeable on 17.9 MHz and quite noticeable on 26...28 MHz). This roughens up the audio on e.g. 10m even more. The same noises are also audible weakly in the background all the time in SSB, and they also seem to creep into the speaker audio amplifier.
• Spurs: The DE1102 uses a crystal oscillator to generate the 2nd LO frequency which is not shielded and therefore easy to receive. Both 10.25 and 20.50 MHz are present and strong, the 3rd harmonic on 30.75 MHz can be detected as a weak image on 9350 kHz. The 4th harmonic on 41 MHz comes in strongly on 19600 kHz, while there is nary a trace of the 5th (51.25 MHz --> 29850 kHz). Possibly the harmonics are mainly even-order.
  I have also found a mixing product stemming from the 2nd harmonic of the 1st LO so far (if you find utility stations shortly above 3 MHz in the middle of the day, check ((f + 10.7 MHz) * 2 - 10.7 MHz) = (2 * f - 10.7 MHz) higher).
  Last but not least, a strong utility station on 10.70 MHz will be heard loud and clear throughout the shortwave spectrum. Apparently RF input finds its way into the 1st IF, around the 1st mixer. (A dual balanced mixer shouldn't have RF-IF isolation that bad, so I guess the leakage is caused by insufficient shielding.) Fortunately there are very few stations in that range; what I heard might have been that Russian naval station in CW that I found listed for exactly 10.700 MHz.

MW reception

Daytime MW sensitivity is close to e.g. the ICF-SW30 over much of the band, however there's some display/microprocessor related ticking/buzzing to be heard on empty channels as well as weaker stations. Sensitivity noticeably drops at the upper end of the band, making the DE1102 fall slightly behind the reference set. (At least the ticking disappears as well.) The RD1220 is about the same in raw sensitivity (and ferrite rod size) but seems to have less gain.

The restricted lows in the DE1102 reduce subjective sensitivity, even more so during speaker playback – the venerable Sony ICF-7600A, a design that's more than 20 years older but features better raw sensitivity (larger ferrite rod in a generally larger set) as well as carefully tuned speaker audio, beats it hands down.

All in all, a decent but unexciting showing, with the internally generated interference degrading sensitivity somewhat.

Battery life

• A set of brand new Sanyo Eneloops, charged only once, lasted about 5 days of fairly heavy use (partly at night with lighting on). Not too bad considering how much one tends to play with new toys. It had to carry out the jobs of mobile and work radio, too.
• The battery indicator is a bit too generous, only dropping from all three to two bars when voltage per cell is a little above 1.250 V, and to one bar at about 1.20 V per cell. At this point it should already show empty, as NiMH rechargeables barely have any charge left then. Still, the indicator range is quite usable for regular NiMHs like 2500 mAh Sanyos, but definitely too optimistic for Eneloops (with typically about 40 mV more) – I ran the set for a while longer at two bars, and the following recharging showed that they only had about 10% of their nominal capacity left. A set of the conventional 2500 mAh cells turned out to be almost half full still (around 1000 mAh left) when I eventually removed them at little more than 1.240 V per cell, wondering about the short run time.
• Multiple sources (Chinese page, message 1881 in the KAITO-DEGEN1102 Yahoo! group) have stated a current draw of about 40..50 mA sans backlight (which takes another ~10 mA), which is quite economic for a set of this type.
• I logged the run times with two different sets of 2500 mAh Sanyo cells during normal use (mainly FM with headphones), and mostly achieved run times of between 30 and 35 hours (a little less during infrequently use due to the cells self-discharging), with a maximum of a little over 35 hours. This equates to about an effective 70..75 mA. Apparently FM operation is more hungry.
• For charge currents, see message 1899 in the KAITO-DEGEN1102 Yahoo! group.

Misc.

Cellphone interference has not been a problem so far. It apparently takes pretty high levels for it to be audible at all, which so far hasn't happened more than once or so. The DE1102 is much less critical than e.g. my vintage Toshiba pocket radios in this regard.

Audio noise spectra

Now for the usual audio spectra:

• Hey, that FM deemphasis looks pretty good. It's closer to 75µs than to 50µs though. (Good for folks in North America, but over here one may want to swap C31 and C32 from 15 nF to 10 nF) No pilot tone filtering, as so often, but hey, I'm not a tape recorder and don't hear up to 19 kHz anyway. The slight bass rolloff (-3 dB @ ~80 Hz) was expected from the schematic, one would need to swap the two 47 nF film caps (C105/C108) for something a little larger, like 220 nF.
• The BASS function gives a fairly wideband boost to the bass and lower mids, mostly for <=500 Hz but with noticeable effect up to 1 kHz. The NEWS tone switch appears to provide a shelving filter for the highs, given that the response further up remains first-order (maybe not too smart an arrangement for this type of tone switch).
• OK, so maybe narrow selectivity isn't quite up to a '7600G – but certainly somewhere between that and the SW7600. The low cut in AM is due to capacitor C37 at the CD8132, a 22 nF, which can be swapped for 100 nF if you don't like that.
• Some audio filtering keeps the wide filter from being useless, but its performance is far from outstanding.
• SSB audio is broadly reminiscent of the ICF-SW7600G, but the comparison would not be entirely fair as the latter's response was taken without off-tuning.

Conclusion

Overall, the DE1102 is a compact, fairly affordable portable that, while far from flawless, shows pretty respectable performance for this type of set. It does best on FM and the lower shortwave ranges. I wouldn't mind if some of the bugs (mainly shielding-related stuff and SSB warble) were worked out, but even as-is, it has much to recommend it. And it's cute. What more do you want? :p

Due to the qualities of the headphone out and FM section, the DE1102 has replaced the Toshiba RP2059 as mobile FM radio together with Shure SE420s (and UE attenuator).Yes, it might be a touch oversized for the kind of application, but do you know any pocket-sized FM radios with good selectivity, sensitivity and dynamic range plus a hiss-free headphone out?

Notes

After not even one year, my DE1102 developed a defect: It became fairly deaf in the narrow AM bandwidth. Interestingly enough, the problem disappeared again soon after and has yet to resurface after more than half a year. (It did eventually when taking out the batteries, but putting them back in cured this again.) Knowing the qualities of the cheapest kind of lead-free solder (namely, being prone to developing bad solder joints), I suspect a bad solder joint. I wanted to have the set modified at some point anyway, but still something like this doesn't exactly speak for quality.

↑ Back to table of contents

Grundig Concert-Boy 1100

The Concert-Boy 1100 is a large domestic portable (about 40 x 22 x 9 cm in size and more than 3 kg in weight, with a detachable carring handle) dating to the mid-1970s; it saw a 5-year production run from 1973 onwards and apparently was one of the first sets for the German domestic market made in Grundig's factory in Braga, Portugal which had opened in 1965 (this one still exists to this day, now manufacturing aftermarket and OEM car radios as Delphi Grundig). In its day, it was above average but by no means a top-flight portable like a Satellit. In those days, the market for more upscale portables was already shrinking, and Grundig had gone public, so first signs of cost-cutting become apparent. In return, development focused on service friendliness – this set, for example, can be opened after removing only two screws at the bottom. Ironically, Grundigs of this period usually hold up well.

Band coverage includes longwave 145 to 280 kHz, mediumwave 510 to 1620 kHz, spread 49m band 5.9 to 6.23 MHz, rest of shortwave 6.1 to 18.5 MHz and FM 87.5 to 108 MHz. It is single conversion throughout, with IFs of 10.7 MHz on FM and 460 kHz for the AM ranges. A total of 10 tuned circuits for FM (2x frontend, 6x IF filter, 2x ratio det) and 7 tuned circuits for AM (2x frontend, 5x IF filter) find use, plus there's a 5 kHz audio notch filter to improve effective AM selectivity on shortwave in particular, and AFC is employed on FM. There are 3 FM and 2 AM IF gain stages. Semiconductors include 13 transistors and 10 diodes. There is a 0..100 logging scale on the inside of the dial window (an awkward place given the amount of parallax distortion, which is far greater than for the dual-bar pointer on the main dial).

This set can be powered by either 6 D cells or mains, where it isn't picky and will run from 110..240 V without adjustment. Like any better portable of its day, it provides a tape I/O via a 5-pin DIN jack. A symmetric 240/300 ohm FM antenna connector is also present. Tone adjustment is possible with separate bass and treble controls. The speaker is a large 7" by 5" (approx. 18 by 13 cm) wideband type, rated output power is 2W sine.

The telescopic antenna measures about 78 cm when extended. It can be fixed in six different angles horizontally (i.e. 60° each) and three vertically (i.e. 45° each) and disappears inside the set when entirely collapsed.

Available finishes were black and wood imitation, my set is of the black variety.

My sample had obviously been through some hard times, as it arrived miserably dirty inside and out and with signs of heavy use, like a junkyard special. Extensive cleaning was required, including a polish of the dial window. Functionally it did better, with only the highs control being non-functional (which was later found to be caused by the pot's wiper having come off and floating around inside, no idea how this happened or how to reattach) and the switch array causing some intermittence (tuner cleaner improved things). None of the bands were way off, only the very widely spread 49m band had an immediately visible deviation (DW on 6075 made an appearance at about 6.11 MHz, i.e. 35 kHz high – still peanuts, absolutely speaking). This speaks for good alignment stability.

• Audio quality:
  • This set has the classic full Grundig sound. I tend to keep the bass control near the bottom end of its range... Other than that, nicely balanced sound with extended frequency response, certainly more refined than e.g. the Redsun RP2000 – a good speaker isn't cheap, and those from Grundig's own production do pretty well indeed.
  • Both amplifier and speaker driver have their limits, of course, so it won't get arbitrarily loud with the bass turned fully up. Still, a respectable output volume can be reached. This is a great set for rocking out. :)
  • In addition to the tone controls, there obviously is a volume-dependent loudness which is always present. To my ears, the effect varies too strongly with volume. Perhaps this was intentional to enable better performance at high output levels.
  • If you notice a ground-loop type hum, turn around the mains plug. I'm not sure why that is, the mains transformer should normally take care of galvanic insulation.
  • Maybe I'll experiment with some damping of the speaker to tame the boominess in the bass.
• FM reception:
  • Sensitivity falls somewhat behind modern-day sets and is more in line with e.g. the Panasonic RF-1410L. It's still OK though. (Do note that judging FM sensitivity accurately here is not easy, the band is too crowded.)
  • Selectivity is a little inconsistent, sometimes 2x 280 kHz ceramic level (next to strong locals), sometimes noticeably better. I guess bandwidth is not too large but form factor falls behind modern-day filters. Still, not too bad for a set of this class and vintage.
  • AFC is always on. Its pulling range has thankfully been limited to a few 100 kHz, but at times it did get into the way of my DXing tests. This set was not built for FM DX, but rather for reliable reception of at least moderately strong stations.
• Shortwave reception (KW2):
  • Tuning, on the upper ranges in particular, is best performed by a safecracker. The tuning knob is not small, but you get from 31m to the upper end in only 3 turns, and things get more compressed towards the top. The tuning dial, spanning 6.1 to 18.5 MHz, is good for little more than band indication, although the specified frequency points coincide well with actual frequency (only the 12 MHz mark is off, I guess it was supposed to read 11 MHz since the 25m band marking is above it).
  • Once you have tuned in properly, frequency stability is remarkable for a set of this type. In comparison, the RF-1410L wobbles around a lot more and also has some long-term drift. With a drift of only a few 100 Hz at most (and this with as large a frequency range as this covered), you won't need to worry about periodic retuning – provided you refrain from selecting another band (some switch cleaning and/or switch wiggling helps here, however) or knocking the tuning knob. For obvious reasons, tuning is somewhat sensitive to mechanical shock. Touching the antenna only moves tuning by few 100 Hz, so this is uncritical.
  • Sensitivity is good. I was able to tweak the alignment a bit and achieve a reasonable compromise between 19m and 16m sensitivity – precise tracking is not easy to achieve with such a wide tuning range, here sets with bandspread ranges hold an advantage. There is a bit of influence of RF tuning on LO tuning, but it's not dramatic and doesn't really hamper peaking up. Ultimately 19m sensitivity on mains compares well with the RD1220 on batteries, which is not too shabby considering that the set most likely does not have an RF stage (i.e. RF preamp), like many bipolar-only transistor sets. These simply make use of the fact that one could now build low-noise mixers, which had been a problem with tubes, and add more gain after the mixer instead. That these low-noise mixers are additive and do not have exceptional dynamic range along with modest RF–IF isolation is another matter.
  • AGC action is middle-of-the-road level as one would expect from an oldschool portable like this, i.e. a noticeably non-constant output level even above AGC knee level, but I'm under the impression that it does beat the RF-1410L at least.
  • In terms of selectivity, the Concert Boy 1100 fits in between sets with 2-element ceramics + IFT and ones with wide 6-element ceramics, as to be expected. 10 kHz selectivity, while decent, cannot quite keep up with the 6-element ceramic filter in the Redsun RD1220. However, the audio notch filtering improves effective selectivity nearby and gets rid of 5 kHz hets pretty well, so in the end the set gives a better impression than the RF-1410L.
  • Overall, this is a set for listening to the stronger shortwave stations in good audio quality.
  • As expected in a single conversion set, images crop up. There is noticeable suppression even on the higher ranges, so frontend Q appears to be quite good, similar to what I saw in the VEF 206. (And like in the Latvian set, an air tuning cap finds use, which are reputed to have good Q. It's a 2+2 gang cap with gear reduction in this case.)
  • When signal levels are high, overload in the form of whistles may become audible. The frontend does, however, do a pretty good job in keeping those to a minimum, seemingly aided by RF AGC.
  • Higher LO harmonics are the most probable cause for FM broadcast stations making an appearance on this band.
• Shortwave reception (KW1 – 49m spread):
  Much the same as above, except that tuning is very easy, since a range of less than 350 kHz is spread over the whole dial. Readout is possible to about 10 kHz, though a realignment would be useful with my sample in order to get the scale closer to reality again (with such a widely spread band, it doesn't take much aging-related drift to shift indicated frequencies noticeably). I've now tweaked the LO trimmer cap a bit and the readout is quite accurate again.
• MW/LW reception:
  Quite respectable overall. Sensitivity is up to the better portables in my collection (not quite ICF-7600A level on MW but not too far off, LW is comparable to ICF-SW7600 and behind ITC Julia, not too bad for not having an RF preamp), nulling with the 15 cm ferrite rod works well, and selectivity is good enough for a mostly het-free experience (although you do notice the difference to modern-day sets). I would have to check whether the occasional whistle in the evening is due to overload or LO harmonics catching shortwave stations. Attempting to provoke close-by 3rd-order intermod by placing the set inside the big AM loop (off-tuned) proved pretty much unsuccessful, the set did about as well as ICF-7600A and ICF-SW7600 here – no RF stage plus RF AGC plus relatively high supply voltage have to show somewhere.

Overall, this is a classic domestic portable that excels at good, stable reception of somewhat less critical stations with great sound. As the saying goes, they don't make 'em like they used to...

There are a few problems that I've been wrestling with:

• The chassis contains two circuit boards at a 90° angle interconnected by solder joints. Now one of the boards has the DIN jack soldered on, i.e. may see mechanical stress during plugging and unplugging. While the jack itself has a number of solder joints and is less likely to develop problems, any up and down movements of the PCB would stress the interconnection solder joints. And indeed, that PCB wants to be pulled upwards a bit for input signal ground to make contact.
  Fixing this is a little more involved, but thankfully the set is fairly service-friendly: You can remove the whole chassis easily after undoing four screws, two more screws later you can pull off the dial mechanics (preferably tune to one end of the range) – getting things back on is the more difficult part – and there you are, readily accessible solder joints. Amazingly enough, I even managed to resolder the ground connection in question and reassemble the set without wrecking anything else (if there is someone with a steady hand, it's not me).
• While trying to find the cause for the audio hum, I managed to kill the input protection diodes, which turned out to be dead short (but since the rest is still alive, they've done their job). Clipping the diodes restored sensitivity, but the new 1N4148s should be installed on the next occasion.

↑ Back to table of contents

Eton / Lextronix E100

(Sample obtained used, mid-2008, originally bought new Mar 2008. S/N: E10-0709032118)

Intro

The E100 is a small PLL synthesized shortwave portable that covers MW (520 .. 1710 kHz in 10 kHz steps or 522 .. 1620 kHz in 9 khz steps, fine tuning in 1 kHz steps), SW (1711 .. 29999 kHz in 1 or 5 kHz steps) and FM stereo (either 87 .. 108 or 76 .. 108 MHz in 100/50 kHz steps); FM coverage and MW steps can be selected by the user. It measures about 125 x 76 x 31 mm³, which means that it is only a touch larger than the Philips AE3405, noticeably larger than the DE105 and a good bit smaller still than the (already cute) DE1102. Weight is specified as 198 grams (I haven't dragged out the kitchen scale to confirm this, but would assume it's without batteries, which typically add about 25..30 grams each). Power supply is via two AA cells, either primary cells or rechargeables are accepted (with the user being able to select the type for correct battery meter display).

The set has a numeric keypad for direct frequency entry, keypad or rotary tuning, a 3-step input attenuator, a high/low tone switch, a locking slider and selectable mono or stereo operation on FM. A total of 200 (non-volatile) presets are provided that are spread over either 4, 5, 8 or 20 pages depending on setup, and there are clock and alarm functions as well. Connectors include DC and headphone, that's it – there was no space left for an antenna jack, unfortunately.

Accessories supplied include a user manual (in a number of languages but not incredibly detailed, and no specs as usual for the company, there only are a few on the packaging), list of European distributors, carrying pouch and apparently some earduds, err, -buds as well, which I did not receive with my set which was (very lightly) used. (It's not like I'd miss them.)

Typical new prices here in Germany have traditionally been around 80 EUR (plus shipping), but more recently some stores dropped this to about 60 EUR, less than you'd pay for a (more full-featured) DE1102. That seems like fairly reasonable pricing now. (Update May 2011: All sold out.) In the US, you can expect to pay more than for a KA1102 and about as much as for a G6 Aviator if you buy the "Grundig" version G100. Canadians are more lucky than that, they can obtain an E100 (the "old" model, not the one I review here, more on that later) for little more than $50CAD.

Versions

As we all know, the E100 is (/was) made by Chinese manufacturer Tecsun. It originated as a slightly modified version of Tecsun's PL200 with a curved front and rubberized surface (which feels nice but has been reported to be subject to wear). This was a single conversion set throughout with an AM IF of 455 kHz.
While there had been some internal changes to past models, like the change of the ferrite rod antenna to an adjustable type that gave better MW performance (in parallel to the PL350, it seems), "ultralight" MW DXers (see dxer.ca, "ultralightdx" Yahoo! group) were shocked to find that new samples of the E100 (serial numbers 0709xxxx) are entirely different inside. It is now a dual conversion design with a high 1st IF of 55.845 MHz and a 2nd IF of 455 kHz (confirmed by the set of spurs encountered). This also means that the MW frontend now is a broadband type, rather a step back for MW DXers as this generally is more prone to overload. It is to be assumed that the current version is more related to the PL450, which came out in early 2008. My sample was originally bought in March 2008, so the revamp must date back to early 2008 at least. A little creative interpretation of the serial number suggests a September 2007 build date. As this model was discontinued in mid-2008, this may actually have been the last batch.

Look and feel

• This is quite a small radio. Next to it, the RD1220 looks outright huge (and fairly low-tech, too).
• About the second thing you notice is the rubberized coating on the case's surface. It gives a nice feel but has been reported to be subject to wear. The buttons and fine tuning knob are made of regular uncoated plastic.
• In spite of the case's plastic being quite soft (which you notice upon disassembly), it feels solid and sturdy. Everything fits together well, and there are no big gaps or anything.
• There are no dedicated retaining features for the fold-out stand or telescopic antenna. The stand just snaps back, and if you push the antenna down it will stay on its rest, but very high-tech this is not.
• The battery compartment door is not hinged.
• Either upright or tilted operation is possible. There are no provisions for operating the set flat on its back. With no rubber feet, it slides around on a table more easily than I like.
• The telescopic antenna measures about 54 cm in length, which reflects the set's small size.
• The display has good contrast and viewing angles. The numerals for the main frequency (or time when off) display are about as large as the ICF-SW7600G's, which is a good size for such a small set. The single orange LED for backlight on the left-hand side is quite dim (in return I wouldn't expect it to drain the batteries like crazy) but gets the job done at night, so I'm not complaining too much.
• The carry pouch is about as comfortable as a radio could want, it's easily the best-padded and softest such accessory among my radios. The set slides in easily. It is not, however, the most rugged accessory and should be treated with care. (I'd also expect the foam padding inside to disintegrate after some 10 years or so.)

Operation

• I would call this a "no surprises" radio: The attenuator (ANT GAIN switch) is an attenuator (prior to RF amp stages) and nothing else, entry timeouts are sufficiently relaxed, and light comes on briefly after a short push of the light button (or after powering on the set unlit) and will stay on permanently if you hold it for a whole longer. The power button behaves in a similar way, the set comes on in sleep timer mode if you push it briefly, a longer push (upon which the display will show "PL200" instead of the sleep timer setting) enables unlimited operation. Overall, while you will need to take a look at the manual initially, this will become much less frequent soon.
• The typical Tecsun system of "system set codes" is neat. Many of the codes are quite straightforward and easy to remember, though you should keep the manual handy for the others. (Unlike stated in the user manual, system set code 22 will only show all the display segments but not the model number. Apparently the new E100 is different from older ones in that regard.) If you can't remember the current settings, hold SYS. SET for some time in order to have them displayed. All of them will be kept even with no batteries in.
  Undocumented system set codes: 98 shows "PL200", 99 always says "On".
• Setting the time can be done by both direct entry (confirmed by ENTER) and increasing hours and minutes. I prefer the former.
• The display is quite informative and presents the time, alarm time and battery indication in off state, while frequency, time and signal strength are displayed during reception. During meter band selection (which can also be continued with UP and DOWN after pressing SW METER BAND), time display is replaced by meter band indication.
• While you can have the display lighting on indefinitely both during operation and when off, turning the set off will activate momentary lighting mode again, which means that the light will also turn off soon after. Presumably this was done in order to avoid accidental battery drain – even when already half asleep, you can just turn off the set without having to worry about the light. The E100 is pretty well suited to nighttime listening overall.
  While we're at lighting, the locking function does not affect the LIGHT button, so that you still have the chance of seeing the lock symbol in the display. However, while accidental activation of the light seems unlikely as long as the set is in the carrying pouch, this might result in unwanted battery drain during transit.
• If there is some criticism, it would be related to tuning. Up and down tuning with the narrow buttons isn't an absolute joy, and the "fine tuning" wheel with a rotary encoder at the side is indeed reserved for the smaller tuning steps (1 kHz on the AM ranges), with some short muting at that – so no band cruising. Pushing the tuning buttons for a while longer will start an automatic search (about the slowest one I've seen, at about 13 kHz or a little more than two channels per second on shortwave). This does not wrap around to the other end of the same band but will continue in the next broadcast band in the respective direction. If you hold UP or DOWN continuously, quick tuning with a decent speed of about 50 kHz per second is activated, which again only covers the pre-defined broadcast bands. Audio is muted during search and quick tuning. To get out of the broadcast bands, slow manual tuning, fine tuning or direct entry are required.
• The mechanical encoder for the "fine tuning" wheel covers about 20 steps per revolution – provided it doesn't decide to step backwards, which happens more often than I like. I'm not exactly a fan of cheapo mechanical encoders, and that's why. (Even better-quality ones are no infrequent trouble spots at a certain age.)
• Direct frequency entry, while not as enjoyable as it could be due to the small buttons, is handled quite cleverly software-wise, much like in the DE1102. If the entered number is a valid frequency in the respective range (accurate to 1 kHz on the AM ranges and 100 kHz on FM), it is taken over even with no second press of ENTER. Otherwise you can also enter the first few digits. 2 in the shortwave range takes you to 20 MHz rather than the 2 MHz I would have expected, but I guess that's a matter of taste. (In the DE1102 which only overs from 3 MHz upwards, there is no other choice than 20 MHz.) But in any case, you can enter just about any number and it will be accepted. (This is not true for e.g. my trusty old ICF-SW7600G which remains on the old frequency given inputs it does not understand, and the ICF-SW7600 even flashes "TRY AGAIN" in such cases.)
• Memory operation is a mixed bag. Unlike stated in the manual, a single push of MEM & EDIT is sufficient to save a tuned station to the next free slot. (This invites filling up the memory with junk presets unintentionally.) In addition, regardless of page configuration, recalling a single-digit memory preset requires entering a leading zero. Maybe this is why they included the memory scanning mode (accessed via M. SCAN / PAGE) which allows zapping through existing presets. Moving presets is one of those things that you'd keep the manual around for. At least you can change the number of pages without any presets getting lost – pretty foolproof.

FM reception

Good above average portable level reception, but not that spectacular either. Tecsuns usually have to make do with only one ceramic IF filter, and that shows in less steep filter skirts when compared to e.g. DE1102 and RD1220. It still manages to pull some stations that give sets with 280 kHz barn doors trouble, and stereo separation seems good, but an FM DX king it is not, and audio distortion could be a touch lower as well (seems like the filter peaks about 50 kHz high, or maybe the tuned IFT is not aligned correctly). I've seen more sensitive sets, but in return it overloads less easily than e.g. the DE1102.

Shortwave reception

• Selectivity with the single bandwidth provided is good. I would estimate the filter to be a 6-element job with an official 4 kHz bandwidth, or maybe even a 3 kHz one. Audio is less muffled than in the DE1102 in narrow, but raw selectivity seems about equal for the most part, with only somewhat stronger 5 kHz hets. The overall result is about ICF-SW7600 level (or not too far off). The filter appears to be about 500 Hz high on my sample, so reception is about equal on the nominal frequency and 1 kHz higher. ("Old" E100s have regularly been reported to be a full kHz off, and their stock filter is generally considered less good.)
• Sensitivity can keep up quite well with the DE1102 over large parts of the spectrum. It doesn't have the same very high sensitivity on the lowest bands, but still is plenty sensitive there, a little more so than the Sony ICF-SW7600 and certainly more so than the ICF-SW7600G. (This evens out by the time you get to 19m or thereabouts, the ICF-SW7600 definitely pulls ahead on 16m. Seems that like the DE1102, the E100 uses a FET in common source configuration as a preamp. The Sonys, like several other sets, employ a source follower only, which has less gain but also less effective input capacitance and thus more constant sensitivity.) Over much of the lower shortwave ranges (e.g. 60m through 31m), clipping some wire to the antenna has little effect, so the whip must be well-matched there.
  In a second row of comparisons, this time with the by-then modified ICF-SW7600G sporting a properly-matched whip antenna (see Telescopic antenna matching mod), the E100 turned out to be be consistently on par to a hair ahead of the Sony throughout the shortwave ranges. So maybe it uses a source follower preamp after all.
• Since there is no external antenna jack, antenna upgrade possibilities obviously are limited. The set seemed to be able to take a Sony AN-LP1, but you may have to get creative when establishing a ground connection on the set, especially if you want to use headphones at the same time.
• AGC action at low signal strengths is good. The E100 certainly does a better job than the DE1102, which runs out of gain and drops volume too early. (However, there's a downside to that – read on.)
• 2nd IF image rejection, as tested on 60m just using the whip, is about equal to the DE1102 and leaves the Sony ICF-SW7600G in the dust (image rejection is the big weakness of the Sony, which pretty much has to be used with the AN-LP1 or a similar frequency-selective antenna with good signal levels for 60 to be fun). Interestingly the E100 pulls this off with a much higher 1st IF, which either requires more filtering, better board layout and/or shielding or a 2nd mixer of the image rejection type – it's the second option here, I guess. Neither of the Chinese sets were able to suppress images on 60m entirely when 49m was really booming in very early in the morning, but what did make it was pretty weak – the Sony showed images stronger than the strongest real station, RHC on 5025!
• As RF gain is on the lower bands isn't too shabby, expect to see some more resulting 2nd-order intermod on the higher bands when compared to sets like e.g. the ICF-SW7600(G).
• Nearby 3rd-order intermod behavior is similar to the DE1102. Since the AM detector also shows the same "grainy" audio when the signal fades away, it seems very likely that this set also uses the TA8132AN / CD8132GP IC as 2nd mixer, IF amplifier and detector. (I'm not overly fond of the AM detector in that chip.) If you do encounter intermod, the middle position of the attenuator switch is quite welcome.
• Very strong stations (the kind that is still full strength and noise-free even with the attenuator fully engaged and may also generate audible images) can overload the IF stages, which results in audio becoming mushy and distorted, ultimately dropping in level. Usage of the attenuator generally helps. Seems like the AGC range itself is not too different from the DE1102 (quite expectedly) but shifted to lower signal strengths. A quick comparison to the Redsun RP2000 on MW using my "AM loupe" showed that the E100 overloads earlier – this probably has more to do with different AM IF internals than with the difference in supply voltage (the Redsun uses an LA1260 which is run off a regulated voltage of about +3V).
• There are a number of internally generated birdies of non-negligible strength to be noted, with some reacting to the attenuator and others not doing that. I would attribute the ones on approx. 2790 kHz and harmonics (apparently caught via the high-impedance antenna circuit and originating from the lower right corner of the set, as you notice when putting your hand there) to an insufficiently shielded / choked DC/DC converter operating at this very frequency.
  Then there are some clean carriers (spurs) to be noted on 13278, 13506, 17856, 18160, 27012, 27126, 27353, and 27468 kHz (plus some more that are not as strong), a standard set for a 55.845 MHz / 455 kHz configuration. (The details.) These seem to be mixing products involving both 1st and 2nd LO which are generated in the 2nd mixer (the 1st LO should not be present there but frequently enough is in consumer-grade receivers); they are not impressed by the attenuator at all and must be coming in after the RF preamp. These are not suppressed as well as in the RP2000 and much less well than in ICF-SW7600 and ICF-SW7600G, which is inconsistent with the set's good 2nd IF image rejection.
  Fortunately most of the internally generated signals are outside or near the edges of the regular broadcast bands (including, usually, the images of DC/DC noise 910 kHz lower which are pretty well suppressed anyway), so in the end you don't stumble across slowly wandering whistles or lonely carriers too often, just don't be totally surprised if you do.
  Here's a little spreadsheet (OpenDocument format) that you can use to check whether a particular band is affected by one of these birdies. It allows entering the DC/DC operating frequency and band edges and then highlights the critical birdies by means of conditional formatting, nothing too fancy but gets the job done.
• One oddity: At about 25..27 MHz, background noise rises with the attenuator fully engaged (i.e. ANT GAIN in L position).
• Speaking of oddities: How a strong RNW on 5955 makes an appearance on ca. 23994 kHz (weak, but undistorted) is anyone's guess.

MW reception

• The attenuator has no effect on this band. Unlike stated in the manual, it is effective on SW and FM only. (Apparently the control always worked this way, as an old PWBR review also mentions this erratum.)
• Daytime sensitivity showed to be only a little behind ICF-SW7600 and ICF-SW7600G and about on par with, say, the ICF-SW30, with the brighter audio and less good AM detector not working to the E100's advantage. As on shortwave, there seems to be plenty of gain on this band. Overall, not too shabby for a set that has to make do with a relatively small ferrite rod, and certainly better compared to the DE1102 with its internally generated interference and high AGC threshold. (Makes me wonder how good a properly-aligned "old" sample might be! I would expect the "new" one to be more prone to overload in areas with strong MW signals, as even a better first mixer can't quite make up for a tuned frontend with its fairly narrow bandwidth.)
• Selectivity is the same as on shortwave.
• As on shortwave, very strong stations may lead to IF-level overload. Unfortunately, the attenuator not working here means that there is not a lot one can do against it. And indeed, in a little MW shootout involving a fairly strong station supported by my "AM loupe" and a number of receivers of mine it definitely brought up the rear, with both the AGC range being exhausted (audio output decreased as signal strength increased, the IF was well into limiting there, even the RP2000 only showed somewhat increased distortion at full gain) and cross-modulation appearing on the next channel (something also affecting DE1102 and ICF-SW7600G but interestingly not ICF-SW7600, which even bettered some sets with frontend tuning).
• In terms of 2nd-order intermod, it did at least as well as the other contestants with a wideband frontend when subjected to high RF levels at about half the received frequency using the "AM loupe"; the RP2000 proved somewhat better (low-Q RF tracking but 2nd-order sensitive RF preamp), and the single conversion sets with high-Q frontend tracking (ICF-7601, ICF-7600A, ICF-SW30) remained unimpressed and showed no intermod (later it turned out that the ICF-7600A did best overall).
• Audio doesn't quite have the "hi-fi quality" (both bass and highs) of the ICF-SW7600, but still is more pleasant on music than the ICF-SW7600G's "communications-quality" audio and is more extended to both extremes than the DE1102's in narrow. The frequency response in MUSIC is reminiscent of the ICF-SW7600's with the latter's TONE switch set to NEWS.
• Switching from 9 to 10 kHz steps in order to access the 1621 .. 1710 kHz range (where one not uncommonly finds pirates from Holland or Southeast Europe here) is quickly accomplished and painless; only the last tuned MW frequency is reset.

Audio quality

• The speaker is a 50 mm (2") job, so don't expect a bass monster – results unsurprisingly are noticeably more pleasant on the much larger RD1220. That being said, it works just about as well as equally-sized drivers in other sets I have. (Like I said, it's a "no surprises" radio.) You may want to switch the tone switch to LOW. Alternatively, just use some headphones.
• The audio amplifier appears to be a pretty conventional CXA1622 job, with the corresponding medium hiss levels. (I do wish there were more radios with a hiss-free headphone out. Using one amplifier for both speaker and headphones was OK in the early '90s – that's when the CXA1622 appeared, in fact –, but these days one should be able to make ICs that integrate a separate headphone amp section.)
• Channel tracking on the volume pot of my sample does not appear to be all that exciting.
• FM frequency response sounds about right for 50 µs deemphasis.
• As indicated by noise spectra taken, the tone control is of the shelving type, i.e. highs attenuation in LOW is fairly constant over a large frequency range. It has maximum effect about in the 2..10 kHz range, some 6 dB. This kind of characteristic, commonly used in audio amplifiers, is not first choice if you want to use the tone control to get rid of adjacent-channel interference; a classic lowpass with -3 dB at maybe 1.5..2 kHz would work better in this case. (It turns out that this shelving behavior is what you get when merely adding capacitance in parallel to an existing lowpass filter cap. According to simulations, the tone control circuit in the ICF-SW7600G does much the same.)

Battery life

This set appears to be quite economic. After more than three months of irregular use (frequently with display backlighting on) on the same set of Eneloops, discharging still yielded almost one third the nominal capacity.

Misc.

Actual tuned frequency in the AM ranges appears to be about half a kHz high on my sample. I was under the impression that this varied a bit, so suspected that some tuning resolution was generated by offsetting the 2nd LO if needed (see ICF-SW7600). However, some LO chasing with the DE1102 revealed this not to be the case – the 1st LO changes frequency on every 1 kHz tuning step on shortwave.
Mediumwave, however, seems to be different – here the 1 kHz steps are obviously not all created equal, and I noticed larger offsets of effective tuned frequency (up to 1.5 kHz) on my set. More specifically, stations on channels with even frequencies (e.g. the classic BBC WS frequency of 648 kHz, R.I.P.) come in best when tuned 1..2 kHz higher, as opposed to 0..1 kHz for odd frequencies. That does smell like the 2nd LO tuning trick at work here with a 1st LO resolution of 2 kHz (meaning that the 1st IF is odd like 55.845 MHz – odd IF + odd tuned frequency = even LO frequency), with imperfect alignment accounting for the intermediate steps not being right in the middle. Still, why would this be required on MW if it isn't on SW? Unfortunately I can't confirm it directly, not having a receiver for approx. 56 MHz.

Conclusion

Overall, the current E100 is a well-performing little set that would serve the world traveller well for catching broadcasts on the AM ranges. (One would only wish that it tuned longwave as well, which now should be possible, though it's unclear how well that would work given the track record of the ferrite rods used on these Chinese sets.) For something that doesn't receive all that much attention, it does a pretty respectable job. While not flawless, it is both smaller and less expensive (at a typical 79€, or even 59€ nowadays) than e.g. the venerable Sangean ATS-606A, though it does lack an external antenna input (and I would guess that the Sangean's 2nd mixer shows better nearby IMD performance).

Its closest current competitor is probably the Grundig G6 "Aviator" from the same stable (though presumably Degen OEM'd), which at much the same dimensions features knob tuning, SSB reception and even air band reception. (G6 inside pics. Yep, it says "ETON E6" on the 'board!) From what one reads, this one behaves similarly to the DE1102, and the (typical Degen) somewhat more complex operation and (typical Eton) not so detailed manual don't make such a good combination. The E100 may be much less flashy, but it certainly is more predictable. Plus it's easily available here in ol' Europe, which still isn't the case for the G6 (the rights for the Grundig brand only apply for North America).

In the meantime Tecsun has also released a model PL-210, which presumably uses the same or very similar receiving electronics as the current E100 but updates the user interface side of things, including more memory presets than you'll ever need and longwave reception. This is nice to see, as it should benefit from the improved operation of their other modern-day models (which generally receive praise in the user interface department). So far you can only buy a PL-210 directly from China.

Notes

In North America, there also is a "Grundig" variant of the E100, the G100, which is identical save for being black. Universal Radio only sell this one now.

It be noted that QC has been reported to be a little hit and miss.

S/N: E10-0709032118

I would speculate that the first four digits of the serial number (0709) stand for either internal hardware revision, batch number or production date (which would mean September 2007 in this case), while the digits that follow are the actual (ongoing) serial number. A little over 32000 samples isn't too much for a radio like that. Serial number ranges that have been spotted are 0505, 0609, 0704 and 0709, with the first three appearing to be "old" samples (with some smaller differences, e.g. early ones don't have an adjustable MW antenna coil) and the last one "new" ones.

↑ Back to table of contents

Sanyo Eneloop rechargeables

For someone who likes to use his portables on batteries yet would not like to waste a bunch of primary cells, rechargeables are the power source of choice.

Now conventional NiMH types have a number of drawbacks, one of the biggest ones certainly being high self-discharge rates (even more so in current high-capacity cells than in those with about 2000 mAh or less). If you have a drawer full of radios that may not be used for some time, you can bet that by the time you want to use a particular one, its cells will be almost flat yet again. That doesn't get any better with age, and if you've got some damaged cells in there, it's even more fun. (Again, the very high-capacity stuff has proven quite unreliable, being easily damaged by overcharging. Out of two sets of four 2500 mAh Sanyo cells for camera use, always charged in a "good" charger not known for blatant overcharging as long as the cells were halfway low resistance, a total of three cells showed excessive discharging indicative of internal leakage. Sure enough, the camera always complained about empty batteries pretty soon whenever it was used.) A good occupational therapy. Conventional NiMHs should be called "chargeables", since that's what you end up doing most of the time…

To the rescue came Sanyo's Eneloop cells, the first exponents of the new crop of NiMH+ / "low self discharge" / "ready to use" cells. First introduced in Japan late 2005, they slowly became available worldwide the next year. Nominal capacities include 2000 mAh in AA (R6) and 800 mAh in AAA (R3) size, with C and D sizes being covered by mechanical adapters (C and D sized cells with minimum capacities of 3000 and 5700 mAh and PTC-based short-circuit protection were eventually announced in 2008, suspected to contain three AAA or AA cells each, respectively). While capacities were somewhat smaller than what was available in conventional NiMH cells, they soon built a reputation for low self-discharging, low internal resistance (with decent performance even at low temperatures – the product launch "down under" was staged in an ice bar, now how cool (literally) is that?) and excellent matching that made creating hand-matched sets of cells redundant. There have also been very few reports of failures (which is to say that I haven't been able to find any, just statements of very low failure rates).

Pricing of Eneloops is about 10 EUR for a set of four AA or AAA cells.

Competing cells, marketed under quite a number of brands but only manufactured by a handful of OEMs, typically have a specified capacity of 2100 mAh in AA size and somewhat worse properties otherwise. They do not all appear to be as rugged as the Sanyo cells.

Currently I have several sets of Eneloops in use, mostly AA (plus 4 pieces AAA), first bought for my camera (a somewhat picky beast) and then displacing more and more of the conventional NiMH cells in my radios (currently none of the 2500 mAh Sanyos are in use and all of the crappy old 2200 mAh noname cells have gone to recycling, but some 1850 mAh Sanyos bought as NOS remain). Do they live up to their reputation?

In a word, yes. While they will eventually lose some voltage when sitting around for a while (but then still give off plenty of charge even at untypically low levels) and I guess internal resistance will creep up as in conventional NiMHs, self-discharging is much reduced, and so is the number of extra charging cycles – actually, I don't think I've recharged any of them for anything but "real" current drain from the devices they were used in. In both my little Olympus (2x AA) and my dad's Fuji (4x AA), they keep on going and going. During charging at 700 mA, they do not get overly warm, and end-of-charge detection works reliably (the charger has a habit of failing to do this given cells with high internal resistance). Voltages and capacities stray very little between cells, to the point that they may be within a mV even after some time. Accidental deep discharging happened once or twice so far, from which the cells seem to have recovered well. Failures? None so far.

The verdict? Eneloops have definitely made life with rechargeables a lot easier. They go a long way towards closing the gap between conventional NiMHs and the user-friendliness of lithium based cells. You still need a decent charger for them and deep discharging should be avoided as much as for other NiMH cells, but other than that they're pretty much bulletproof. Recommended? You bet!

Postscript: By now, you can also obtain higher-capacity "XX powered by Eneloop" cells with 2500 mAh (HR-3UWX, vs. HR-3UTG for the "classic" Eneloops or HR-3UTGA for the 2nd-generation ones). These are mainly geared towards digital cameras and such, with the expected results of thinner membranes: Higher self-discharge rates, less cycles, less robustness. Inner resistance still seems fine though. I think us radio nuts are better off sticking with the standard types, which are cheaper to boot (plus you can also buy them in flashy colors if you're so inclined).

↑ Back to table of contents

Sony ICF-2001D

In Feb 2009, Helmut G. Vogel generously lent me his early production ICF-2001D, S/N 30809 (mid/late 1985 or thereabouts), for evaluation purposes.

Intro

How does one review a receiver legend? Well, let's start with the basics: This model (along with its North American counterpart ICF-2010), first available early 1985, was Sony's top shortwave portable and the first ever consumer-level set to feature synchronous AM detection. While the ICF-2001D was discontinued when the ICF-SW77 arrived in about 1991/1992, the '2010 remained available until 2003 at a typical US$379 (MSRP US$449). An 18 year production run may well be a record for consumer electronics. (Only the ICF-EX5, a Japan market set, still beats that – also introduced in 1985, it didn't begin to disappear from store shelves until early 2009, only to be replaced by an updated "Mk2" model later that year.)

What we're looking at is a lap-size portable measuring an official 28.8 x 15.9 x 5.2 cm³ (a longish shape) and weighing about 1.8 kg including batteries – which would be three D cells (main) and two AA cells (backup). There is a slide-out telescopic antenna which measures an impressive 120 cm in length while still making a fairly robust impression. A magnetically retained (!) flip stand allows for a low angled position on the desk. A shoulder strap for carrying the set can be attached and was originally supplied.

Tuning ranges vary depending on version. There is an AM range potentially extending from 150 kHz to 29999.9 kHz (but limited to 26100 kHz in e.g. German version) covered in 1 kHz or 100 Hz tuning steps, the FM band is covered from 76 MHz (or 87.5 MHz) to 108 MHz in 50 kHz steps, and there may be an aeronautical (air) band ranging from 116 to 136 MHz, tuned in 25 kHz steps.

For cruising the waves, you get a number of tuning options: First there is a tuning knob at the side, with two user-selectable step sizes and thus speeds on the AM ranges. Then there is a numeric keypad for direct frequency entry. Finally you get no less than 32 individual buttons for presets, which double for things like direct broadcast band selection.

On the AM ranges, you have the choice among several modes: Wide and narrow IF bandwidths can be selected, synchronous detection toggled on and off, and finally you can also set up the receiver to tune in SSB transmissions in either the upper or the lower sideband.

There is a two-part LC display with a clock / timer and a frequency section. Frequencies are shown to 100 Hz on the AM ranges and to 1 kHz elsewhere.

Antennas for both AM and FM(/AIR) ranges can be connected (3.5 mm jacks), as well as a power supply and headphones (mono). A (mic-level) line-out is also provided. The set was supplied with an antenna adapter that takes coax.

The main opponent in reception will be the venerable ICF-SW7600, freshly recapped by aforementioned gentleman (his third "patient" in a row, all with the caps measurably dried out and with PCB corrosion on one sample). Two fairly full backup AAs and three 4000 mAh D size NiCads (also rather full) were installed in the 2001D.

Operation

• Display viewing angles are best when looked at from the bottom. Seemingly the tilted position is preferred.
• The frequency numerals have about the same size as on the ICF-SW7600(G) or E100. This is not incredibly large for a set this size. But then this also is a mid-80s set.
• The display backlight is microprocessor controlled and automatically delays its timeout if any of the (electronic) controls are used while it's on. However, the lighting itself with a single green LED on the right-hand side can only be called pitiful – dim and uneven. Even the ICF-SW7600G (which does not win any prizes for backlighting to begin with) does better than that.
• The buttons seem to be microswitches behind a rubber mat. Thus they don't feel as precise as on newer sets, but it's OK nonetheless.
• The recessed sliders and switches are not a joy to operate. I always have trouble finding the attenuator switch.
• The smoothly turning tuning knob with an optical encoder (at least it feels like one and is not jumpy) is nice to operate. Its position at the far right of the set is not ideal for left-handed people though, it pretty much has to be spun with the right hand.
• Carrying the set is a little awkward. This one came without the shoulder strap but I wouldn't consider that to be terribly useful either. A handle would have been useful. (In fact, some people have modified their '2010 with one.) As-is, the set is a little clunky.
• On the AM ranges, the presets also store reception mode and IF bandwidth settings, which is fairly neat.

Shortwave reception

• Manual tuning can be carried out in either 1 kHz (FAST) or 100 Hz (SLOW) steps on the AM ranges. While the tuning knob has a good feel, the momentary muting upon each step (up to a certain tuning rate) makes things rather "chuffy". The muting behavior is reminiscent of the ICF-SW7600. Single 100 Hz steps show noticeable "plop" noises as the PLL locks (something which is said to be quite annoying in the ICF-SW77 with its yet smaller steps). Tuning speed in FAST is fast enough for band cruising, quick rotation of the tuning knob will get you along at about 160 kHz per second. (The band cruiser may be annoyed by the chuffing, however.) 5 kHz steps are only used during automatic search. When you have reached the end of a band, there is no automatic wraparound – the set's age is showing here.
• Raw sensitivity is extremely close to the ICF-SW7600 over most of the shortwave spectrum (both running off rechargeables). This was to be expected given the very similar frontend circuitry – I've been calling the SW7600 a "mini 2001D" due to the similarities in architecture. Interestingly enough, the '2001D/2010 enjoys a good reputation for sensitivity even off the whip – I'd guess operation with proper grounding (e.g. mains supply) plays a role here. The 2001D did, however, show better antenna matching near the bottom end of the shortwave range – weak pirates slightly above 1.6 MHz (as well as 160m ham stations) came in much weaker on the SW7600, but attaching a bit of wire to the latter's telescopic antenna evened out the difference (while having no effect on the 2001D). A whip antenna tuner makes a handy accessory for either set.
  It be noted that this particular set has seen a preamp FET repair, a very common defect in this model since protection diodes are installed for the telescopic antenna but not the external antenna input. Here the dead 2SK152-1 was replaced by a 2N3819. Improved protection for the FET was eventually implemented in the early '90s (prior to 1992).
• AGC action at low levels is as good as you'd expect in a top-flight portable, slightly improving upon the ICF-SW7600. (The improvement might, however, also be due to a better-performing AM detector – Sony's synchronous detection ICs appear to be using a pseudosynchronous detector for envelope detection.)
• As far as selectivity is concerned, the two bandwidths do what they say – wide (seemingly a nominal 9 kHz, 6-element filter?!) is fine for listening to strong undisturbed stations, but hardly first choice for 5 kHz separation, while narrow (a nominal 3 kHz, 6-element filter, with some additional audio lowpass filtering kicking in as well) gives a nice and tight filter bandpass for tricky cases at the cost of fairly muffled audio. The wide filter appears to be slightly off-center or asymmetrical, as sound in sync USB and LSB is noticeably different.
• The three-position tone switch is effective. The middle position engages a high-cut, and the lowest position gives a stronger high-cut combined with a low-cut.
• Synchronous detection works very well. Its locking range is limited to a few hundred Hz around the carrier frequency, but it really takes a lousy signal for it not to lock. (In that regard it's more similar to the AR7030's synch detector than the ICF-SW7600G's which cannot keep up here.) It does have its quirks, like the lock and USB/LSB indicator not always being lit even if lock has in fact been obtained (this is coupled with operation of the 2nd s-meter LED – press BATTERY CHECK, and you'll see the sideband LED light up!), or the sideband selection by tuned frequency. Sync detection is very effective against hectic selective fading distortion as well as AM detector distortion in general, especially to be noticed in the wide bandwidth setting. Even for rather weak stations it frequently makes things more listenable, thus giving the set the edge over the ICF-SW7600. (The ICF-SW7600G does not do that much worse on halfway strong signals, but its "communications-quality" audio response hides much of the sync detector's benefits while bringing out the noise.) You may want to back off the RF gain on very strong stations though, or else distortion may creep back in. Opposite sideband suppression is not breathtaking but OK. Drifty pirate or clandestine stations with some frequency wobbling are a problem for the sync, as they will drift off tune after a while (but that's no different from the 2nd-generation sync in the ICF-SW7600G).
• SSB reception, with the narrow filter and a sideband-selective product detector (as usual for Sony sets with synch detection), works well, with no wobbling or such. 100 Hz steps are a little on the coarse side of things, however. Strong stations will have a bit of distortion at the beginning due to the AGC rise time. In addition, the audio is somewhat muffled as the carrier frequency always is right in the middle of the filter bandwidth – an offset of +/-1 kHz für LSB/USB would have been useful.
• Image rejection is very good, much better than in the ICF-SW7600(G). (That was to be expected.) It even beats my previous image rejection champs, E100 and DE1102 – where those will show weakish audio, there's no more than a weak het in SSB.
• Spurious signals also showed to be well-suppressed. The strong ones are audibly weaker than on the ICF-SW7600(G).
• The signal strength indicator, carried out as a line of LEDs, covers a fairly large and useful range. For most of the shortwave range, about 4 lit LEDs are the equivalent of the ICF-SW7600's TUNE LED being on.
• The RF gain control is a bit of a joke. While it is effective (it's an attenuator in front of the 2nd mixer btw), its effect abruptly sets in about halfway from the top. (It be noted that this may not be typical but rather a problem of this sample; pot cleaning had been required in the past to make it work again at all.)
• The attenuator, similar to the RP2000 (or – at least early-model – DE1103), is located after RF preamp stages for the whole AM range.
• AM audio is somewhat lighter in the bass when compared to the ICF-SW7600 but not unpleasant otherwise. The 3-position tone switch is fairly effective but doesn't turn the wide bandwidth into a compromise bandwidth either.
• I noticed a few weak interfering signals / hets which seem to be internally generated as other sets did not show them. Compared to later samples, you still see less shielding on the back of the circuit board of this one if you take a peek inside, so I guess this was improved later on.

MW/LW reception

• MW sensitivity is somewhat better than on the ICF-SW7600(G), with less noise on weak stations, overall more in line with the ICF-7600A (my current sensitivity champ in its size class) – that was to be expected with the somewhat larger ferrite rod. The narrow filter position and better weak-signal AGC action) also play in, however.
• Nearby strong-signal handling, as tested with inductive coupling to my "AM loupe" and the strongest MW local, is somewhere between ICF-SW7600 (very good) and ICF-SW7600G (middle of the road), as I did manage to overload the IF stages (splatter and cross-mod on neighboring channels)... closer to the SW7600G I'd say (The SW7600 remained unimpressed in this case, in spite of its wider filter.) Backing off the RF gain or switching in the attenuator brought things back to linear operation, with the attenuator being more effective at higher preserved sensitivity. This would point towards the 1st mixer as the likely culprit, which according to the schematic has a fairly low supply voltage and thus was possibly running into clipping. In the ICF-SW7600G, the attenuator helps only a little at best, so the intermod is probably generated by the 2nd mixer or the preceding IF amp.
• 2nd-order intermod seems to be somewhat better controlled compared to other portables with a wideband MW frontend. That makes sense given that the higher signal levels from the ferrite rod allow for some source degeneration (for better linearity at reduced gain) at the RF amp. Don't expect any miracles though.

FM reception

While the set does appear to be decently sensitive, selectivity with the two stock 280 kHz filters (barn doors) is hardly to be called exciting. Some modern-day low-loss 180s or 150s would work wonders here.

Air band reception

In terms of real signals, I caught one (1) volmet station. I could hardly conceal my excitement. Seems like this is not a terribly interesting region in terms of air band. There were only a few spurs and mixing products, but then FM band signal levels aren't exceptionally high here either.

Sound

Speaker audio, while frequently criticized, is perfectly fine for the speaker size (looks like a 10 cm / 4" job and consequently outclasses my sets with smaller speakers). Nice balanced portable audio, maybe not breathtaking but perfectly adequate on FM. Even on the AM ranges where the low end is noticeably reduced, it still manages to sound somewhat more "grown up" than the ICF-SW7600 overall, although the difference is smaller.
There is some handwritten data on a label on the back of the speaker driver, including a resonant frequency of 175 Hz, impedance of 4.2 ohms and a 1984 build date.

Hiss levels on headphones are average, maybe a touch more than on the ICF-SW7600 if only one channel is used, but lower if it's split to supply both drivers.

Quirks

There are a number of documented quirks on this set. One is the battery-to-circuit-board contacts for both sets of batteries (solder pads which contacts press against, aided by the back case), which if subpar may cause "Error 3" messages or loss of time and memory contents. The backup batteries do not sit terribly firm either. While there is a main supply voltage line that goes to the control PCB for display lighting, this was not used for implementing a "backup backup" supply, i.e. supply of the control circuits from the main batteries if backup supply is out. This is all the more surprising considering that the older ICF-7600D/2002 has it.

Many of these sets suffer from reduced sensitivity on the shortwave bands due to a blown Q303 (2SK152-1) FET. Protection diodes are effective for the telescopic antenna but not the external antenna input!

Conclusion

Overall, in spite of several idiosyncrasies, this is a fairly decent upperclass portable. With some better IF filters and a tunable antenna (I'd also suggest an audio mod for some more bass in the wide AM mode at least), it's likely to be good for some fun on the short waves. It improves upon lesser sets not so much in terms of pure reception capability but rather reliability (image rejection, frequency readout and tuning steps) and flexibility (e.g. bandwidth selection). A notable exception is the synch detector which does improve reception even on weak signals. Overall, it's no surprise that the '2001D turned out to be a popular platform for modifications – good basic substance with a few quirks.

Would this be the ideal receiver for me? Well, it's a little clunky, and I don't have a charger for D cells (and still no mono adapter for headphones either, believe that?). A number of people use it as a nightstand radio, and indeed it seems to be well-suited for such semi-stationary use.

Notes

There is a host of information on these sets available on the web. Just a few helpful links:

Apparently late-production ICF-2010s (post-1998 or thereabouts) are equipped with AM IF filters sourced in China which show less good quality than the previously-used Murata filters and may exhibit asymmetry in particular (see icf2010 group message 2779).

Alignment notes

First off, unless you're exactly sure you know what you're doing and very careful to boot (it seems very easy to damage the ferrite antenna wires), better leave alignment to an expert!

The sync / SSB alignment procedure as outlined in the service manual aims to accomplish the following:

• First of all, VCO mid-point tuning frequency is adjusted to 3640 kHz (via CT601) at 1.5 V (VT1 setting) in SSB mode. The VCO runs at 8 times the 2nd IF.
• Following this, the USB/LSB switchover point is adjusted in synch mode via VT2 (which controls a comparator's switching voltage).
• Finally VCO mid-point frequency is checked again to make sure that there has been no PLL reference induced error (which would have to be corrected via CT4).

Carried out like this, there is no way to accomodate non-centered IF filters. You can tweak VT1 afterwards for a symmetrical response in SSB, but then displayed frequency will be off and there will be no change in synch mode. Alternatively you could choose main tuning frequency a few steps higher or lower than nominal at the beginning until you get a symmetrical response in USB and LSB, but then this display offset will be present all the time. (On more advanced receivers like the venerable AOR AR7030, the displayed frequency is corrected by the offset required.)

If you notice a slight frequency shift towards the higher shortwave bands (i.e. you had things adjusted to zero beat on MW/LW yet there is an audible offset up there), a tweak of the 6.275 MHz PLL reference xtal is needed (CT4). Try adjusting CT4 for zero beat on the highest-frequency (reliable) shortwave broadcast or time signal station you find (taking into account the previously found frequency offset, of course), then check back at MW/LW. Chances are there will be a slight offset down there now. Using CT4, aim for the same offset on both frequencies (best stored in presets), then zero beat again via CT601.

I had been wondering why they chose to make the 2nd LO frequency non-adjustable. The answer is quite simple: It doesn't matter! Not within a few kHz at least. This is because it is also used for pre-mixing in the PLL for the 1st LO. If it's off, the 1st LO will be off by the same amount. This merely shifts the effective 1st IF, which doesn't matter much because the filtering up there is relatively wide (15 kHz or thereabouts) and thus quite tolerant. This approach also reduces frequency errors introduced by the PLL reference oscillator being off. At the same time, it introduces the aforementioned problem of not being able to cater for off-center IF filters as easily.

↑ Back to table of contents

Tecsun AN-200 AM (MW) loop

Intro

Santa was a little early in bringing me this compact tunable AM loop just before Christmas 2009... I had actually ordered it together with the 2010 edition of Sender & Frequenzen but it turned out to be backordered. It would seem that others were equally aware of effectively 19€ being a good deal. Normally you can also pay twice that.

What you will get when ordering the AN-200 is the compact tunable AM loop itself, an instruction sheet in reasonable-quality English (which claims it's a MW/LW antenna) and a 1/8" male to male stereo cable that can be used to connect the antenna's pickup winding to a receiver's antenna input.

Performance and handling

• The antenna itself will peak up between about 500 kHz and a good 2.0 MHz, great for 160m. (Maybe I got lucky, as some production tolerances are to be expected. The loop would only be expected to tune ca. 520 kHz to 1710 kHz with a bit of safety margin.)
• Q seems quite good, with bandwidths no more than a few 10 kHz even on 160m.
• I found it to give some healthy gain over the built-in ferrite rod when used with inductive coupling to my trusty Sony ICF-SW7600G – I wouldn't be surprised if it were close to 20 dB with the set inside the loop. You can easily get to the natural noise floor even with the loop being next to the receiver though. A barely detectable signal (O=1) could be brought up to audibly noisy but perfectly listenable (O=3), and an O=3 signal found itself boosted to O=4 levels (very solid reception, almost undisturbed) easily. In order to hear anything more, I think I'd have to put up something much bigger in size, like a Beverage or similar, and that's just not doable here. Direct coupling by using the 1/8" connector, by contrast, yielded unspectacular levels and seems to drag down antenna Q considerably, too.
• Directionality seems about the same as with the Sony's internal ferrite rod.
• As there is no reduction gear, tuning requires a bit of care.
• The compact dimensions of the antenna make it ideal for use with portables. (It'll still barely fit a Redsun RP2000/RP2100, inner diameter is about 21.5 cm.) However, there is no way of keeping a pocketbook-sized or smaller radio inside the loop – I'm using an external box as sort of a stand to bring the set closer to the loop's axis.
• If there is one quirk, it is the antenna wires being on the outside of the loop and thus easily touched. This tends to throw off the tuning as your fingers make for some fine inter-winding capacitance, so a bit of care is required when handling the antenna.

(Those wondering how I got the ICF-SW7600G to receive 160 meters with the ferrite rod please be referred to the review. A DE1103 / E5 would do this out of the box.)

Look and feel

Externally the AN-200 is quite attractive and equally decorates your desk when not in use. It is also fairly well-made. The tuning knob is reasonably sized (Ø ~3 cm, ~1.8 cm height), doesn't wobble and turns smoothly. Judging from the feel and tuning behavior, it controls a varicap of the plastic variety. Do not ask me how to disassemble the antenna – I'd guess the two halves of the plastic foot are clipped together, as there are no screws visible anywhere.

Conclusion

/me likes! The AN-200 works well, looks nice (a distinct advantage of a commercially made product) and doesn't give the impression of falling apart tomorrow, all this while not costing an arm and a leg. Thumbs up.

So far, the antenna has held up well, even with a bit of rough handling. Amazon reviews indicate that people have received the occasional dud though, in which case it's best sent back. A good one definitely is a keeper.

Note: A small 9" dia. loop with 25 windings occupying about 1" has about 250 µH of inductance (going by the Bob's Tesla Web Lab formula), giving an X_L of about 800 Ω to 2.7 kΩ across the MW band. Since my sample will tune up to pretty much exactly 2 MHz, that means minimum capacitance (including inter-winding) is about 25 pF, down from a maximum of about 400 pF. Not a bad little tuning cap there, and wire routing seems quite solid, too.
I probably don't need to mention that near the upper end of its tuning range, the antenna is rather touchy in the literal sense – approaching the windings with your fingers (thereby increasing inter-winding capacitance) throws off tuning considerably. (So far, none of these loops has been caught singing Polly Scattergood's "Please don't touch", but you never know, HI.)

Here's something for you if you feel like simulating inductive coupling operation in LTspice. This was based on what I saw with an AN-200 / ICF-SW7600G combo. Coupling factors K1 and K3 may not be 100% accurate but should be in the right ballpark.

↑ Back to table of contents

Homebrew lower shortwave loop

(Rebuild of homebrew AM loop, Nov 2010.)

Intro

My homebrew AM loop recently mutated into an antenna for the higher MW and lower shortwave bands. It gained a proper stand using a piece of big heavy beam and lost 5 turns and 250 pF in the process, was rotated by 45 degrees, now has strain relief measures on both ends of the windings, and the tuning capacitor was relocated to a spot much closer to a corner, thus remedying several problems that plagued the original first-try design (which liked to tip over, had way too much coupling to ground when standing on the floor and wasn't too exciting in terms of wire tension; it also made use of one ground wiper and had excessive high-impedance lead lengths as the capacitor was located in the center). What remains is 4 turns of 4 mm² insulated braided wire spaced at 2 cm on a square frame of about 93 x 93 cm standing on a corner, with two 500 pF sections on a split stator capacitor in series.

The tuning range now extends from about 1450 to a little above 4800 kHz, a bit lower than I would have liked. However, going to three turns would have shifted things upwards too much according to several calculation aids. As-is, wire length is between approximately λ/14 and λ/4.

A first pickup loop using the same wire (with a twisted section going from the loop to the receiver) also was added, though the excessive wire thickness makes handling cumbersome. Due to high impedance at this point and modest wire lengths, pickup loop conductivity is not nearly as critical, so thinner wire can be used with no problem and I'm likely to revise this point. The pickup loop is triangular here, about 1/3 loop diagonal wide and 1/2 diagonal long, extending between the corner opposite the tuning capacitor and a center beam. Its symmetry follows the rest of the loop. With a coverage of about 1/4 to 1/5 of loop area, loop impedance is transformed to a value that's still well in the kΩ range. It ought to be much smaller for a 50 Ω input (with tighter coupling, too), but a high-impedance input will do fine here.

Finding the best spot for a coupling loop is easy: it should be where touching a loop winding throws off tuning the least. (After all, we don't want to degrade performance due to capacitive coupling.) This turns out to be exactly opposite the tuning capacitor, at the center of the windings. Not too surprising, as the high impedance point is at the tuning capacitor (also allowing whip antenna coupling there) and the low impedance point should be on the other side.

I later added another, smaller coupling loop using thinner, more flexible wire, again in triangular shape. Its base of about 27 cm runs right next to one of the middle wires opposite the variable capacitor for optimum coupling (so right in the low-impedance region where current is maximum), its sides are about 20 and 24 cm respectively, and again the remainder of the wire is twisted and led to the receiver. This seems to come much closer to a 50 Ω impedance.

Evaluation

The whole affair is undoubtably rather big (about 1.30 m wide and 1.35 m high), a little heavy and thus somewhat cumbersome to handle. While you can place a little 9" loop like the AN-200 just about anywhere, such a big loop has to sit somewhere on the floor where there is enough space for it, rather than in the spot with best reception.

Inductive coupling yields big signal levels as before, well into the noise floor. (I can still make portables break into sweat on MW.) Peaking is fairly sharp, so that reduction drive on the tuning cap and big knob (Al-over-plastic volume knob from a defective late-'70s audio amplifier, basic model but nice build) comes in handy. Q seems to be around 100 on 160m, rather low for that kind of wire gauge, but possibly explained by insulation-related loss.

Directionality is there but, seemingly typical for loops, is most pronounced when signal sources are located about on the same level (which makes sense if you consider its approximately toroidal pickup pattern, same as for a dipole). As expected for an antenna relatively close to ground (I don't have a mast and rotor here...), signals with higher incident angles are preferred. Nearfield directivity appears to vary with frequency.

The connection from pickup loop to receiver is normally carried out as a coaxial cable for good reason: The signal difference between "both legs connected" (differential mode) and "ground connection removed" (common mode) turned out to be only about 12 dB with the "twisted-pair" cable on the big pickup loop. It's not a big deal, since the signal picked up from the loop easily dominates any feedline signal pickup and the feedline is only about 2 meters long anyway, but should be kept in mind for longer connections.

Loop self-resonance (?) seems to occur at about 11.7 MHz when using the big coupling loop. With the small coupling loop, this second peak shifts to about 13 MHz.

The small pickup loop seems to come much closer to a 50 Ω impedance. On 160m, there is no difference in signal level whether or not a FET source follower is used in front of a 50 Ω input (but noise is lower without it), while by 80m the difference has risen to about 6 dB more when using the source follower. A medium-impedance input (nominal 600 Ω) cannot keep up with either. All of this indicates a source impedance that is very low at the low end of the tuning range and then rises to about 50 Ω and a little beyond towards the high end. Indeed, an impedance variation about as large as frequency variation is to be expected when using this coupling method. I may still try a gamma match or somesuch, even though for receive-only purposes it's plenty sufficient as-is, given that received background noise is much higher than receiver noise when the antenna is tuned properly.
The FET-dominated noise floor in the low-impedance region does not come unexpected either, as their higher voltage noise compared to bipolar transistors means that noise figure will be worse at very low source impedances. (The J310 used in this case is by no means particularly noisy for a FET, but its 6 nV/√(Hz) voltage noise level is equivalent to a 2.2 kOhm resistor's. With a low source impedance, that's enough to be detected over the noise floor of a 50 Ω receiver with a ~12 dB noise figure, which amounts to 1.7 nV/√(Hz) or 3.4 nV/√(Hz) EMF. Since the FET has a high input impedance on the lower frequencies, the latter is better suited for comparison.)

↑ Back to table of contents

Sony AN-LP1 shortwave loop: Servicing notes

The AN-LP1 is an active loop antenna for approx. the 75m through 13m range, once designed to work with Sony's mid-90s shortwave portables – notably ICF-SW7600G and GR, ICF-SW100 and ICF-SW1000T. Several other portables are also able to use it, minus the automatic turn-on feature which is only available on the aforementioned sets. I reviewed it many moons ago, with the result admittedly suffering from being badly structured, as all my early reviews were – but there you have it.

Now my unit wasn't entirely perfect, even when not taking the mods performed by the previous owner into account: Unrolling the wire to the antenna frame gave quite a bit of scratching noises during reception, the bandswitch could make somewhat better contact as well, and the peaking on some reception ranges seemed noticeably off.

So after managing to get off a sticker that the previous owner had placed on the back last night (which obstructed one screw), I took my AN-LP1's control unit apart today (2011-05-15). It's not too hard if you're reasonably handy and have the service docs at hand, mostly the usual stuff like making sure small screws don't grow legs and such. A Phillips 0 sized screwdriver is fine.

Reinstalling the takeup reel is a bit fussy as the cable needs to be threaded back in, which is a little annoying as you may have to do that a few times. Reinstalling the screws in the battery compartment is another minor hurdle, you may want to hold the unit upside down or use a magnetic screwdriver (my small ones aren't).

Alignment of the teeny tiny trim pots is reasonably easy with a very small flat-head precision screwdriver. The main problem is that the front cabinet needs to be off, so you'll need to find some other way to turn the miniature surface-mount rotary switch that changes bands. (I found that a somewhat larger flat-head screwdriver did a good-enough job.) I recommend turning the band selector on the front at the same time, so that things will still line up at the end.

Instead of a SSG, I used plain ol' background noise, with the takeup reel put back in its usual place and the loop connected. Works well enough. The receiver used was my trusty '97 ICF-SW7600G.

Following the alignment procedure, I found that a number of the bands were quite a bit off on my sample. Not sure whether this is due to aging or it came like that from the factory. I don't think the control unit had ever been opened before. A few bands were reasonably close, but those were a minority. The only range I didn't touch was 5 MHz, which I'd always found to work great.

I mostly used the specified alignment frequencies, with only a few exceptions:

• 4: Originally this peaked well above 4 MHz, spec'd is 3950 kHz. I decided to peak it at 3850 to get a reasonable compromise between 80m and 75m performance. (The SSB section of the 80m ham band starts at 3600 here, and 75m stations are rarely encountered beyond 4010.)
• 6: Taking the geometric mean of expected band edges 5850 and 6400 gave me about 6120 instead of specified 6075, so I peaked on 6100. Really minor tweak.
• 10:Was a bit tricky to peak up. Checked at 9400 and 10000 to make sure sensitivity was about equal there.
• 16: They suggest 16500 for that one, I'd say it should be a touch higher. Checked on 15600 and 17900 to make sure none of the band edges suffered too much. (PS: Err, why 15600? Should have been more like 15100.)

I took some time cleaning the takeup reel contacts and wiping down the corresponding contact surfaces on the PCB with my usual sealing contact cleaner (Teslanol t6 "Oszillin", about the equivalent of the usual, least aggressive type of DeOxit, I guess). If that doesn't do it in the long run, I may try the kind of contact grease used on car batteries. (The original grease is some greenish stuff.) Sure seems a good bit less scratchy for now.

The bandswitch was another candidate that could make somewhat better contact. After I couldn't find any holes in this thin, surface-mounted thing, I didn't feel much like opening it up. Apparently you could remove the metal cover (certainly so after desoldering), but I wasn't too keen on having miniature switch internals flying about or potentially not getting things back together. Know your limits.

Now for final performance evalution in the markedly improved propagation conditions of May 2011 (really no comparison to last winter):
The realigned AN-LP1 manages to get background noise above the ICF-SW7600G's internal noise floor pretty much all the way from 3.5 MHz to about 25 MHz. (That 20 MHz range has picked up nicely.) The only really weak spot with sub-whip levels is in the 8.5 MHz vicinity, but that was to be expected with the 7 and 10 MHz bands being as far apart as they are.
Using the 14 MHz range on 20m somewhat reduces the levels of 2nd IF images from 19m on the ICF-SW7600G, but since relative frequency difference is a lot smaller compared to the 49m/60m case, the effect isn't nearly as pronounced. On 60m, the AN-LP1 pretty much makes the set bulletproof, with an estimated 20 dB increase in image rejection.
There still is noticeably directivity on the 40m band, but it's pretty much gone an octave higher.

↑ Back to table of contents

AOR AR7030 PLUS

(Sample obtained new, 2002.)

Intro

The AR7030, here in the "Plus" version with an optical encoder for the tuning knob rather than the plain mechanical one of the base model, is a "prosumer" communications receiver that covers 0 .. 32 MHz continuously in minimum tuning steps of about 2.5 Hz. Introduced in 1996, it is a dual conversion design with IFs of 45 MHz and 455 kHz. Two ceramic filters are being cascaded for particularly steep slopes. A combination of high input IP3 mixer and a wideband frontend (with only some lowpass / highpass filtering above / below ca. 1.7 MHz) is being used. Both 50 ohm coaxial and 600 ohm spring terminal inputs for wire are being provided; the coaxial input can also take a whip antenna which enables a JFET-based preamplifier to provide a high-impedance input.

This receiver is the brain child of John Thorpe, previously responsible for a number of Lowe receivers, and arguably his most advanced design. His previous designs were known for good audio quality and AM synchronous detection, and this basically continues in this model. Further emphasis is placed on frequency stability, and all oscillator frequencies are being derived from a single TCXO by a DDS PLL synthesizer with low phase noise.

I did find myself wishing more than once that I had gotten the noise blanker and notch filter option as well. Oh well, it was quite the expense for my 20-year-old self as-is. Never really got that much use out of it. Come to think of it, I've often had more luck buying things used than new.

A Discussion of Niggles

I am not going to bore you with a full review of this unit, which has probably been done in depth before. Instead, I am going to focus on a few select issues.

Headphone Audio Quality

The AR7030 uses a '90s Philips audio processor with volume control developed for TVs and such (not the only consumer-level Philips IC in this design). This seems to have poor performance near the low end of the volume control range, as audio becomes thin and distorted there, and channel balance suffers as well. It's fully there at the 20% level, which already is a bit louder than I'd like to listen to with 300 ohm Sennheiser HD580s, hardly the world's most sensitive cans. For comfort, I had to resort to my least sensitive headphones. Some of my more sensitive ones (e.g. Sennheiser HD590, ca. 110 dB/V) also revealed some wideband noise.

These issues can be addressed relatively easily by using either a passive attenuator or an external headphone amplifier. I used a FiiO E11 to good results. Despite the unit's internal DC/DC converter, what little interference issues there were could be easily avoided e.g. by not placing the unit right next to loop antennas of all kinds, and in the electrostatic domain things were totally clean.

Synchronous Detector Low-Level Tones

The synchronous detector basically is really nice, with one annoying quirk: With a good, clean signal, you can often hear a low-level tone in the background. I found this super annoying at times. If memory serves, the designer stated that this could have been done away with but it would have made the synchronous detector more complex and more expensive. Oh well.

Oddball Frequency Steps

The smallest frequency step is only approximately 2.5 Hz – in fact it's 2.655 Hz. This probably relates to the oddball reference frequency of 11.13625 MHz, which is also quadrupled to derive the 2nd LO frequency of 45 MHz - 455 kHz = 44.545 MHz with a minimum of complexity and extra phase noise. When trying to step in e.g. 5 kHz SW broadcast band steps, this results in an ever larger error accumulating as you go along. Quite annoying. Such quirks actually are a rather common trait of leftfield trick circuits, innovative and economic as they may be.

While the TCXO probably is quite accurate as-is – I certainly didn't notice more than a few dozen Hz worth of frequency deviation – this also prevents you from being able to use a 10 MHz lab frequency reference.

Handling of Large Antennas

As good as the first mixer is (the 7030 still places in the upper third of the Sherwood Engineering Receiver Test Data Table, and it doesn't look like any other receiver its age quite reaches or betters it) and as reasonably as sensitivity has been chosen, it is actually not super hard to provoke 2nd-order intermod in this model, due to the minimal frontend filtering.

The fix is the usual one – either engage some of the built-in attenuation (not particularly effective on 2nd-order intermod but it helps) or use some sort of frequency-selective antenna (e.g. a tuned loop), a preselector or other such filtering.

Leaking Backup Battery

The unit uses a rechargeable 3.6 V NiCad battery pack for clock backup. These are known to leak on a regular basis (much like the ones that commonly graced PC motherboards in the early '90s), about 15 years is reported to be a good estimate for their life. I presume being trickle charged in operation doesn't help. As such, with my receiver being unplugged most of the time I'd hope the thing lasts longer than that. I certainly haven't looked inside in a good while.

Entry last modified: 2020-03-19 – Entry created: 2020-03-19

↑ Back to table of contents

Glossary

AM

Amplitude Modulation, also: mediumwave band (see MW), or (Sony) the whole LW/MW/SW range.

AGC

Automatic Gain Control (older term that basically means the same: AVC = Automatic Volume Control); a measure to ensure that stations are virtually always equally loud regardless of signal strength. Better receivers allow changing AGC time constants and even turning it off which may be helpful in certain DX situations.

BCB

Broadcast Band, see MW.

BTW

A harmless acronym: by the way.

Dual conversion

See IF.

DX

Originally meaning "distance unknown", this acronym is generally understood as reception of distant, hard to hear stations, or chasing for them.

FET

Field Effect Transistor. Commonly used in receiver front ends for RF amplification and mixing. On shortwave and lower, one typically uses n-channel junction FETs, while on the FM band and higher dual-gate MOSFETs are also popular (they can have very good noise figures up there, but are not so great at low frequencies due to high levels of 1/f noise).

FM

Frequency Modulation. This is employed by broadcast stations on the FM band (wideband FM, 150 kHz bandwidth) and the Citizen Band (CB) (narrow-band FM, ~10 kHz). On shortwave FM is basically only used above some 25 MHz (like CB on 27 MHz or part of the 10 m ham band), while it's a widespread standard modulation type on higher frequencies.

Ham

Infml.: radio amateur.

IC

Integrated Circuit; a compact part that offers the functionality of conventional circuits with separate capacitors, coils, resistors and all. A receiver like the ICF-SW7600G(R) uses ICs for demodulation (AM, Sync, SSB, and FM in conjunction with an external filter), for frequency synthesis (PLL), as audio amplifiers, and of course the microprocessor controlling the whole thing can also be seen as an IC.

IF

Intermediate Frequency (of a superheterodyne receiver). Superheterodyne receivers or superhets mix the incoming signal to a constant IF (after some RF pre-amplification, usually) which simplifies filtering enormously when compared to older concepts (superhets were invented in 1917 or so). Since the mixing process generates two signals twice the IF apart (see Image rejection), general coverage shortwave receivers use a rather high (1st) IF, however the higher the IF the more difficult / expensive it is to find good filters. Therefore they're usually dual conversion designs with a high 1st IF (which used to be just a few MHz in the days of tube receivers and lots of separate bands with bandpass filters, and is usually 45...70 MHz today with exceptions like the models with the "cost-effective" typical FM broadcast IF of 10.7 MHz allowing the use of FM IF filters there) which typically uses one or two crystal filters (typically 30, 15 or 8 kHz @-6 dB) and a lower 2nd IF with good filter choice (typically 455 kHz, or 450 kHz on older designs). This usually keeps image frquencies away pretty well, but very compact receivers tend to exhibit leakage issues around the 1st IF filter. Dual conversion receivers also have better dynamic range than single conversion designs built with the same efforts. (See IM.)

IFT

IF-level transformer. When tuned by means of parallel capacitance, acts like an additional LC filter circuit. Tuned IFTs on IF level are typically used to improve ultimate rejection and reject spurious filter responses.

IM or IMD

Intermodulation (distortion); a by-product of strong signals generated by amplifiers, mixers and the like when signals no longer fall into the linear working range. Attenuation of signals can help to eliminate the resulting "ghost stations" (and parts / circuits with better dynamic range, of course). On shortwave the most common types of IM products are 2nd order (usually found far away from the signals generating them, like at the sum of the frequencies of two strong stations) and 3rd order (found within or next to crowded bands). 2nd order IM products can be defeated rather easily with high-/low-/bandpass filters, so 3rd order IM products are seen as being more critical. Less expensive shortwave receivers tend to have wideband frontends, thus they can generate both types.

Image rejection

The ability of a superheterodyne receiver to suppress stations that are twice the IF (or one of the IFs) away.

Intercept point

This is a calculated value given in dBm that can be used to describe how good a receiver is in rejecting intermodulation products of a certain kind (mostly: 2nd order --> IP2, 3rd order --> IP3). The larger, the better. You'll mostly find the input IP3 published, which for many portables is below 0dBm and can reach +30dBm and more on very good receivers. Note that in practice with a high number of signals it's not always the mixer with the highest IP3 that performs best. In addition, you need to know where the input noise floor is (i.e. how sensitive it is when looking at a complete receiver) to judge dynamic range properly – a mixer intended for VHF use is likely to have a low intercept point, but it'll have a low noise figure, too.

LO

Local oscillator; oscillator that is used to generate one input signal fed to a mixer in superheterodyne receivers. It be noted that high LO levels tend to improve strong signal handling, but depending on the mixer may lead to overly high LO radiation (which for FM receivers is straight in the VHF air band).

LW

Longwave; the respective broadcast band (mainly used in Europe) ranges from 150 kHz to 281 kHz, with a 9 kHz channel spacing.

MW

Mediumwave; the respective broadcast band ranges from 531 kHz to 1611 (1700) kHz, with channel spacing being either 9 or 10 kHz depending on region.

PCB

Printed Circuit Board – the basis for all those ICs, capacitors, resistors and stuff.

RX

Receiver.

SSB

Single SideBand; see the ICF-SW7600G additional information section.

SW

Shortwave; generally the frequency range 3...30 MHz, contains various broadcast and other bands.

---

Want to drop me a line?

Created: 2004-06-12
Last modified: 2020-03-19