Oh, I forgot one currently available (at least in the US) radio that seems to enjoy a good reputation: The Sangean DT-200V. This seems to apply mostly to its AM part though, the FM is supposed to be decent but not that exceptional and build quality doesn't seem to win any awards either.
Quote:
Originally Posted by axle
I guess vintage is the best way to go. It's a real mystery why a good portable radio isn't really made anymore... does no one listen to radio these days? I thought technology was advancing.
|
Some of those new highly integrated chips seem to be quite good (they do pull some interesting tricks), but if the primary goal is performance rather than minimal space taken up you'll typically resort to 20+ year old technology. (Heck, some of the very best FM tuners ever are 25 to 30 years old.) Pocket radios for the most part seem to be as simple as it gets and frequently are little more than an application circuit for the all-in-one ICs used (e.g. SRF-59 with CXA1129N), and performance comes out accordingly. (You can make a simple circuit that performs rather well but even that requires some effort.)
I guess the problem is that things that make a radio perform better also cost, in terms of money, space and power consumption: Want to have better frontend selectivity? Then an analog radio will require more tuning caps and inductors (+ board space), and a digital will need more varicap diodes and inductors + board space. Those inductors typically are high-Q air coils, not that small by pocket radio standards. Oh, and you have to align all these circuits, too. Want a dedicated frontend? That might cost a couple BJTs or FETs, but most importantly those have some power consumption, and don't think that an FM IF / AM RF/IF chip would necessarily draw less than its all-in-one counterpart. (Chip choices also shrink as you move down the supply voltage ladder from 3 V to 1.5 V.) Want at least two halfway narrow IF filters? Then an IF amp might be helpful, a few more mA. Want a dedicated MPX chip? There goes your board space, and up goes the power consumption. Not using an all-in-one chip typically means an extra power amplifier has to be used, there goes some more board space (and many of those are plagued by hiss to some extent, plus output power may be an issue at 1.5 V). Sufficiently large output coupling or other electrolytic capacitors need some space, too (you'll want at least 220µ per channel for output coupling - yes, codecs and SoCs for MP3 players can run DC coupled these days, but that doesn't mean that this kind of progress would move on into separate amp chips very quickly -, and power buffering will probably amount to about 470µ or at least 220µ depending on supply voltage), and the most reliable and/or best-performing ones typically aren't the smallest. Last but not least, a sensitive AM section will require a ferrite bar of some size (among radios of the same vintage and integration level, those with the largest ferrite bars also have the best sensitivity). Now what do people want? Radios that are small and inexpensive and run eternally with possibly a single AA or AAA cell. High-efficiency DC/DC converter chips that allow getting pretty much any voltage from 1.5 V exist (there goes some more board space), but that only means a single cell will be drained more quickly. The SRF-M40W is a brick that uses two AA cells and for a PLL rig is quite economical at an IIRC 28 mA measured current on FM, which should make for 50+ hours of listening with alkalines but that already means twice the power consumption of an SRF-59. The RP-20 seems to draw about 40...50 mA from its two AAA cells, that's calling for rechargeables.
We have one advantage these days: Rechargeables have become much better. Given that even a more power hungry radio is no more so than a thrifty MP3 player, one could probably pull off something that runs on two (or maybe even three) AAs and draws up to about 50 mA these days (50 hours of playtime still is quite a bit). Now whether people would buy something that would have to be used with a belt clip due to its size is another question - they'd certainly ask why this is as large or larger than your average hardddrive based MP3 player. You also have to be able to sell enough of these to keep the price down - something beyond the 100€/$ mark would not sell very well. But if you put in the same effort as in the very first Walkman radios in the early '80s, you could doubtlessly make one that performs better today - better RF transistors are available now, and ceramic filters are cheap enough that you can easily use two 180s without worrying (OK, a bit of filter matching may not be the worst idea).
I would certainly love to see something with a high-performance 3-gang frontend (the RP-20 is 2-gang but darn good anyway, so one should be able to build a yet better 3-ganger) and two decently matched 180 kHz filters plus perhaps a low-loss 280 kHz one or an IFT. (That would be about on par with an entry level FM tuner, actually somewhat better in the IF section and somewhat worse to on par in the frontend depending on whether you apply US or European standards. Real high end still is a far cry away but that config should be very usable.) All that with a dedicated MPX that gives good stereo separation and has a low stereo threshold (with correct region-dependent deemphasis, a mono/stereo switch would also be included), a good-sounding amp chip that's not too hissy, and all the other goodies required for good sound. (Two headphone jacks would be neat, particularly if only one of them doubled as an antenna input, possibly with a supplied impedance transformer to connect to a dedicated antenna.) I wouldn't mind analog tuning if that gives better frontend Q (air tuning caps seem to fare best in this regard, dunno 'bout the mica stuff), frequency counter ICs and simple LCDs seem to be dirt cheap anyway. Whether an AM part would be included, good question. It would require more than average effort in the IF filter section (e.g. 2 2-pole ceramics + IFT, maybe some amping) and with the AGC to make it worthwile, and one would have to think of a way to include a decently large ferrite bar without the set getting too clunky (it would have to be detachable or rotatable). Question is, who could pull that off? One of the dedicated radio manufacturers I guess - Sangean or Degen perhaps. Sony seems to have no ambitions in that direction.
