Armstrong 509 tuner

[Jim Audiomisc]

WRT SDR The FunCube Pro Plus has the advantage of using a 24bit ADC, plus including some RF band filters. The snag is that the sample rate limits the bandwidth to only 192kHz. So a bit narrow for standard Band II FM. Useful for many other purposes, though.

I still prefer older tuners for FM anyway.

 

[Paul R]

Is 192kHz not enough for FM? IIRC Bessel functions are involved, so I'm probably out. I used one once, but I think I got away with it.

 

[Paul R]

I would hope not! But it will be of course
"Progress" is very much a double edged sword and I confess to being a Luddite! Much modern tech makes "consumers" "end users" out of all of us, no matter how much knowledge and experience in electronics one has amassed. I demand from test gear etc that it be repairable by myself, not only repairable but modifiable, able to be re-purposed if required. Often the only progress is in small size, light weight and relative cheapness but the actual performance of older gear is very good and as a bonus it can be kept running properly almost indefinitely.

In the software world there's the concept of 'code smell'. The comms hardware you describe has a bit of a smell, now that generic processing is so powerful.

A classic Tek scope, OTOH, really doesn't and should be cherished.

But my grandparents old multi-standard TV smelled really bad in engineering terms.

 

[Jim Audiomisc]

Is 192kHz not enough for FM? IIRC Bessel functions are involved, so I'm probably out. I used one once, but I think I got away with it.

The standard estimate for FM bandwidth of FM radio comes out at about 2 x (75 + 15) kHz for mono. Hence the 200kHz channel allocation. However for stereo and low distortion you may tend to need more. This is why old 'supertuners' tended to offer a choice of IF bandwidth. Narrower than 200kHz helps to improve rejection of adjacent channel transmissions, but increases the distortion *particularly* for stereo.

Hence 192k would work, but probably give iffy results for stereo. I've not checked, though, so may have a go sometime.

Decoding FM is easy for the FCPP because it gives IQ output so you can simply measure the phase rate versus time. However there is another snag in that it has a few Hz 'hole' near DC into which the signal can vanish as it transits. In effect, a sort of 'crossover distortion'. I dunno if other SDRs have this or allow thought the phase noise from the mixing. The have a higher sample rate, hence a wider bandwidth, but that and the use of 8bit sampling and no RF filter tends to mean they're more easily overloaded by unwanted inputs at other frequencies.

Keep meaning to experiment with this but then devote the 'round tuits' to something else... Not a worry given other tuners that work fine.

 

[davidsrsb]

Even the premium high resolution SDR receivers have found it beneficial to use an analog preselector filter.

 

[Arkless Electronics]

Beam deflection valve mixer, Collins mechanical filter... old tech still competitive today in absolute performance terms.
I always quite fancied a Racal RA17... can't really justify it though!