Jasen Betts wrote in a message to Roy J. Tellason:

 RJT>> The other part that jumped out at me as being a little unique
 RJT>> was a 565 PLL chip.  I have heard of this one before,  but have
 RJT>> *no* idea what you'd do with it.  Any of you guys have any
 RJT>> thoughts on the matter

 JB>> ISTR that a PLL is a kind of frequency follower, other than using
 JB>> it in a frequency synthesiser (in a radio tuner) or frequency
 JB>> multiplier etc I don't know what you could do with it,

 RJT> Sounds about right.

 JB>> HMM, maybe you've got a parallel-port radio tuner kit there.

 RJT> Could be...

 RJT> I've been bumping into all sorts of circuits lately,  radio 
 RJT> circuits among them.  I'd like to know,  though,  how some stuff
 RJT> (scanners and other things that deal with a *lot* of frequencies)
 RJT> manage it.

 JB> AIUI the use a superheterodyne (I can spell it but not really
 JB> understand it)

Yes,  in contrast to one site in oz that's offering all sorts of tutorial
material but where he didn't quite spell it right.    I understand
it,  but the thing is,  even with double conversion,  you still want some
RF gain in front of the first mixer stage,  and some selectivity in there, 
and that's where I get hung up.  How the heck do they deal with that many
frequencies?  I guess you can switch coils in and out with diode switching
or whatever,  but that sure hits me as getting awfully complicated.  I
guess one of these days I'll have to get a hold of a scanner schematic
diagram and see how they do it.

 JB> basically what they do is generate a sine wave near the desired
 JB> frequency (say 60Khz below it) and use that to modulate in antenna
 JB> input this results in a prequency shift of the program signal, then
 JB> they use a fixed frequency detector/demodulator setup.

I'd say that 60 KHz is awfully close -- you'd have a problem with image
response.  The tradeoff there is the degree of selectivity you have in
front of the mixer,  the more selective your tuned circuits there,  the
closer the IF can be,  but I don't think I've ever seen anything as close
as 60 KHz.

 RJT> It's been quite some time since I read anything on this subject,  
 RJT> but I remember early gear that switched coils, even to the extent 
 RJT> of plugging different ones in,

 JB> old clunk-clunk VHF TV tuners....

I was thinking of ham stuff,  actually,  I have a 1967 ARRL handbook
somewhere and I've run into other stuff way back when that had switched
coils.  Come to think of it,  I have a grid-dip meter that uses that setup
too.

Most of those TV tuners that I've seen used coils in series.  Some of the
"coils" were actually only a single turn,  or a single loop of
wire,  or sometimes a stamped bit of sheet metal,  riveted in place on the
switch wafer, for dimensional stability I guess.  A few very early TV
tuners that I ran across used "strips",  little phenolic
assemblies with contacts on the outside and a complete set of tuned
circuits on the inside,  some of which even had adjustments.  But those
were pretty uncommon.  You could have some fun working on one of those by
switching some of the strips around,  and putting them in whatever order. 
:-)  They were still around when UHF first hit,  and one mfr. had an
"adapter" that required you to change out one of the strips from
an unused channel.

Heck,  most newer TV sets and VCRs even have a single tuner module that
goes across all of the TV bands.  I wonder how _they_ do it,  that's a
pretty wide spread...

 RJT> and I have my doubts that modern stuff does anything of the sort

 JB> more recent TV tuners use varicap diodes .... diodes sepcially
 JB> designed have a junction capacitance which varies depending on to
 JB> their bias voltage...

Actually most diodes will do that to some extent,  the ones you refer to
being particularly optimized in some way.  I've read that you could
actually use ordinary rectifier diodes for such a thing.

 JB> but ISTR they still use split the spectrum into 3 bands VL VH and
 JB> UHF

That's true,  come to think of it,  they do have those band select signals
in there,  I'd forgotten about that.

 RJT> I suspect that this is where things start getting less into
 RJT> ordinary "tech" stuff and more into that "black
art" portion of
 RJT> design that I've been avoiding all these years.  :-

 JB> it'd be real hard to design an RF circuit that'd work from 100Khz
 JB> all the way up to 1Ghz... ISTM most devices get at most factor of 3
 JB> betweens ends of each tunable band, that may be a limitation of
 JB> tuning devices or it may also involve limitations in the filters
 JB> employed in the receiver.

I think that the 3:1 limit in tuning is largely because of the 10:1 limit
in practical tuning capacitors.  I'm not sure how far you can go with any
given tuned circuit,  though,  before other factors come into play.  Things
like the Q of the tuned circuit,  the selectivity you get out of it. 
Fiddling with this sort of stuff also involves a lot of math,  something
I've not been terribly fond of,  and by the time I got my hands on a decent
calculator that had the required functions I'd lost interest in RF and
moved on to other areas instead.  :-)

---