Today the si570 magic oscillator seems to satisfy all our VFO requirements
in terms
of stability, frequency range and frequency increments, but will it always
be there,
for don't forget that the Plessey SL600 series was the answer to a maiden's
prayer in its time
but now is no longer available?

"gareth" wrote in message
...
Today the si570 magic oscillator seems to satisfy all our VFO requirements
in terms
of stability, frequency range and frequency increments, but will it always
be there,
for don't forget that the Plessey SL600 series was the answer to a
maiden's prayer in its time
but now is no longer available?


All things become obsolete eventually, old Bean. It's called progress.
--
;-)
..
73 de Frank Turner-Smith G3VKI - mine's a pint.
..
http://turner-smith.uk

On Sun, 22 Nov 2015, gareth wrote:

Today the si570 magic oscillator seems to satisfy all our VFO requirements
in terms
of stability, frequency range and frequency increments, but will it always
be there,
for don't forget that the Plessey SL600 series was the answer to a maiden's
prayer in its time
but now is no longer available?

Probably not. But I say that because this is still early for that sort of
IC, that in the future there's likely going to be lots of improvement.

Those Plessey ICs were neat, but they had limited use. They weren't
intended for consumer use, but for "communication receiver" use. That's a
fairly small market. I have no idea how long they lasted, but they
probably did better than one should expect.

COntrast that with the MC1350P, which was an 8pin gain stage intended for
IF amplifiers and which included a really good gain control stage. I
gather it's no longer being made, but it lasted decades. I even did once
see it in a car radio, and I can imagine it going into TV sets. It was
broad enough building block that it probably found enough use.

The more specialized something is, the more limited the use, and thus it
probably won't last long. So logic ICs have remained over decades (or at
least morphed to other families while keeping the same functions, but
those Signetic analog PLLs from the early seventies are long gone. They
had a use, especially right at that time, but later their value diminished
because other things came along to do the same function, but better.

An IF amplifier can survive, an FM detector (like the old RCA CA3089) can
survive because it can be used in consumer electronics. There were lots
of tv-specific ICs that came and went, depending on whether they were
actually used much in tv sets.

Communication ICs have a relatively small market, and they also suffer
because new devices (and concepts) come along to replace them. There were
a slew of ICs intended for AM broadcast radios, but some were quite
limited in how much else they could be used for, while later generations
combined some or all of the FM broadcast radio into the same IC,
obsoleting the old. There were also a few iterations, so that RCA CA3088
from 1971 was neat when it came out, but saw limited use, while later ICs
of the same function got more travel.

I forget who made it, but there is or was an IC intended for car radio use
that actually broke things dramatically, for the AM band the receiver
upconverted (so less image problems) to an IF higher than the AM broadcast
band. I have no idea how much use that saw, or if it's still available.
AM radios have generally never seen innovation, everyone content to use
single conversion to 455KHz.

Around 1970, National Semiconductors issued the LM372, some stages of gain
and an AM detector, intended for TRF use. I remember one article about
it, and maybe have one around somewhere, but it really went nowhere, I
have no idea how long they kept making it.

But somewhat later, Feranti came out with the ZN414, a similar device
except only 3 pins. That was popular with home builders, but I'm unsure
where they thought the IC could be used commercially. Somehow it kept
going, and today someone makes a descendant, the MK484. I think they may
be showing up in dollar store radios, but that's only a guess. Even circa
1970, nobody was making TRF AM band receivers, except the hobbyist.

National at the same time also issued the LM373, a section of gain with
controllable gain, and a detector stage that could be used as an AM/FM/SSB
detector. Lots of potential, that got some articles, but I'm not sure if
it was ever used in a piece of commercial equipment. Hobbyists can't keep
such neat devices alive.

I don't think National still exists, that factors in too. Motorola spun
off its semiconductor division, and that weeded out some neat devices.

But things change. Precisely because of ICs, things can be made denser,
and thus new techniques become viable.

There was the TDA7000, and FM broadcast receiver that used an odd
detector, but also converted down to a low (I think 70KHz) IF, using
active filters for the IF selectivity. But so long as it was used on the
FM broadcast band, the images would fall between stations, so it didn't
matter that image rejection wasn't very good. In fact, that concept
derives from a 1947 scheme, from GE, converting the FM broadcast band to a
200KHz IF and a pulse counting detector, which was later resurrected by GE
in the early sixties when they showed a semiconductor receiver of the same
type, using a tunnel diode as the mixer/oscillator.

I'm not sure what's used in my Sansa Fuze MP3 player, but it seems to be a
decent FM broadcast band receiver. But it's too small for a superhet with
IF transformers in there, and I can't imagine them adding a whole FM
receiver unless it added very little to cost.

Now, you can get ICs that convert down to a low IF, and then convert to
digital and process there. SOme of it is controllable, but they are
seeing use in portable shortwave receivers, nobody could imagine this sort
of thing 40 years ago. So fewer external components, even compared to
analog ICs like the Plessey from forty years ago. They apparently offer
good selectivity, and various bandwidths (though what's offered depends on
the implementation in a given receiver), something often not seen in cheap
receivers because extra filters cost money (and space).

I remember 20 years ago seeing some of this sort of stuff, but it was in
ICs designed for cellphones. It wasn't clear from what material I did get
how flexible they were if you needed them for other things.

But these now wipe out most previous work, why build up with discrete
components and analog filters when it can be done in one IC? It's not
perfect, since the cheap ones aren't completely flexible. I have a
Grundig G8 I got for curiosity sake, if I modified it and added my own
computer controller I think I can get better IF selectivity, but there's
absolutely no way to use it for SSB except by injecting a signal at the
signal frequency.

So I guess you go back to transistors, albeit by emulating the ICs using
multiple transistors. In the early days of solid state, people got by
with as few transistors as possible, then ICs showed you got better
results with more transistors. They are now a lot less expensive, so
while you have to wire more up, you can get better gain stages by using a
balanced modulator of some sort, and those balanced modulators can be had
using transistors. Then you get "the latest techniques", well one step
behind digital processing, without worrying about not being able to get
that neat IC in the future.

Michael