Wideband receiver architectures
 

Just curious... does anyone know what sort of architecture you find in your
typical HT or mobile "receives everything from 100kHz - 1.3GHz!" radio to
generate such a vast range of frequencies while still keeping the prices so
low (plenty are available for $200, and I can't expect the LO makes up a
particularly large percentage of that total price)? I would tend to lean
towards something PLL-based given the typically channelized nature of
available frequencies, but a single PLL would still need a VCO with a huge
tuning range. So... perhaps a pair of mixed PLLs, with one doing the fine
tuning and another the coarse tuning (having been multiplied up from a low
frequency)?

I attempted to decipher the schematic included with my Yaesu FTM-10R, but
everything is so tiny it was rather hopeless!

I don't suppose any of the reviews ever go through and sweep the LOs while
watching for the worst case spur conditions, do they?

---Joel

Wideband receiver architectures
 

Joel Koltner kirjoitti:
Just curious... does anyone know what sort of architecture you find in your
typical HT or mobile "receives everything from 100kHz - 1.3GHz!" radio to
generate such a vast range of frequencies while still keeping the prices so
low (plenty are available for $200, and I can't expect the LO makes up a
particularly large percentage of that total price)? I would tend to lean
towards something PLL-based given the typically channelized nature of
available frequencies, but a single PLL would still need a VCO with a huge
tuning range. So... perhaps a pair of mixed PLLs, with one doing the fine
tuning and another the coarse tuning (having been multiplied up from a low
frequency)?

I attempted to decipher the schematic included with my Yaesu FTM-10R, but
everything is so tiny it was rather hopeless!

I don't suppose any of the reviews ever go through and sweep the LOs while
watching for the worst case spur conditions, do they?

---Joel


Solution for LO maybe is DDS.
The practical solution of DDS is is extremely simple and cheap.
Please google word "DDS"

-- postipoika

Wideband receiver architectures
 

Joel Koltner wrote:
Just curious... does anyone know what sort of architecture you find in your
typical HT or mobile "receives everything from 100kHz - 1.3GHz!" radio to
generate such a vast range of frequencies while still keeping the prices so
low (plenty are available for $200, and I can't expect the LO makes up a
particularly large percentage of that total price)? I would tend to lean
towards something PLL-based given the typically channelized nature of
available frequencies, but a single PLL would still need a VCO with a huge
tuning range. So... perhaps a pair of mixed PLLs, with one doing the fine
tuning and another the coarse tuning (having been multiplied up from a low
frequency)?

I attempted to decipher the schematic included with my Yaesu FTM-10R, but
everything is so tiny it was rather hopeless!

I don't suppose any of the reviews ever go through and sweep the LOs while
watching for the worst case spur conditions, do they?

---Joel


I suspect that they first upconvert to somewhat higher than the 1.3GHz
upper frequency limit, then use a roofing filter with however many itty
bitty IFs it takes to get back down to a reasonable bandpass filter.

I also suspect that the receiver receives everything kinda well, not
superlatively.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" gives you just what it says.
See details at http://www.wescottdesign.com/actfes/actfes.html

Wideband receiver architectures
 

"Postipoika" wrote in message
...
Solution for LO maybe is DDS.

That's a big "maybe" -- even the fastest DDSes aren't much better than 1GSps
or thereabouts, giving a realistic tuning range of 0-400MHz... and those parts
are quite power hungry and not at all cheap (e.g., Analog Devices' AD9858 is
several watts and nearly $50 even in quantities of 1000!). On the other hand,
one of the much lower frequency DDSes in conjunction with a mixer would
probably work well -- although it just shifts the problem to how to get nice
and pure, very wide range fixed frequencies to mix with. (E.g., DDS does
0-200MHz, so now you need just, say, a 200MHz+/-IF, 400MHz+/-IF, 600MHz+/-IF,
etc. oscillators to mix with...)

Of course with the very fine-grained frequency resolution of a DDS,
multiplying the output up would seem to be an obvious approach, but I'm told
this isn't typically used due to the relatively high spur levels of the DDS
itself.

---Joel

Wideband receiver architectures
 

Joel Koltner wrote:
Just curious... does anyone know what sort of architecture you find in your
typical HT or mobile "receives everything from 100kHz - 1.3GHz!" radio to
generate such a vast range of frequencies while still keeping the prices so
low (plenty are available for $200, and I can't expect the LO makes up a
particularly large percentage of that total price)? I would tend to lean
towards something PLL-based given the typically channelized nature of
available frequencies, but a single PLL would still need a VCO with a huge
tuning range. So... perhaps a pair of mixed PLLs, with one doing the fine
tuning and another the coarse tuning (having been multiplied up from a low
frequency)?

I attempted to decipher the schematic included with my Yaesu FTM-10R, but
everything is so tiny it was rather hopeless!

I don't suppose any of the reviews ever go through and sweep the LOs while
watching for the worst case spur conditions, do they?

The HT-class wideband receivers typically will have a first IF in the
45 or 70 MHz range, and use a number of different PLL's one for each
different band. Each PLL is literally less than a square cm on the
PCB, and there's a lot of them.

At least one Sony from the mid 90's had a 10.7MHz second IF, and if
the frequency you wanted was anywhere near the first IF of 70MHz, then
they skipped the first IF and went straight to the second IF.

For FM the 10.7MHz IF is the ending point... for non-WBFM modes there
was a third IF of 455kc with some ceramic filters.

Some of the HT's have 5kHz channel spacing on the HF band with no
finer tuning available (OK for SWLing but nothing else), but the
slightly better ones have finer tuning in the HF range thanks to a the
2nd or 3rd LO being a PLL too.

Image rejection is nearly completely nonexistent.

Tim.

Wideband receiver architectures
 

In article ,
Joel Koltner wrote:
one of the much lower frequency DDSes in conjunction with a mixer would
probably work well -- although it just shifts the problem to how to get nice
and pure, very wide range fixed frequencies to mix with. (E.g., DDS does
0-200MHz, so now you need just, say, a 200MHz+/-IF, 400MHz+/-IF, 600MHz+/-IF,
etc. oscillators to mix with...)

From reading the occasional data sheet and appnote, I have the impression
that (as Tim Wescott says) they upconvert to a high IF, like 3 GHz,
and then convert down from there.

In that case they might have an LO that tunes from 1.7-3 GHz, which is
a much smaller range, proportionally, than 100kHz-1.3GHz.

--
Wim Lewis , Seattle, WA, USA. PGP keyID 27F772C1

Wideband receiver architectures
 

On 16 Jun, 22:22, Postipoika
wrote:
Joel Koltner kirjoitti:

Just curious... does anyone know what sort of architecture you find in your
typical HT or mobile "receives everything from 100kHz - 1.3GHz!" radio to
generate such a vast range of frequencies while still keeping the prices so
low (plenty are available for $200, and I can't expect the LO makes up a
particularly large percentage of that total price)? *I would tend to lean
towards something PLL-based given the typically channelized nature of
available frequencies, but a single PLL would still need a VCO with a huge
tuning range. *So... perhaps a pair of mixed PLLs, with one doing the fine
tuning and another the coarse tuning (having been multiplied up from a low
frequency)?

I attempted to decipher the schematic included with my Yaesu FTM-10R, but
everything is so tiny it was rather hopeless!

I don't suppose any of the reviews ever go through and sweep the LOs while
watching for the worst case spur conditions, do they?

---Joel

Solution for LO maybe is DDS.
The practical solution of DDS is is extremely simple and cheap.
Please google word "DDS"

-- postipoika

My Yaesu FT-857D and FT-817ND transceivers use a PLL and DDS. The PLL
cleans up the DDS output and the DDS gives the fine tuning.

The 857 receives up to 450 MHz, and uses a 68.33 MHz first IF.

Leon

Wideband receiver architectures
 

Andy writes:

Look in the tech manual for a spectrum analyzer that covers a few
Khz to 1.3 Ghz and you will see exactly how it's done.

Typically, the LO runs above 1.3 Ghz and all signals are converted
UP to a high IF where the first filter is a wideband SAW. A SAW can
have several Mhz bandwidth way up there.

Then the signal is downconverted to an IF that is compatible with
the SAW bandwidth to achieve image rejection. The IF filter here
is usually something like a crystal or ceramic filter, unless the
requirements
for instantaneous bandwidth forces a better group delay response.

The hardest part is keeping the signal linear ( intermod,
crossmod)
in the presence of multiple simultaneous signals until the SAW and
next IF filters are reached.

Lots of tradeoffs have to be made.

These tradeoffs have already been made in spectrum analyzers, so
if you get a few different block diagrams of different units, you
will
see what specific frequency plans the designers have decided to use.

The frequency plan of the cheapest spectrum analyzer would
probably fit your purpose....

Coming up with a good frequency plan in view of the nonlinearities
of the mixers used and the bandpass characteristics of the filters
one employs is a real challenge sometimes.

Andy in Eureka, Texas

Wideband receiver architectures
 

Thanks Andy, that was most helpful.

---Joel

Wideband receiver architectures
 

On Jun 16, 5:05 pm, "Joel Koltner"
wrote:
Just curious... does anyone know what sort of architecture you find in your
typical HT or mobile "receives everything from 100kHz - 1.3GHz!" radio to
generate such a vast range of frequencies while still keeping the prices so
low (plenty are available for $200, and I can't expect the LO makes up a
particularly large percentage of that total price)? I would tend to lean
towards something PLL-based given the typically channelized nature of
available frequencies, but a single PLL would still need a VCO with a huge
tuning range. So... perhaps a pair of mixed PLLs, with one doing the fine
tuning and another the coarse tuning (having been multiplied up from a low
frequency)?

I attempted to decipher the schematic included with my Yaesu FTM-10R, but
everything is so tiny it was rather hopeless!

I don't suppose any of the reviews ever go through and sweep the LOs while
watching for the worst case spur conditions, do they?

---Joel

The one I had used triple conversion superhet. Mircoprocessor
controlled.
The first IF was like 336Mhz to 10.7Mhz to 455Khz. There was no front
end. just direct conversion to the IF. So all the micro had to do was
control the oscillator from 336Mhz to 900Mhz using up conversion on
the low bands and down conversion on the higher bands. Cell phone band
was locked out. Guess what you can pick cell phone band on the image
frequency back in the day.

73

N8ZU

Wideband receiver architectures
 

The one I had used triple conversion superhet. Mircoprocessor
controlled.
The first IF was like 336Mhz to 10.7Mhz to 455Khz. There was no front
end. just direct conversion to the IF. So all the micro had to do was
control the oscillator from 336Mhz to 900Mhz using up conversion on
the low bands and down conversion on the higher bands. Cell phone band
was locked out. Guess what you can pick cell phone band on the image
frequency back in the day.
========
Are analog cellphones still used in the USA ?
I thought by now it is all digital sigs which can not be received with a
'normal' receiver , so what's the reason to 'lock out' the cellphone
frequencies ?

Frank GM0CSZ / KN6WH

Wideband receiver architectures
 

wrote in message
...
Guess what you can pick cell phone band on the image
frequency back in the day.

Given how much time was spent discussing the need for image rejection and the
sometimes-elaborate techniques for achieving it, I suspect a lot of my college
professors would be incredulous to learn that many receivers do just fine
without it.

I've noticed that many Yaesu radios lock out the images of the old AMPS band,
which sucks because the image is where there are also 100%-legal-to-receive
paging bands. Yaesu lost a sale to Icom over this.

Hopefully the stupid law will be dropped one of these days -- it's not like
any large percentage of all cell phone conversations are carried out on an
AMPS anymore, and most of the cell phone carriers are quietly shutting down
their old AMPS equipment altogether.

---Joel

Wideband receiver architectures
 

On Mon, 07 Jul 2008 06:08:21 +0100, Highland Ham wrote:

The one I had used triple conversion superhet. Mircoprocessor
controlled.
The first IF was like 336Mhz to 10.7Mhz to 455Khz. There was no front
end. just direct conversion to the IF. So all the micro had to do was
control the oscillator from 336Mhz to 900Mhz using up conversion on
the low bands and down conversion on the higher bands. Cell phone band
was locked out. Guess what you can pick cell phone band on the image
frequency back in the day.
========
Are analog cellphones still used in the USA ?
I thought by now it is all digital sigs which can not be received with a
'normal' receiver , so what's the reason to 'lock out' the cellphone
frequencies ?

Frank GM0CSZ / KN6WH

The same government logic that made TV broadcast standards
backwards-compatible for over 50 years.

__
Gregg