Speaking of quadrature demods, wouldn't best performance be had by
using a 23.3ns ) wideband delay instead of a single or
double-tuned LC. I just found one with 30MHz BW...

Probably not, unless there's something I'm missing. You don't want a
wideband circuit here -- the quadrature detector takes advantage of
phase shift through the passband. Basically it uses a mixer to compare
a signal with a phase-shifted version of itself, with the output being
proportional to the actual phase difference. With a single-tuned
circuit, the 3 dB point corresponds to 45 degrees of phase shift, and
that's where you center the quadrature coil.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------

In article , John
Miles wrote:

Speaking of quadrature demods, wouldn't best performance be had by
using a 23.3ns ) wideband delay instead of a single or
double-tuned LC. I just found one with 30MHz BW...

Probably not, unless there's something I'm missing. You don't want a
wideband circuit here -- the quadrature detector takes advantage of
phase shift through the passband. Basically it uses a mixer to compare
a signal with a phase-shifted version of itself, with the output being
proportional to the actual phase difference. With a single-tuned
circuit, the 3 dB point corresponds to 45 degrees of phase shift, and
that's where you center the quadrature coil.

Phase shift = delay for a given frequency of course. 90deg at
10.7MHz is 23.3ns. Being wideband means the delay and amplitude change
over a 150kHz BW are MUCH less than with a single LC. Seems to me that
would give the best performance and lowest distortion...

Michael

"Michael Dunn" wrote in message
.. .
In article , Doug Smith W9WI
wrote:
What is your reception goal for this receiver?

- High fidelity?

- High sensitivity?

- Adjacent-channel rejection?

- Overload tolerance?


Pretty much all of the above. Cost is secondary. Though I do like
simple elegance, and I'm not going to go as far as the SUPRX mentioned
by Bob, though it must be a fine rcvr.

Mark mentioned:
could design not only AFC, but also automatic tuning of an RF
preselector and perhaps even an antenna tuner, that would be tres
cool.

All of my older FM tuners have a three-gang "bread slicer", A.K.A. variable
capacitor. It tunes (1) the LO, (2) mixer input, (3) RF stage input. Since
the OP wasn't interested in a synthesized LO, a multigang capacitor would
seem to make sense.

With some imagination, one could arrange a scheme so that the LO would lock
to
a synthesized signal every 200 kHz step. Thus one-knob tuning of multiple
stages, yet with synthesizer stability. Details are left as a student
exercise.

73 de bob w3otc

"R J Carpenter" wrote in message ...

All of my older FM tuners have a three-gang "bread slicer", A.K.A. variable
capacitor. It tunes (1) the LO, (2) mixer input, (3) RF stage input. Since
the OP wasn't interested in a synthesized LO, a multigang capacitor would
seem to make sense.

With some imagination, one could arrange a scheme so that the LO would lock
to
a synthesized signal every 200 kHz step. Thus one-knob tuning of multiple
stages, yet with synthesizer stability. Details are left as a student
exercise.

73 de bob w3otc

You'd also probably get higher Q with air var. cap.
But everything else is more complicated with this approach

SioL

Michael Dunn wrote:

Hello fellow radionuts (radionauts?). For some reason, I've decided
to design and build an FM tuner. Just to see if I can make a good one
I guess. A few questions have presented themselves:

-PLL or quadrature (IC) decoding?

-Stereo decoding: single IC or "discrete"?

-Throw in SCA! Why not? PLL?

-Anyone in S Ontario (KW) area?

-I'm still having a hard time figuring out AFC. I can see how I might
do it with a PLL - the DC error voltage being used to pull the LO to
the correct frequency. How was it done in the old days? ;-) If I
could design not only AFC, but also automatic tuning of an RF
preselector and perhaps even an antenna tuner, that would be tres cool.

All thoughts greatly appreciated.

Michael Dunn
VA3SSP
mdunn @at@ cantares.on.ca
Tech museums: http://www.cantares.on.ca/museums.htm
Design portfolio: http://www.cantares.on.ca/portfolio/

Here's a design I used quite successfully:
www.web-ee.com/Schematics/FM_RCVR/FM_RCVR.pdf

Bob
K4QQK

In article ,
Michael Dunn wrote:
Hello fellow radionuts (radionauts?). For some reason, I've decided
to design and build an FM tuner. Just to see if I can make a good one
I guess. A few questions have presented themselves:

-PLL or quadrature (IC) decoding?

Quadrature. Probably double tuned. (One of the old, now obselete,
demodulator IC's had a good description on double tuned quad coils,
but damned if I can remember which one. I searched for it, as there
was a newsgroup discussion of them going on at the time).

PLLs are feedback loops. Feedback loops are driven by error. In this
case error would show up as distortion.

You could also mix down to a lower IF than 10.7 and use a frequency to
voltage converter. (A precision one shot triggered on each cycle).

-Stereo decoding: single IC or "discrete"?

-Throw in SCA! Why not? PLL?

Narrowband IF ICs like the MC3361 make a good SCA demodulator. Basically
a superhet using the onboard oscillator and mixer that runs a 455 kHz
IF and a quadrature demodulator.

I did one with the MC3359, and I think it's on my ftp pages
at ftp://ftp.eskimo.com/u/m/mzenier .

-Anyone in S Ontario (KW) area?


-I'm still having a hard time figuring out AFC. I can see how I might
do it with a PLL - the DC error voltage being used to pull the LO to
the correct frequency. How was it done in the old days? ;-) If I

Just about all FM detectors (ratio det, discriminator, quad)
output down to DC (from the offset from mistuning) , which can
then be integrated/lowpassed to create the AFC voltage. On
frequency may be at 1/2 the power voltage for a quad detector, not
zero volts.

could design not only AFC, but also automatic tuning of an RF
preselector and perhaps even an antenna tuner, that would be tres cool.

Signal strength is a different signal from AFC. AFC comes after the
signal has been limited, so that's sort of information is lost.

Random comment: Elektor Electronics magazine, the european project
magazine, had a fancy FM tuner project about a half a dozen years ago.

Mark Zenier Washington State resident

In article ,
says...

PLLs are feedback loops. Feedback loops are driven by error. In this
case error would show up as distortion.


Well, that's a pretty extreme oversimplification.

Error above the loop bandwidth shows up as distortion. To use a PLL to
recover FM, you use a wide bandwidth that covers the entire baseband
spectrum of interest. That way, the error *is* the signal. Very
standard technique.

-- jm

------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------