Re: OPT equalising circuit

In article 
<1eafb781-0b1f-49ee-b4d7-fe74736d7b5c@[EMAIL PROTECTED]
>,
 Alex <apogosso@[EMAIL PROTECTED]
> wrote:

> On May 6, 7:17 am, John Byrns <byr...@[EMAIL PROTECTED]
> wrote:
> > In article <0BATj.7774$ko5.7...@[EMAIL PROTECTED]
>,
> >
> >  "Alex" <apogo...@[EMAIL PROTECTED]
> wrote:
> > > On the stereo-decoder note, what is the phasing of 19KHz pilot with 
> > > respect
> > > to the 38KHz (suppressed) carrier: do the zero-crossings coinside
(every
> > > second one) or peaks coinside (every secon ones)?
> >
> > > I would guess, the peaks. That way it would have been easier to
derive 
> > > 38KHz
> > > simply by a class C frequency multiplier.
> >
> > Your guess is incorrect, the 19 kHz pilot tone and the regenerated 38
> > kHz carrier must be phased so that the zero-crossings coincide, that
is
> > the way the original system specification was written.  If the decoder
> > is adjusted so that the peaks coincide it will be impossible to
achieve
> > any stereo separation at all.  If the phasing is such that the
> > zero-crossings do not coincide, but are close, it is possible to
achieve
> > full separation by adjustment of the matrix separation control, but
the
> > signal to noise ratio will be degraded.
> >
> > Regards,
> >
> > John Byrns
> >
> > --
> > Surf my web pages at,  http://fmamradios.com/
> 
> I see, John.
> Thanks for the info.
> 
> If it is so that the sero-crossings coincide then it better suits a
> PLL-based multiplier. Probably the early designers had that method in
> mind, not the harmonic frequency multiplier...

I don't know, it could easily be that the early designers had the 
harmonic frequency multiplier method in mind.  Consider the following 
argument.  First IIRC there is a 90 degree phase ****ft between the 
primary and secondary of a double tuned transformer, my whole argument 
depends on that fact so please correct me if I have that point wrong.  
Second a 90 degree phase ****ft at 19 kHz translates into a 180 degree 
phase ****ft at 38 kHz, which simply results in a reversal of the left 
and right channels, while a 90 degree phase ****ft at 38 kHz results in 
no recovery of the stereo difference signal and hence no stereo 
separation.

Now as long as the phase ****ft in the 19 kHz filters and amplifier is a 
multiple of 90 degrees at 19 kHz the phase ****ft at 38 kHz will be a 
multiple of 180 degrees at 38 kHz, whose only effect will be to reverse 
the two channels.  Let's assume that we have a single tuned 19 kHz 
filter preceding the grid of a 19 kHz amplifier tube and a double tuned 
transformer with a center tapped secondary winding connected to the 
plate circuit of the 19 kHz amplifier tube, providing us with a multiple 
of 90 degrees of phase ****ft at 19 kHz.  If we ground the center tap of 
the 19 kHz transformer secondary and connect a diode from each end of 
the secondary to the grid of a 38 kHz amplifier tube we will have a 
frequency doubler to generate the local 38 kHz carrier needed to 
demodulate the stereo difference signal.  This type of frequency doubler 
will cause a 90 degree phase ****ft at 38 kHz because the peaks of the 
resulting 38 kHz signal will be aligned with the peaks of the 19 kHz 
pilot, rather than having the zero crossings aligned as they should be.  
But we aren't done yet, the 38 kHz signal out of the diode frequency 
doubler is distorted and contains harmonics of 38 kHz.  If we connect 
another double tuned transformer to the plate circuit of the 38 kHz 
amplifier, it will remove the distortion on the 38 kHz waveform, and at 
the same time introduce another 90 degree phase ****ft at 38 kHz 
effectively canceling the effect of the 90 degree phase ****ft introduced 
by the doubler circuit.  The secondary of the 38 kHz transformer then 
drives the stereo demodulator or switching diodes.

So it appears to me that the original designers of the pilot tone stereo 
system had LC type decoders in mind, not PLL decoders.

Another factor might be that the original designers wanted to keep the 
amount of 19 kHz leaking into both stereo channels the same.  I haven't 
done the math yet, but it isn't obvious to me that this condition would 
accrue if the signal specified aligned peaks rather than aligned zero 
crossings.  I will have to check to see if this is actually the case.

> In case of Patrick
> Turner it indeed makes no difference, because he fiddles with at least
> two LC tanks to get the phase right. In his circuit the phase
> adjustment is so broad, that he can possibly swap left and right
> channels if he wanted to.

I haven't looked at Patrick's stereo decoder in a while, I will have to 
take a look and refresh my memory before commenting.  Some of Patrick's 
designs do tend to be a little Baroque.

I just took a very quick look at Patrick's tuner design as I am posting 
this, and the first thing I noticed is that the Ratio Detector won't 
work as drawn, but presumably it works just fine a built.  The stereo 
decoder design appears to be much more straight forward than his first 
design, which was quite unusual and likely didn't have very good 
performance.


Regards,

John Byrns

-- 
Surf my web pages at,  http://fmamradios.com/