On Sun, 25 Aug 2013, Percy Picacity wrote:

In article ,
"gareth" wrote:
snip

You're still missing the point that in addition to the peak response, there
is
also a deep null.

No I'm not! It can be adjusted with the 'phase' control to null a signal
*at IF* near to the wanted one. Adjusting the position of the null has
no affect on beat frequency with the wanted signal, or the beat
frequency of the unwanted signal (it gives the BFO a less strong IF
interfering signal to beat with but it does not affect the frequency of
the beat note, just the loudness). Tuning the BFO has no effect on the
null. The two controls do not interact, though they both have an affect
on readability.

I dug up an early article by Lamb about the filter (not the QST article
but some other publication). And there doesn't even have to be a notch.
Ajust the control a certain way and there's no notch, it's just a very
narrow filter.

The notch is just iciing on the cake, the filter was there to get a narrow
enough bandwidth so the audio image isn't there. There were some
construciton articles in the sixties in various magazines for adding cw
selectivity to SSB transceivers, which of course at the time often had
only an SSB suitable IF filter. And one scheme was to gang a few of those
phasing type filters, the ganging narrowed the skirt. SO they'd use
triodes, the crystal from the plate of one to the grid of the next, the
phasing capacitor from the cathode of one to the grid of the next, the
triode acting as a phase inverter instead of the transformer. And while
there were trimmer capacitors in each section so they could all be
aligned, no phasing control was brought to the front panel.

I said I never used the phasing control on the Sp-600, and one of these
days when I get my $20 at a garage sale TMC GPR-90 going (I don't think it
needs much work, I just need to get around to it), I doubt I'll use the
phasing control on it. The description of such filters always sounded to
me like the notch ability wasn't so useful, since it interacted with the
peaking of the actual crystal filter. It's not like having a separate
notch filter to wipe out offending interference. Circa 1936, the phasing
control probably helped a lot, all the receivers fairly simple and nobody
wanting to make things complicated in circuit or price, and of course the
bands weren't as crowded. But nowadays, it is something from the 1930s.
A great thing when you need a simple crystal filter, or to start with to
get the receiver going (and then replace with a better filter), but there
are better schemes out there already.

Michael VE2BVW