[IRCA] Importance of IF skirts - was - Winradio G313-E

["Bob Foxworth" <rfoxwor1@xxxxxxxxxxxxxxx>]

John Bryant writes a fascinating piece, with which I can't find
any fault. I'd like to add a minor historical footnote, below...

> Like Brandon, I've been using the WiNRADiO 313E exclusively for DXing
> for the past year, plus.
>         /* snip  */
> Lastly, an attribute of most serious DSP receivers.... virtually
> vertical sidewalls on the IF envelope.  With any analog IF filters...
> even the vaunted Collins mechanical filters, the acutal window that
> you are receiving the signal through is shaped like a truncated
> equilateral triangle.  Where your main DXing filter may be 2.7 kHz
> wide at 3 dB down, it may be 15 or more wide at 30 dB down.  The
> result of that is that you use narrower filters that you really want
> to, to conquer adjacent channel QRM..... AND since you have touse a
> wide enough filter to let intelligible audio in, you also allow Loud
> adjacent splatter, etc. in "under the filter skirts."  With a total
> DSP package, from demodulation on down, the IF filtering, of course,
> is done in the software.... hence, filter skirts that are pretty-much
> VERTICAL.


I worked at ITT for a while in the 1980's following my broadcasting
experiences, until ITT got out of the HF business. So all this
is first-hand knowledge. Although the HF circuits were mostly
all gone, as were the rhombics. Only six log-periodics were left.

ITT WorldComm used to run a lot of HF P-T-P links across the
North Atlantic in the 1960-ish time frame. Many of them were
voice, or RTTY in a voice-grade channel, and were ISB (this
was independent sideband, with totally unrelated programming in
the LSB and USB channels). A lot of this was actually done by
Mackay Radio, prior to being acquired by ITT in it's gobble-em-up
days. They had 10 kW xmitters at Hauppauge, and receivers
at Southampton, 40 miles to the east, all using rhombics. (Their
competitor RCA operated from Rocky Point and Riverhead,
in similar fashion). WSL 500 kc/s was part of this operation.

They achieved this ISB by using exciters on the transmit side with
paired Collins mech. filters at 455 kHz to generate separate
sideband signals. The filters were paired so that the response
curves intersected at about the -6 dB point. I recall they were
2.1 and/or 3.1 kHz wide. This composite RF envelope was
heterodyned (not multiplied)  up to the HF exciter frequency,
and contained both LSB and USB energy.

What is perhaps more interesting is what they did on the receive
side. They used modified SP-600 receivers with provision for
phase locking the HFO to a stable local oscillator reference
that was created by mixing a roughly 2 MHz VFO with a roughly
2.5 to 4 MHz crystal reference and multiplying it up to the needed
HFO freq, which was anywhere from about 4 to 22 MHz, and
sending it to the first mixer. The SP-600 itself then needed to
be tuned only close enough to peak the ant and mixer RF circuits
and to let the HFO lock onto the substitute HFO. This allowed
stably "tuning" the rcvr to exact cycles. The 2 MHz was in an oven.
This was not trivial to set up properly.

The 455 kHz IF was treated similarly by adding a new, outboard
IF strip using two custom made L-C filters with bandpasses of
23 to 25, and 25 to 27 kHz. The SP-600 was also modded to
use an added xtal osc of either 430 or 480 kHz (upright/inverted
sidebands) to mix the 455 IF down to 25. These custom 25 kHz
filters reportedly had very steep skirts, which were not obtainable
at 455 kHz. And the independent sideband energy could be
cleanly picked off and sent to two detectors. It was up to the
receiver side to separate the sidebands as well as possible.

In addition frequently two SP-600s ran in space-diversity mode
being fed from two side by side rhombics. Both ITT and RCA
had dozens of them, many pointed across Europe. Mackay/ITT
built combiners allowing the AGC to be slaved in both RX's,
combining the audio demod, among other things.

I still have a in-house manual for the CNY-2061-D which is the
ITT designation for the modified SP-600 which describes
the mod. process and alignment for the SP-600. Much of
this material was lost in November 1986 when the developer
Racanelli razed the site (for an industrial park)* after ITT sold
and walked away from it, after the end of all HF ops. I used
to have an example of the 25 kHz IF as well as some other
stuff I eventually wanted to give to a museum but no one
I could find wanted it, and I got rid of it in 1990 when I sold
my house and married. There was no eBay then...

*Adams Ave, Hauppauge, LI NY  Rumor has it that, if you
dig deeply enough under some office buildings there, you
might find a dozen old SP-600's, vacuum caps, etc etc
RBL's that were 'dozed under.  Just a rumor, mind you.

The inability to get truly vertical skirts in IF was one of the
main things Gordon Nelson always wished he could remedy.

It is truly important for weak signal work, as John correctly
notes.

- Bob



_______________________________________________
IRCA mailing list
IRCA@xxxxxxxxxxxxxxxx
http://arizona.hard-core-dx.com/mailman/listinfo/irca

Opinions expressed in messages on this mailing list are those of the original contributors and do not necessarily reflect the opinion of the IRCA, its editors, publishing staff, or officers

For more information: http://www.ircaonline.org

To Post a message: irca@xxxxxxxxxxxxxxxx