On Wed, 10 Jun 1998, Nolan Lee wrote:

> At 04:56 PM 6/10/98 -0400, you wrote:
> 
> >The SE3 and critters like it phase-lock the reinserted carrier to the
> >station's carrier. That's a BIG difference, because the reinserted 
> >carrier acts to re-
> >enforce the station's carrier and thereby you achieve an increase in intel-
> >ligibility of the recovered station's audio (NOT necessarily fidelity, altho
> >that's frequently much improved, too.).
> 
> Bear with me on this, I'm a bit rusty on theory and I'm a barefoot
> redneck, so type slow. <grin> I always thought that reinserting the
> carrier to the product detector was only used with SSB reception, and
> not AM. Does the SE3 only work on SSB? I musta' missed something
> somewhere. I'm assuming that the SE3 is tapping the final IF of the
> receiver and feeding it thru it's own product detector and audio
> circutry. Did I go astray here? 

In selective fading, the different frequency components across
the 6 kHz or so of AM transmitted bandwidth are receiving different 
amounts of attenuation, because of different paths through the
ionosphere, etc. With a standard diode detector for AM, 
one way to look at it is that all the freqs present in the IF passband
are mixing with the strongest one present, which should be the carrier.
If the carrier and part of one sideband has faded, the remaining
individual freq components have to beat with whatever audio freq
is now the strongest, giving the wrong recovered audio freqs.
With a synchronous detector, the circuitry phase locks to the
proper freq and phase the carrier should be, even if it drops
out at times. Thus there's always a proper carrier to receive
audio with. This can be done by actually locking to the carrier,
with a slow time constant PLL loop which can hold a constant carrier
freq an phase relationship even during long fades, or by seperating
the carrier out from the combined AM signal with a sharp filter,
and amplifying it seperately, so that it remains strong even during 20 db 
carrier fades, (CV-157 does this) or by the use of a Costas loop, the 
freq and phase can be recovered from the USB and LSB audio signal even
with DSB transmitted. ( article by Jack Webb, Dec '57 CQ mag is
a fine hollow state sync detector of the later type, if I've 
recalled the date properly)

Depending on the type of sync detector, it could be used on AM, 
DSB, or SSB reduced carrier, the modulation system internationas SW
broadcasters are said to be using in the future. Plain SSB, it
won't do anything for.

> The carrier reinsertion in my primary radios is stable to within
> something like 1 part in 10 to the 8th power with all kinds of
> fancy PLL error correcting stuff and such. Granted, I can't do
> anything about the received station's carrier drifting, but if I
> tune to WWV or any other station using the AM mode and switch the
> mode switch to LSB USB or ISB, there is no typical SSB type tonal
> distortion. Unless you actually look at the mode switch, you can't
> tell if it's AM or SB. I usually listed to stations in one of the
> SB modes because the selectivity and senstivity is better.
> 
Indeed this works just fine - works even better with a ISB rx like the
R-1051 where you can feed recovered USB into one earpiece and LSB
into the other.

> Since the internal IF drift in a R390A probably exceeds this by a factor
> of a few thousand times or so, with no PLL ability, I fail to see how a
> piece of gear that compensated for the transmitting station's carrier
> drift and mentions tuning errors of less than a hz would work worth a
> damn. Did I miss something here?
> 
The sync detector will have a lock in range - how close the rx has
to be tuned to the tx before the PLL circuitry of the sync detector 
can capture the signal. For the Sony 2010 rx, that's several hundred
hz. Doesn't matter if it's the tx or rx that's not exactly on freq.


> Wouldn't something like the SE3 perform much better on something that
> used PLL? I mean what good is all of this if your radio still drifts?
> I don't need anything that I have to keep retuning.

The sync detector will likely work over a wider range than you are willing
to accept as mistunign on an SSB station, so if the rx can copy SSB
well enough you haven't haken an axe to it's faceplate for excessive
drift during a SSB conversation, it should be stable enough to work
with a sync detector.

> > Nah, it just points out that some people are fanatics. :-) 

No one here, of course  :)

John  KK6IL  jlkolb@cts.com


---
Submissions r-390@qth.net
