Nice work George. It appears that the 10 Hz change is indeed detectable
with the waterfall display. Now, we just need to apply it to a strong
signal with microchirp. Looks like the resolution with MixW is about the
same as Digipan. I'll try it later today.
-Paul, W9AC
----- Original Message -----
From: "George, W5YR" <w5yr@att.net>
To: "Paul Christensen" <w9ac@arrl.net>; <n4lq@iglou.com>; "A-USA,ex3"
<al_lorona@agilent.com>; "'Paul Christensen'" <paulc@mediaone.net>; "Ten
Tec" <tentec@contesting.com>
Sent: Friday, February 01, 2002 23:17 PM
Subject: Re: [TenTec] RE: More About Clix
> Paul and others,
>
> As promised, I did some brief ad hoc testing tonight to get an idea of how
> well a 10 Hz frequency change would show up on a waterfall display. The
> attached screen shot shows part of my MixW screen while tuned to WWV on 10
> MHz, with the receiver on USB and detuned to 9999.000 to produce a 1000 Hz
> beatnote.
>
> Looking at the display, you will see the red line directly under the
10,000
> label which is my 1000 Hz marker. Note that the 9999.000 frequency is
shown
> on the extreme left side of the waterfall showing the receiver dial
> setting. The bright trace under the red line is the 1000 Hz beatnote. Note
> that the observed frequency of the beatnote by MixW is 999.5 Hz showing
> that the PRO was a whole 0.5 Hz off in its calibration against WWV. I had
> previously check the 500 Hz tone in AM where receiver calibration is out
of
> the picture and MixW reported a frequency of 500 Hz, proving that the
> soundcard was as close as I could measure it.
>
> Now, in order to introduce an equivalent change in received frequency,
> rather than try to get WWV to key their atomic clock for an small offset,
I
> set my RIT for 10 Hz and then as quickly as I could I would invoke RIT by
> pushing the button and then as quickly as I could I would puch the button
> again to turn off the RIT. I have no real idea of how long the frequency
> was perturbed, but the departures are clearly visible in the display. The
> less-bright dots on either side of the main track are produced by the time
> tics and thus they provide a very rough indication of time down the
> waterfall, which incidently was running at X8 speed. The soundcard
sampling
> rate is fixed by MixW at 11025 +/- any adjustment available by the
> program, again using WWV as a primary frequency source.
>
> So, I suspect that a waterfall display *probably* could show the
microchirp
> situation if someone wants to try it.
>
> 72/73/oo, George W5YR - the Yellow Rose of Texas
> Fairview, TX 30 mi NE of Dallas in Collin county EM13qe
> Amateur Radio W5YR, in the 56th year and it just keeps getting better!
> QRP-L 1373 NETXQRP 6 SOC 262 COG 8 FPQRP 404 TEN-X 11771
> Icom IC-756PRO #02121 Kachina #91900556 IC-765 #02437
>
> All outgoing email virus-checked by Norton Anti-Virus 2002
>
>
> "George, W5YR" wrote:
> >
> > Paul, I use MixW for most digital modes and SpectroGram for general
> > spectral estimation. I have found that I can readily see a 10 Hz change
in
> > frequency on the waterfall under steady state conditions. When I get out
to
> > the shack tonight, I try to do a rapid 10 Hz up and back down change and
> > see what shows up.
> >
> > One critical factor to all this, of course, in addition to the soundcard
> > sampling rate is the waterfall speed. I speed up the normal MixW
waterfall
> > by a factor of eight, and it is surprising how much more detail you can
see
> > in a PSK31 signal. Virtually every keystroke from a slow enough operator
> > can be viewed.
> >
> > So, the microchirp detection by waterfall should use the highest
possible
> > sampling rate and the highest possible waterfall speed. I'll try all
that
> > tonight and let you know how it works out here. YMMV . . .
> >
> > 72/73/oo, George W5YR - the Yellow Rose of Texas
> > Fairview, TX 30 mi NE of Dallas in Collin county EM13qe
> > Amateur Radio W5YR, in the 56th year and it just keeps getting better!
> > QRP-L 1373 NETXQRP 6 SOC 262 COG 8 FPQRP 404 TEN-X 11771
> > Icom IC-756PRO #02121 Kachina #91900556 IC-765 #02437
> >
> > All outgoing email virus-checked by Norton Anti-Virus 2002
> >
> > Paul Christensen wrote:
> > >
> > > In theory, I believe that some of the sound card analysis programs can
work, but I'm not sure if they sample fast enough to detect
> > > the change in pitch...although we're probably speaking of less than
50-100 mS for detecting the duration of a "microchirp." Seems
> > > like any PC should be fast enough, but I just don't know if enough
will be captured to the screen. The PSK31 "waterfall" displays
> > > look like the trick because a pitch shift should be seen by a
horizontally shifted waterfall line. However, I haven't seen anything
> > > out there that will give accurate results in a small enough pitch
shift window like 10-50 Hz.
> > >
> > > I use a Fast Fourier Transform (FFT) program called SpectraPlus for my
Hi-Fi SSB work. The resolution down at the low end of the
> > > audio spectrum is good but the FFT sample size at low frequencies must
be very large in order to offer enough resolution to see
> > > small pitch changes. On the high end of the audio spectrum, the FFT
size can be made smaller, but then resolution is lost because
> > > of the display's semi-log scale. I may experiment with it this
weekend.
> > >
> > > -Paul, W9AC
----------------------------------------------------------------------------
----
|