[Elecraft] PSK31

[Charles Greene]

Dan,

Well, the shape is the same if you mean it has three peaks and two 
valleys.  But the frequencies and absolute and relative amplitudes are 
different.  For example, take my USB setting using Spectrogram v6:

FL1, 2.48 khz BW, BF1 138, peaks at 353@-71dB, 1074@-63dB, 1922@-75dB

FL2, 2.48 kHz BW, BF2 122, 752@-66, 1455@-65, 2306@-78

FL3, 2.48 kHz BW, BF3 161, 360@-68, 1163@-70, 1772@-79

You get a whole new set using LSB.

I have the V B mods and 14 matched xtals, but haven't made any other mods 
to the filters.


At 09:18 PM 7/22/2003 -0400, you wrote:
>Chas,
>I know the center frequency changes and the entire passband shifts to the
>left or right, but the shape of the passband (should) remains the same.
>
>If you are referring to what happens if you lower the passband so that it
>crosses 'zero' frequency (actually goes into the opposite sideband), then
>you will get a 'funny' distortion - but then that is just the way any
>specrtum analyzer falls down in the task because it cannot differentiate
>from one sideband to another, nor can it display very low frequencies
>because the FFT at zero frequency is undefined mathematically, so you just
>get a spike - but that is not really what is present.
>
>What I do see is an apparent change in the passband ripple (even if I do not
>change the BFO) - but that is mainly because of the coarseness of the FFT -
>chosing more resolution will give a more true and smoother picture (and
>takes a longer time), but usually the faster display is sufficient to show
>the corner frequencies, and that is mainly what I look for.
>
>Is that what you were referring to, or are we still not communicating?
>
>73,
>Don W3FPR
>
>
>----- Original Message -----
>From: "Charles Greene" <W1CG@QSL.net>
>To: "Don Wilhelm" <w3fpr@peoplepc.com>
>Sent: Tuesday, July 22, 2003 8:17 PM
>Subject: Re: [Elecraft] PSK31
>
>
> > Don,
> >
> > I don't understand.  Of course the filter changes shape when you change
>the
> > BFO freq.  Try it.  It changes significantly.  Why wouldn't it change?
> >
> >
> > At 10:21 AM 7/22/2003 -0400, you wrote:
> > >Chas,
> > >
> > >Just one comment -- if the filter SHAPE changes when you change the BFO,
>you
> > >can conclude that Spectrogram is not showing you the precise shape of the
> > >filter.  You may have to use a finer sampling resolution to see a better
> > >picture -- but then, it probably won't make that much difference anyway,
>so
> > >we don't have to make a big deal of it except for 'intellectual
>curiosity'.
> > >
> > >73,
> > >Don W3FPR
> > >
> > >----- Original Message -----
> > > > .... Then I used a noise generator and Spectrogram to
> > > > align all my narrow filters to this frequency.  The freq I use is 1000
> > > > hz.  You can change it by BFO setting, but when you change the BFO
>setting
> > > > on FL1, it changes the filter shape too....
> >
> >
> > 73, Chas, W1CG
> > K2 #462B
> >
> >
> >


73, Chas, W1CG
K2 #462B