re: [amsat-bb] PSK-31 satellite

["Frederick M. Spinner" <fspinner@xxxxxxxxxxx>]

>http://smallwonderlabs.com/swl_psk31.html :

>"Modifying either version to go elsewhere should not be problem.  Yes - 
>tuned
>circuits and a new LO frequency. Not all frequencies are created equal - so
>to speak - spectral purity could be an issue if you're moving far..."
>
>---
>
>The low-cost of this kits is due to off-the-shelf crystals from >Digikey
>http://www.digikey.com/ . The direct conversion (DC) Warbler uses 7 >xtals,
>the others - heterodyne rigs about 10. At $15 to $25 a xtal this adds >up!
>At less than $2 each one can see the beauty...
>

This is important for prototyping, but not a production sized run of rigs...

>As I believe that the bandwidth of a multiple (cascaded) >crystal "ladder"
>or "lattice" filter is mainly a percent function, a DC rcvr on 10 M >will 
>have a VERY WIDE bandwidth. The 80 M Warbler has a bandwith of >about 1 
>KHz,
>the 20 M model, due to the 9 MHz xtals, about 3-4 KHz. (Which is my >guess 
>why many SSB rigs often use 8 to 12 MHz as an IF.) I don't have >the 
>bandwidth specs on the 40 M rig.
>

Actually, the bandwidth is a function of the value of the coupling 
capacitances (as in any bandpass filter) in the lattice (or ladder).  The 
other catch is that the impedance of the filter will change with the 
capacitiance/bandwidth.  It is essential to properly match the impedances on 
a filter to have it work correctly or at all.  It is probably possible to 
increase the bandwidth of the 80m rig, for example, but then you would have 
to add four broadband ferrite transformers or lossy resistive or LC 
impedance transformers, complicating the design and adding parts count.  
However, you are probably correct in saying that simply scaling the design 
with scale the bandwidth as well.  This is all covered fairly well in QST 
July 1987 pages 24-29, Designing and Building Simple Crystal Filters, by Wes 
Hayward, W7ZOI, for the "Cohn Min-Loss" filters that both of those rigs use.

The 80m PSK rig for an example uses the "native" 1.5K Ohm impedance of the 
SA612, and uses a "lossy resistive" match on the input/output to the other 
stages/components.  I'm sure the 1K bandwidth is less than the designers 
would have otherwise wanted, but the filters are built to the impedance of 
mixers, and they used that as the design constraint.

>On the issue of 10 M (during high sunspot cycles) skip getting into a
>satellite - or NOT getting into a satellite (the bigger problem to my
>mind): With a dipole I can see a partial problem. Has anyone done
>experiments to see how high an angle will get though the ionosphere >at
>different times

Commercial and military do it all the time.  You have a Near Verticle 
Incidence of Reflection situation-- meaning the signal will come down in a 
smaller skip zone than "normal" for hams using low angle.

Since signals actually *bend* through the ionosphere (refract) at HF, not 
reflect off of it, a high angle radiator antenna will go through an ionized 
ionoshere under conditions that the low angle antennas will not-- meaning 
that it could theoretically work the sat even when skip is going on.  Note, 
the signal would still refract, which means that AOS and LOS can be 
influenced as the beam into space could be bent.  This also implies though 
the intereference potential from "skip" is fairly great.  The "working" the 
sat on the second hop situation is still possible with HF as well.  (So is 
some forward scatter and back scatter)  But it still is possible that under 
high sunspot/ionization conditions that even if the signal is sent straight 
up the signal could not escape the ionosphere.

>
>While I think that the PSK sat is very interesting, perhaps cheap, >and 
>potentially fun, and remembering Peter's quote: "based on a >proposal made
>30 years ago by G3VEL," I have to agree with others that point out >that 
>we've got to make to leap to microwaves.

Absolutely agreed.

>And lastly the issue of an "easy-sat"/repeater-in-the-sky versus >"space 
>science." It seems that in the PSK world one must have a 200+ >MHz CPU to 
>run most PSK programs.

I have run Digipan on a 486/75 no problem at all.  I would imagine that much 
of the CPU time is in the "FFT" waterfalls, windows calls, etc., etc., etc.  
I'd hope that an embedded PSK31 could make do with much less processor.

Fred W0FMS

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com

----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org