[hpsdr] Alex test results

[John Ackermann N8UR]

I've done network analyzer tests of a production Alex unit.  The data 
will probably end up in or linked to the Alex wiki, but there's a first 
cut of a web page with plots for each band on RX and TX plus some other 
info at http://www.febo.com/pages/alex/index.html .

In summary -- it works.  On the receive path, the worst-case insertion 
loss through the system is about 1.5 dB (on 10M).  There is at least 70 
dB of RX rejection from 60 MHz through 200 MHz, and more than 90 dB from 
75 MHz to 175 MHz.

On TX, the worst-case insertion loss is 0.9 dB (on 6M); below 10M, the 
loss is less than 0.5 dB.  Rejection at the 2nd and 3rd harmonic of each 
ham band is at least 40 dB except for 30M, where the 2nd harmonic is 
about 26 dB down.  30M uses the same LPF as 20M, and the cutoff 
frequency is a bit too high for optimum 30M performance.  In-band return 
loss is around 20 dB (SWR of 1.25) or better on most bands, though 10M 
is more in the 13 - 15 dB range (SWR of 1.6 to 1.4, so still nothing to 
worry about).

There is one important thing to note if you want the best rejection of 
VHF/UHF signals.  As noted in the Alex instructions, you need to put a 
short ground wire between the RX and TX boards, and also from the boards 
to enclosure.  If you don't, there is likely to be a "hump" in the 
frequency response somewhere between 90 and 130 MHz that could be as 
much as 30dB worse than necessary (though you'll still have >50dB 
rejection, so it's not the end of the world).  I have details and photos 
of how I added the grounding linked to the page above.

Congrats to Graham for the great design, and to Scotty et al for 
bringing it to life!

73,
John