[hpsdr] QSD gain and noise.

[Philip Covington]

On 9/26/07, Robert McGwier <rwmcgwier at gmail.com> wrote:
> ***** High Performance Software Defined Radio Discussion List *****
>
> I measured the Flex 5000A production model I have.  The MDS in 500 Hz at
> 50.1 MHz is 6 dB worse than it is at 80 meters with preamp on.  Since
> the preamp defaults to on at 10 and 6 meters,  with the preamp off, the
> best I could do was compare at 80 meters and 12 meters.  The MDS was 1
> dB worse at 24 Mhz than it was at 3.5 MHz.
>
> I do not understand where the notion that the noise floor in the 5000
> rises 20 dB from best to six meters.  I suspect that this was due to
> measurements we made in the SDR-1000.  It just isn't so.  There has been
> some discussion that there is some theory behind this.  It is faulty.
> The reason the noise rises in the 5000 is the choice of switch and this
> is perfectly born out by the measurements I just gave above.  The switch
> does begin to lose aperture above 24 MHz.
>
> This is not a theoretical fault of the QSD.  It is a practical fault of
> the parts used from my measurements.
>
> Bob
> N4HY

I think the 20 dB increase in NF for the F5K has been misquoted.
Before now, the only data point we had was from screen captures of
your beta F5k which shows approx 10 dB increase in NF from 14 to 50
MHz.  You also indicated that considerable hand tuning of the circuit
was done to improve the QSD from beta units to the production units.

Any which way it shows the QSD can be improved (judging from the
considerable improvement between the SDR-1000 and the F5K) and more
work needs to be done.

Phil N8VB

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
***Disclaimer***
This email contains my personal opinion* however correct or mistaken,
and should not be construed as "ill will" towards any person or
manufacturer.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

* <http://en.wikipedia.org/wiki/Opinion>

 1190892632.0