[hpsdr] Update to gnuradio code / soundcard notes

[Tom McDermott]

Thanks, Phil for the note about clock selection.

The gnuradio code has been updated to select Mercury as the 
122.8 and 10 MHz clock references (Hermes ignores this).
http://svn.tapr.org/repos_sdr_hpsdr/trunk/N5EG/gr-hpsdr/

C0 = 0x0   C1 = 0xF8 (48 Khz).

I'll think through some better way to do this in the future, without
adding too many parameters to the model. Also cleaned up the
sample rate initialization a bit.

Some possible issues with the sound card:
a) Not all sound cards support 16000 sample rate, 
b) You may need to override the default sound card selection.

The SSB demo app uses an audio sample rate of 16000
samples/sec. You can change it in the GRC audio sink block,
but then it is necessary to change the decimation in the low pass
filter (which is set to samp_rate/16000 currently). The demo app
has a global variable called samp_rate which controls the Hermes
receiver sample rate and the decimation ratio of the LPF.

The audio device name is blank in the sample SSB receiver
demo for Hermes (selecting the default sound card).

Here are some notes about the audio card from the Gnuradio
documentation for the audio sink block:
--------

Not all sampling rates will be supported by your hardware.

Leave the device name blank to choose default audio device.
ALSA users with audio trouble may try setting the device
name to plughw:0,0

The audio sink can have multiple inputs depending upon your
hardware. For example, set the inputs to 2 for stereo audio.

The specific audio device to use can be specified as the
device_name parameter. Typical choices are: 

  - pulse, hw:0,0, plughw:0,0, surround51, /dev/dsp
----------

-- Tom, N5EG