Re: [Discuss-gnuradio] relation between the amplitude and the voltage of the signal

[Marcus D. Leech]

On 03/25/2011 08:01 PM, Yan Nie wrote:
> Dear all,
>
> I'm wondering how to correspond the amplitude set in software to the voltage of 
> the signal. I did some test, which demonstrate that setting the amplitude to 
> 32767 and the GPA gain to 20dB could reach full scale 2V p-p. I'm confused 
> about the amplitude setting in software, if 32767(which is 2^15-1) reaches the 
> full scale, which means the amplitude is 16-bit data, but the ADCs are only 
> 14-bit. How could the amplitude setting correspond to the voltage of the 
> signal? What about the receiver side?
>
> A simple question: if using usrp_source_c to receive complex signal by LFRX, 
> does both of RX_B and RX_A ports should connect to antenna? This is for double 
> checking if my system setting is correctly. Really appreciate any of your help!
>
> Wish you a great weekend!
>
> Yan
The numerical view within the flow-graph corresponds only very-roughly 
to what happens in the analog world on the other side of
  the ADC/DAC.

In the UHD universe, using complex-float representation, everything is 
scaled to +/-1.0 by UHD so that flow-graphs can be reasonably
  "generic", and UHD takes care of scaling for the appropriate hardware.

For both transmit and receive, if you want a high-precision correlation 
between what the flow-graphs world-view is and the outside world,
  you'll have to *take measurements*, and calibrate appropriately.   
You calibration is going to be slightly different from daughtercard to
  daughtercard--just a fact of life in the analog world.  Total gain 
can vary by a dB or three, even for identical settings in the digital world.


--
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org