Re: [Discuss-gnuradio] USRP/LFRX Gain Question

[Johnathan Corgan]

On Fri, 2008-12-05 at 08:09 -0800, dan s wrote:

> I've been using a waveform driver with known amplitude to calibrate
> the counts reading on the USRP/LFRX when I use the usrp_cfile_rx.py
> script file with decimation of 32 (2 MHz sampling).
>  
> I performed the calibration using 0dB gain on the USRP, and 10 dB gain
> for various frequencies from 10 to 150kHz.  The difference in measured
> voltage between 0dB and 10dB was about a factor of 3.04 for all
> frequencies.
>  
> I know that the voltage gain with 10dB should be a factor of 3.2.
> Does anyone know if this is purely in an ideal state?   I'm guessing
> that is the case and my setup is working properly?

10dB power delta is ~3.16 factor in amplitude change.

An amplitude change of 3.04 is 20*log10(3.04)=9.65 dB.

The AD9862 datasheet says the PGA gain error is +-0.3 dB.  I suspect if
you look at the tolerances in your measurement setup you'll find you are
within spec.

-Johnathan