Re: [Discuss-gnuradio] USRP1 gain

[Marcus D. Leech]

On 10/09/2013 11:03 AM, Nemanja Savic wrote:
> You are right Marcus, but it is negative gain, and I wonder why USRP 
> source allows me to specify more than 20 db of gain for wbx, without 
> any warning. In fact what happens when u specify some gain value, how 
> is configuration of the path set?
>
> Thanks
Gain is distributed by UHD starting at the antenna, and moving towards 
the ADC.

In the case of WBX, the attenuator sits between two LNA modules.

And, I'll emphasize again--the way gain-control is implemented in RF 
circuits is with a variable attenuator.  If you purchase an RF
  Variable Gain Amplifier chip with a voltage-control input, it's 
overwhelmingly likely that the way it's implemented internally is with
  a voltage-variable attenuator.

The thing to "care" about is whether at some gain-setting value 'X', is 
my signal-processing-chain "happy" with the results.
  If you're doing things like precision power measurement, you *HAVE* 
to use an external calibration.  You can't just play
  math games with adding up nominal circuit-element gains, because each 
element in the gain chain that provides a gain
  will have device-to-device variability of 1dB or more.


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