Re: [Discuss-gnuradio] gain control extreme non-linearity--is my hardware broken?

[Lee Patton]

40-50 dB is at the top end of what the RF VGA can do (max of 56 dB set in the code). My guess is that the output of the DBS direct-conversion chip is saturated for your input signal when gain is >= 40.  What is your input signal?

I think the best way to see if you're saturating is to plot the average output power vs. commanded gain.  I attempted to do that  (http://lists.gnu.org/archive/html/discuss-gnuradio/2005-12/msg00257.html ) but I think I calibrated with too low an input signal.  


On Mon, 2006-01-02 at 22:19 -0500, Marcus Leech wrote:
> Matt Ettus wrote:
>
> >Lee Patton wrote:
> >  
> >
> >>I have seen that problem too. It is the result of incorrect python
> >>integer/float semantics during the setting of the gc2 variable in the
> >>set_gain() method found in db_dbs_rx.py. I will post a complete patch
> >>once I am finished investigating some other gain issues that I have
> >>noticed. 
> >>
> >>For now, the following change should work:
> >>
> >>old: gc2 = int(round(31 * (1-gain/24)))
> >>new: gc2 = int(round(31 * (1-gain/24.)))
> >>    
> >>
> >
> >
> >In the mean time, I have checked this in.  Thanks for finding the problem.
> >
> >Matt
> >
> >
> >  
> >
> Applied the fix, and it seems to have mostly worked--there's still some 
> weirdness
>   between the 40 and 50 settings, but it's much improved.  I thought 
> that perhaps
>   my hardware was broken--like one of the VGAs wasn't working.