[time-nuts] DIY VNA design, directional coupler

[Charles Steinmetz]

Alexander wrote:

> there is a current mode feedback device [which does not follows the gain
> bandwidth product role ] and has 1000V/usec rise time 92dB THD at 30MHz
> 3nV/rtHz noise

Yeah, but look at its 1/f input voltage noise corner -- it's at 2 or 3 
MHz!!!  So the baseband input noise density is over 1000nV/sqrtHz at 
1Hz, with potentially devastating effects on the phase noise performance 
in phase measurement applications.  (See Figures 18 and 56 of the cited 
datasheet.)

Even if the internal amplifier topology exhibits relatively low AM-PM 
conversion, starting that far behind is not a promising way to design a 
low-PN widget.

NFL

Best regards,

Charles