[evlatests] Some initial 3-bit RF bench tests

[Jim Jackson]

We just ran some tests on the 3-bit digitizer board in the lab and 
got some interesting results.

We inserted a directional coupler in front of one input to the 3-bit 
digitizer board and connected a spectrum analyzer to it.  We can see 
a strong birdie being generated on this signal line at about 13 GHz. 
We see smaller birdies across the spectrum that are obviously the 
intermodulation products of this birdie and the 4096 clock and it's 
harmonics. The birdie is about 1MHz wide and definitely moves the 
same as the birdie we see on the spectrum plot from the captured data.

Though we have not analyzed it closely, 13 GHz is roughly consistent 
with the path length from the input connector to the digitizer 
chip.  We are going to look at the input of the board with a network 
analyzer to see if that gives us any clues.  We did try a magnetic 
2-4 GHz isolator and some absorber material that works at 
12GHz.  They both reduced the birdie level somewhat but did not eliminate it.

Cheers,
Jim