[evlatests] RFI birdies in deformatters.

[Mike Revnell]

I've been trying various things with one of the deformatters downstairs 
and so far nothing I've done has a measurable effect on the level of the 
birdies.

The level of the signals (-55dBm or ~1mV) is high enough that it gives me 
some hope of being able to do something about it. Unfortunately this 
involves changing the board layout somewhat.

I had a look at the board layout and I've obviously learned quite a bit 
since I did it. I can see a number of things I can do that are likely to 
help without doing great violence to the rest of the board. The down side 
is I can't be certaion they will fix the problem entirely.

If it is, in fact, a transition issue and there is improvement then the 
current boards could be set aside until after the new correlator is 
installed when the analog output is no longer of interest.

The cost of extra components would be a couple bucks for a linear 
regulator chip. Most of the changes would be a more careful layout of 
power and grounds delivering power to the output stage of the DAC.


Mike Revnell
National Radio Astronomy Observatory
Socorro, New Mexico
mrevnell at nrao.edu
revnell at ieee.org