I see some darn nice work here guys!
Not having built a Felix myself, I can't really comment too much, but the design looks good.
But I can comment on one commercial power filter I tested that is similar, tho simpler.
I will post a schematic here ASAP.
I took this filter to a buddy's lab who has great measument gear. We ran spectum analysis of it from 1Hz to 40Mhz. The testing was both informative and frustrating.
Here is a summary of what we found. (the tests below seem easy, but it took many hours of work).
Plugged into a typical 120V/60Hz noisy AC line, the filter seemed to have almost NO effect. There was a slightly lower noise circa 10Mhz, maybe -2dB, that's all. No other real changes anywhere else across the spectrum. Certainly not any improvement near the audio band. But as the line noise was constantly changing, it was difficult to compare with and without. But after numerious tests, we saw no overall improvement.
Then we plugged in a typical SMPS line lump. More noise on the line, for sure (~80Khz) but the filter did little or nothing to remove it. Ditto for a fluorecent lamp and some other noise sources.
Next test was to put a heavy load on the filter. We used a 500W oil filled radiator heater. No change at all.
However - what we did find was this. If a noisy device was plugged into the filter instead of the AC socket ahead of it, there was a dramatic increase in noise. The filter was not letting the noise drain back out into the AC mains and away. There was a slight increase in noise on the mains, so a tiny bit did leak out, but not much. Maybe a dB or so. Mostly it stayed on the output side of the fitler and was much higher than with no filter.
This showed us that you certainly would NOT want to plug a noisy and a quiet device into a filter like this. The results would be much worse. Apparently the backward impedance of the filter kept noise from being drained away, as it is on a normal AC connection.
Finally in frustration to make the filter do "someting", I plugged in a heat gun ahead of the filter. The heat gun motor threw tremendous amounts of noise into the AC line, much of it in the audible spectrum. The filter did do a very nice job of cleaning that up. At last.
Our conclusions:
1) This filter did little to no good cleaning up high frequency noise above about 40KHz.
2) The filter prevented noise from draining out into the AC line. This was NOT good as it meant much more noise at the filter output than with no filter in place.
3) Plugging a noisy and a quiet device into the same filter would be a very bad idea. More noise for both.
4) The filter was effective at removing motor noise, much of it in the audible band, from the line. This would be a good use for the filter.
Of course these do NOT apply to the Felix, but are just an example to show that things don't always work as you think they might!