X10 signal problems and a "fix"

[JeffVolp]

The value of the shunt capacitor (leg of the "T") is not that important as long as it has a relatively low impedance at 120KHz.  A 1uF capacitor has an impedance of 1.3 ohms.  The physically larger 2.2uF reduces that to about .6 ohms.  Obviously, size is a consideration for the Leviton filter, which is probably why they opted for the 1uF.

FYI - I use a 4.7uF in the TSS.

Jeff

[pomonabill221]

DAMN... I just tested my homebrew filter, and I must have messed up!!!
  My problem is that I used air core inductors that are VERY sensitive to each other when they are next to, or near each other 
they are 1" diam. X 24 turns (#14 wire), with .22uF in parallel.  These calcualted out at 7uH, but air cores react with each other when they are close... darn it!
  I ended up using just ONE inductor/.22uF parallel combination with 2uF to neutral (thanks Jeff for the info), and the notch is down at 120khz, like it should be.  It does load the X10 carrier a little, but I will try this as a first shot.
  I need to get some KNOWN ferrit or powdered iron cores.  All the ones I have (a few) don't match anything that I can find.  They are from the input filter on SMPS for old computers, and are probably not usable for 120kHz.
  Anyone know a good source for cores? or what the SMPS computer supplies are, so I can wind my own?
  Thanks!