Maybe I didn't explain that very well. Maximum attenuation will be at the notch, which should be about 120KHz. Attenuation outside the notch decreases. While dependent on the Q, attenuation below 80KHz and above 200KHz should not be significant. That dovetails nicely with the input stage of most X10 modules, which have a tuned bandpass filter centered at 120KHz. You just need to block in-band noise that would get through that input bandpass filter.
Actually, a low-pass filter like the XPPF is much better at isolating noisy electrical devices from the powerline. However, the physical size of the inductors needed to build high-current low-pass filter with sufficient rejection at 120KHz would make the unit large and expensive. I am not aware of any being produced. Maybe I should consider offering one for special applications...
Jeff
The XPPF sounds very similar to the Leviton 6288 - big hulking inductors (115uH). It relies on the series impedance of the Inductor to perform it's filtering rather than the LC resonant point of the 6287.
I agree these filters are more appropriate for broadband noise sources. They are also less prone to variation due to component tolerances. I have a number of the 6288 style (smarthome) filters that vary in effectiveness due to component tolerance variation.
I've never seen a 6288 style filter in anything approaching 15 amp (typical is 5A). The inductors in this type of device would be very large indeed to keep the coil resistance and power dissipation down.
When I read the ACT filter description, I had envisioned a 6288 style lowpass filter with a high frequency bypass cap (to give the High Pass function). In reality, I don't know why anyone would want to add a cap to provide the High pass feature. This is my long way of saying that you are most likely correct. This is likely a bandstop filter, not a lowpass + highpass as described.
Thanks for keeping me honest,
Boiler