Noise canceller, an idea (maybe a stupid one...)

[A.Z.]

Hi folks, while looking at some circuits I started thinking (bad habit, I know ) and came out with the following idea, again, probably is stupid, yet I'd like to discuss it to find faults

First of all, let's state that the idea is to deal with "local" (in house or neighborhood) noise, that is, the one from SMPSUs, induction ovens, inverters, LED strips and then some; now, such noise is both "transmitted" along the power lines (be those domestic or outside a house) and, in some cases, radiated and ... picked up by our antenna, till now nothing new

Now let's think for a second to the so-called X-Phase "noise eliminator", for the ones unfamiliar with it, I'm referring to circuit like this one (just an example, you may find other similar units/schematics) https://ok1tk.com/qrm-eliminator/ now, that unit uses a "noise probe" (noise antenna) meant to receive mostly noise, the noise from such an antenna is fed to a circuit which allows to shift its phase up to around 180° and to adjust its level, the result is then mixed with the signal received by the regular (true) antenna, at that point, by adjusting level and phase it is possible to cancel out ONE type of noise

Now, let's step back to the "line noise", let's say that we pick up the noise from the AC mains, for example we could use a 1KpF capacitor connected to the neutral of an AC wall plug, at this point, we do know that the "signal" on the AC line is on 60Hz (50Hz in EU), so we may consider whatever else we pick up as noise

At this point let's say we shift by 180° ALL that "noise" picked from the AC mains, adjust its level and mix it with the signal from the antenna, we should now achieve "local noise" (as said at the beginning) cancellation, won't us ?

[radiolistener]

At this point let's say we shift by 180° ALL that "noise" picked from the AC mains, adjust its level and mix it with the signal from the antenna, we should now achieve "local noise" (as said at the beginning) cancellation, won't us ?

You cannot eliminate noise by shifting all noise picked from AC mains by 180°, because such noise consists a lot of components from many sources located at different random distances. It means that phase of each noise source depends on exact distance from picking point. And combination of phases picked at specific point depends on point position in 3D space. So, if you do 180° phase shift, it cannot help to eliminate noise for antenna located at another location in 3D space, because such antenna has a different combination of noise phases. But you can reduce some strongest one noise source in that way.

For example let's assume there are 3 noise sources: A,B,C. We have antenna which receive these noise sources with phase shift A=15°, B=72°, C=233°. And we have test point which receive these noise sources with phase shift A=57°, B=73°, C=7°.

As you can see there is no way to apply some phase shift for signal taken from test point in order to get A=180°, B=180°, C=180° relative to your antenna. You can eliminate just one source A or B or C, but not them all simultaneously.

In reality there are hundreds of thousands of noise sources instead of 3, so it will be even much more hard...

[A.Z.]

You're right, again it was just an idea floating over the top of my mind, and I slipped about the phase (and then some)

[Andy Chee]

How do "noise cancelling" earphones/headphones work? (I'm assuming they do "work", though I've never experienced using them!)

Obviously they don't phase shift 180 degrees, but I presume the noise pickup microphone goes through an inverting amplifier and then summed with the speaker output?

Isn't this in principle how the MFJ-1025/1026 noise cancellers work?