In-house interference from mains switches.

[neazoi]

Hello,
Whenever I open or close a mains switch in the house, a "crank" is heard in a near-by AM radio. this happens only durring the switch openning or closing (switch change state)
What is the thing that causes the interference? Is it caused by tiny sparks on the main switch during state changing?

 

[kripacharya]

Sparking at the switch point is one possible source.

I suspect you will get a louder "crackle" sound when you switch OFF compared to when you switch ON. Is this true ?

The other reason is that when you switch something on or off, there is a sudden change in the otherwise "smooth" flow of current. A sort of pulse is created. Such short pulses create wideband RF noise, and could be what is being picked up by your radio. Especially if it is close to the same wiring as your mains switch.

 

[neazoi]

Sparking at the switch point is one possible source.

I suspect you will get a louder "crackle" sound when you switch OFF compared to when you switch ON. Is this true ?

The other reason is that when you switch something on or off, there is a sudden change in the otherwise "smooth" flow of current. A sort of pulse is created. Such short pulses create wideband RF noise, and could be what is being picked up by your radio. Especially if it is close to the same wiring as your mains switch.

So the second case is like a CW keying without envelope mechanism. a fast change in current will create interference in a broad spectrum.
In the first case, what is the mechanism of generating RF from the sparking? I mean why does the spark generate RF without a resonant circuit?

 

[Audioguru]

The "A" in AM stands for Amplitude. The carrier is modulated by the amplitude of the signal. Interference caused by electric sparks and lightning are demodulated in an AM radio and produce sounds.
AM radio sounds awful. AM radio has high distortion and misses most of the high frequencies that we can hear.

When I turn off an electric switch that powers a few compact fluorescent light bulbs then the switch makes a loud "POP" caused by a spark but it is not heard in my FM radio. The spark is "wideband" creating many frequencies.

FM radio is much better. It does not produce amplitude caused sounds, has low distortion and has a frequency response that produces most of the sounds we can hear. Use it instead of AM radio.

 

[kripacharya]

So the second case is like a CW keying without envelope mechanism. a fast change in current will create interference in a broad spectrum.
In the first case, what is the mechanism of generating RF from the sparking? I mean why does the spark generate RF without a resonant circuit?

Here.. check out this link. Its about the earliest form of radio transmission/ reception

 

[THINKER_KAM_MAN]

just a suggestion

i wonder if its the inductance of the wiring that causes the sparks
also of course the inductance in the wiring and the stray capacitance
will caused your house to be tuned at some frequency.
and yeah there is also transmission line effects.

memee but i think its all caused by the rapid change of currents (dI)
when you switch on. so Vmax = L*dI/dt = high voltages which flash
over at the switch contacts.

but if you want to stop your switch sparking, you need spark suppression in the form of a resistor in series with a capacitor
across the switch, say 400V 0.047uf and 100R, may be.
the resistor is for limiting the surge currents into the capacitor.

 

[kripacharya]

just a suggestion

i wonder if its the inductance of the wiring that causes the sparks
also of course the inductance in the wiring and the stray capacitance
will caused your house to be tuned at some frequency.
and yeah there is also transmission line effects.

memee but i think its all caused by the rapid change of currents (dI)
when you switch on. so Vmax = L*dI/dt = high voltages which flash
over at the switch contacts.

but if you want to stop your switch sparking, you need spark suppression in the form of a resistor in series with a capacitor
across the switch, say 400V 0.047uf and 100R, may be.
the resistor is for limiting the surge currents into the capacitor.

It is of course the inductance which causes the spark. This is because current through an inductor changes linearly (not rapidly as suggested), but when a switch is suddenly switched OFF, there is nowhere to go, and hence voltage rises and rises until it jumps the air-resistive gap in the switch thus causing the spark.

There is no "rapid change of current". Rather the dI stays constant, while dT becomes very small, hence V rises to "compensate".

This is why a spark is sometimes formed when a switch goes OFF.
If however the switching action takes places somewhere close to zero-cross of the line current, then the effect is much smaller, and why the spark is also smaller or non-existent.

There is no need to think of tuned frequencies or transmission-line effects.

 

[THINKER_KAM_MAN]

just a suggestion

meemeeehaa

but suppose dI/dt = 10sec would you get a spark?

meeehummm
i think when you close the switch at t=0 I=0 so current starts to flow. Imax = RL = steady current. but Thor = L/R, L is small R is small. well just like a
Il = Ima * e-L/R its not linear
pulse into a transmission line,
meehumm, will the pulse find a resistor of Z = transmission line to
be not reflected??? don't think so, hooo, do i expect the wave to
be reflected back to the end of the line??? yeah radio waves
they will be waving at me all right.,