Eleminating radiation from chargers

Question

When I move my EMF radiation meter near a charger it shows me that there is radiation from the charger, and even the device itself radiates when the charger is connected to it. The chargers emit radiation even if the device is not connected to them.

I understood that the radiation is caused because of the way the modern chargers are working. They are working in a way that they switch on and off very quickly. It is done by a DC/DC converter.
It results in a very high frequency noise (radiation) that travels on the charger's wires and even on the connected device.

But I noticed that when I connect the ground of the output of the power supply to the ground of the wall socket, there is no radiation!

So I have some questions:

1. Why it eliminates the radiation?
2. Is it safe to do that? (I know that the two sides of a transformer need to be isolated from each other, but it is just in principal).
3. Will it damage the device, the charger or the socket?
4. If it's ok, why the chargers are not designed in the way they will not radiate? (The chargers have only 2 connectors to the wall socket so why they don't have the third connector to the socket's ground?)

Thank you!

Answer 1

The coupling capacitance in the isolation transformer causes common mode noise voltage from the primary side switcher not the low voltage side. So this Common Mode DC pair extends the CM noise to the laptop. Some have a torridal ferrite to absorb some of the CM signal but when the laptop is floating there is not much attenuation unless RF caps were used to the primary earth ground. This current is limited by the Earth ground leakage test limits often 250uA or 500uA depending on country. So the winding coupling capacitance often radiates with line frequency modulated 50kHz or whatever switching rate. Then external unbalanced mic’s often induce some of this E field into the high CM input impedance and causes nonlinear conversion to differential mode noise.

Earthing gnd shunts and attenuates the CM high impedance Signal to low levels. That is simply an impedance divider.

This also works with any VGA video cable as this cable has earth ground from the 3 pin power plug.

The only risk is a ground fault to the grid could energize the metal edges of the laptop case momentarily or an open ground elsewhere but to a machine with high ground current may pass to a user touching a good ground.

A better solution would be to have a perfectly balanced high inductive CM choke with RF caps to earth ground to both shunt voltage but at lower CM noise current levels But this adds great expense it seems so it is not done for the few that might benefit from it.