[EdHolland]

Grea
One way to look at the weird behaviour of a radio is that it is a voltmeter - it measures or "hears" the difference between its antenna and chassis terminals.

Now if there is a good antenna and low impedance ground connection, reception is good and noise free. Simple! Most of us deal with something different. Another more common situation is a fair antenna, but no chassis connection except the capacitors linking to the line and neutral. In this case a few things can happen. First, it can be fine - clean mains that has a low rf impedance to ground = good reception. Alternatively, if there is rf noise on the mains line, this adds to the aerial voltage, and a vareity of fuzz and buzz interferes with our stations. The radio is acting honestly to what is across the chassis and antenna terminals, but it isn't what we want to hear - a good ground should clean this up. Another effect is where there is good signal from the antenna, and 50/60Hz on the chassis from the way the ac power line is connected. Sometimes these can mix, resulting in background hum on strong stations because the mains frequency modulates the received station inside the radio. I have seen this several times, and again it can often be remedied by connecting a good earth/ground to the chassis.

It is interesting to experiment - also to see what the original instructions recommended. These sets can receive on a short length of wire, but the balance of signal could then be coming down the mains line - the radio doesn't care. You could even connect the antenna terminal to a ground, leave the chassis disconected and still get plenty of stations from what comes down the ac line, noise and all!

And so, this is why it is recommended to have a good antenna, and a separate earth/ground connection to the chassis. Done right, it defines the paths for RF signals, sending them carefully to the radio circuits, and separates them from the ac line and any RF and interference on it.

Ed