Things tend to be safer with transistors these days. I remember a homebrew short-wave power (tube) amplifier that swallowed 400 milliamps at 2500 volts ... Izt nicht für Finger poken und Mitten grabben! Das sightseeren Rubbernecks keepen Hands in das Pockets und watchen blinken Lights!
Done that too. But later... And is was pure amazement. Nowadays, kids consider as granted a credit-card size TV to watch a Grand Prix Down Under. Live.
Nope. Just a passive circuit that was resonant with the radio signal so it could 'select' it, and then rectify it through the galena crystal which acts as a diode. There's no amplifier without an external power source.
No, I didn't say that. The LC circuit always does its (passive) job. An amplifier on the other hand needs an 'active' component (vacuum tube, transistor, opamp..) that requires power.
Let's give it a try..
A coil and a capacitor, connected in parallel, form an LC circuit. It is said to be 'resonant' for one frequency. The larger the coil and/or the capacitor, the lower the frequency (or the larger the wavelenght if you prefer). Something like the string on a guitar.
The behaviour of such a circuit is as follows: when it is fed at one end with radio signals picked up by an antenna, and connected to the 'ground' at the other end, it simply shortcircuits all the antenna signals to the ground, except the signal that happens to match its 'resonant' frequency. Which means that for that signal, you can measure a certain voltage between the ends of the circuit. Actually, an LC circuit has a very low resistance (impedance) -hence the 'short-circuit'- for all non-resonant frequencies. It is a 'filter'.
Once rectified, the tiny energy picked up by the antenna -and selected by the LC- is sufficient to generate an audio signal in the phones. The voice coil is actually brought to live by a very small fraction of the energy radiated by the transmitter.
With good quality parts, you get good selectivity, or a narrow, sharp spike in the 'bandwith curve', which means that adjacent frequencies don't interfere. Progressively, airband receiver design evolved to better selectivity. You pprobably remember when channels where 50 KHz apart (or worse?), then 25, and now 8.333...
Sometimes, we intentionally design circuits with 'poor selectivity', at least over a certain range of frequencies. We want our glider antenna to work for frequencies between 118 and 137 MHz, which is quite a broad 'bandwidth'. One way to achieve this is by using a large diameter tube for the radiating rod, typically 1', rather than a thin wire. Additional LC circuits are designed to 'reject' signals lower than 118 and higher than 137 MHz, to minimize interference from FM radio, lower TV bands a.s.o.
Basically, we want a broad open gate with steep walls on the side (engineers call it 'skirt selectivity'

.
Combinations of L and C can be designed as 'bandpass filters' (broad or sharp), and also low-pass filters or high-pass filters.
Well, here we are. And dinner is served. Does this make your fingers get itchy enough to pick up the old soldering iron again, JJ?