I've always been too embarrassed to ask, but the notion of "ground" has always puzzled me. The way I understand it, in theory, is that the earth acts as a giant sink/pool of electron and thus will maintain its potential. Thus, if I connect one electrode of a battery to the ground, the other to a lamp and the other end of the lamp to the ground, the lamp should turn on.
However, if I try this with a 5V battery and a small incandescent light bulb, planting a small wire into the dirt, it doesn't work. And hum, why should it, it's not like the dirt is particularly conductive...
So is this something that only works for high potentials? When you have a lot of contact area between the electrode and the "ground"? What exactly constitutes a valid ground?
Answer
The term "ground" means several things in electronic circuits, radio, and in power electronics.
In electronic circuits, "ground" is merely the one node in the circuit that other voltages are implied to be in reference to. A voltage is always between two points. The statement "This line has 5 V on it" would therefore be meaningless without the implied assumption that is really means "This line is at 5 V higher potential than the ground net". In this case "ground" has nothing necessarily to do with the earth. The fact that the implied 0 reference node is called "ground" is historical only.
In radio, "ground" can be a integral part of a antenna. For example, most commercial AM stations (around 1 MHz) have a vertical antenna that is basically a big wire sticking up into the air. The RF signal is fed into the bottom of the antenna with the other end of the feed from the transmitter connected to earth ground close to the antenna. This causes the ground itself to act like a mirror to the RF signals when viewed from the half-space above the ground. Due to this mirror effect, a receiver sees a whole dipole, even though the bottom part is just a reflection of the top part.
Since the wavelengths of commercial AM radio are large (around 300 m), the most handy thing to use for the mirror is the surface of the earth. Small antennas for much shorter wavelength can use a metal ground plane for the same effect. However, since the ground plane should extend for a wavelength or a few around the antenna, for 300 m wavelengths a deliberate conductive metal plane nearly a km accross is usually not logistically and economically possible. In the case of a small metal plane, we still refer to it as the "ground plane" of the antenna, even if it has no connection to earth ground. Again, this is due to history. Originally earth ground was used, then the same term kept for the same concept, even when real earth ground was no longer envolved.
In electric power systems, "ground" usually does mean a real connection to the earth. You are right in that soil is often a poor conductor. This is why ground stakes, which are for providing deliberate connection to earth ground, are usually several meters long going into the ground. By making the ground stake long, you get a lot of dirt conducting in parallel, which helps. Even so, getting a good low-Ohms connection to earth ground is not always easy.
Here in North America, electricity is delivered to houses via a center tapped secondary of a transformer. The secondary produces 220 V, so either side to the center tap produces the 110 V that are used for ordinary circuits. All three transformer connections are wired into the house, then the center tap is connected to a real earth ground.
Metal water or sewer pipes that go underground to connect to a larger line underneath the street make good ground connections, both due to their length and diameter. My house doesn't have town water or sewer, so there is a 1/2 or so copper rod driven into the ground near the breaker box, with a thick copper cable connected between it and the neutral in the breaker box. I don't know how far down the copper rod goes, but probably a few feet at least, maybe 10 or more feet. This probably varies by local electrical codes, but I haven't checked.
No comments:
Post a Comment