On tonight’s tech net, the main items of discussion centered around coax, antennas, power supplies and batteries. All things that we as hams have to deal with. Due to the length of my explanations here, I’m going to skip batteries for now, and just throw down a few thousand words about coax and antennas:
Coaxial cables come in many different types and sizes. One question was why are there so many different sizes of coax? Without getting deeply into the mud, I suggest that there are broadly two main reasons: loss and power handling capabilities. Let’s talk about power first, since it’s pretty straightforward. The more power a coaxial cable needs to handle, the bigger it must be. Part of that is like battery cables, the center and outer shield have to be beefy enough to handle the power. Another part is that the more power going through a coax cable, the higher the voltage difference is between the center conductor and the shield. Just like with high-voltage transmission lines, as the voltages get up into the thousands, the two parts of the coax must be moved further and further apart to keep them from arcing over. I have seen coax in use at a 50,000 watt TV station that is larger in diameter than my arm (and I’ve been working out!).
Regarding loss, the general rule is that the smaller the diameter of the coax, the greater the loss is going to be. Also the higher the frequency we’re using, the greater the loss is going to be. In the Wi-Fi consumer electronis industry, we routinely use 50 ohm coax cable smaller than the lead in a pencil because of the need to be able to snake the cable in and around very tight spots. We accept the fact that the loss of this cable in the 2.4 and 5.0 Gigahertz WiFi bands is substantial but, due to the short lengths we usually use, it’s acceptable.
Below is a quick list of coax cables you are likely to meet as a ham. There are hundreds of types, so this is a very pared-down, ham-friendly list. RG-174 is often used by QRP (low power) operators, because it is small and light, and easily carried in a back pack or go kit. RG-214 and RG-8 are a bit over 1/2 inch in diameter, and like LMR-400, are your best bet for carrying VHF and UHF signals for 50+ feet without too much loss. In fact, please use LMR-400 rather than either RG-8 or RG-214 for VHF+. RG-214 has a double shield, and is quite expensive, but good for use in repeater cabinets. RG-8 and its cousins are good for 2-30 MHz use, and is much cheaper also. Note that these thick cables can also withstand 5,000 volts, and RG-6 (75 ohm cable TV) is only good for 350 volts.
So why did we settle on 50 ohms for most coax? Actually, 72 ohms is best for least receiving loss of very small signals, and that’s why the cable TV industry like it. It turns out that 50 ohms was calculated back in the 1930’s (and re-calculated many times) to show that it was the best impedance for transferring power. For receiving, 72 ohms is still best (and some hams use 72-75 ohms for that reason), but 50 is a good compromise. It also turns out to be a happy value, close to the natural impedances of certain types of monopole and dipole antennas.
Taking the simple example of a steel wire antenna poking up from the center of a car’s roof, if it’s about 19 inches long, we can say that this is a quarter-wave antenna in the 2 meter ham band. It radiates (and also picks up when receiving) signals from directions generally best represented by imagining throwing a big, inflatable doughnut over it, like the ring-toss game. The pattern favors the horizon in all directions out from the car, with some pickup also from maybe 30 degrees above and below the horizon, as this drawing shows:
This 1/4-wave antenna has a slight gain (meaning you get more power or received signal than you put into it!) compared to an isotropic antenna. Isotropic is a theoretical pattern described by a beach ball–it both transmits and receives well in all directions. Our 1/4 antenna gets about 1.2 dB (abbreviation for decibels, which relate in logarithms of the powers of 10, so 3 dB is a doubling or halving of power) more gain at the horizons only compared to an isotropic antenna, because we squished that beach ball with giant hands down in the middles, top and bottom. It’s all the same amount of energy that we started with, but we’ve squirted a larger amount of it in a direction that’s more useful to us!
Speaking of squishing a beach ball, now let’s squish it down in the middle even more by changing the antenna to a 1/2-wave type. A half-wave antenna simply squirts even more energy at the horizons (so the signal goes further across a city, for instance), and even less up or down. We also lose more of that fat, doughnut shape, and it starts to look a bit more like a frisbee, but (and here’s the big deal) more of it squirts out towards the horizons, which are the useful directions we really want our signal to go. Here is an image showing “Unity”, or about like a 1/4-wave antenna, then two more antennas that have 3 and 5 decibels of gain over a dipole (that’s the little “d” after the “dB”, it’s the thing we are referencing ourselves to).
That’s all antenna gain is–we refocus the energy from that theoretical beach ball in directions we care the most about. A beam antenna becomes more like the reflector on a flashlight. Held in the open, a flashlight bulb somewhat weakly puts light out in all directions. However, that same bulb, now placed in a flashlight with the reflector right behind it, can really put out a strong beam, but only in one direction. Dish antennas are very close to flashlight reflectors, and accomplish pretty much the same task: they reflect back any energy in a single (well, mostly) direction, causing increased signal in that direction.
So, broadly speaking, gain-type antennas create this wonderful gain because they redirect the antenna’s energy in the directions we care most about. Make sure you walk away from this little tutorial with one important thing in your head: Gain antennas don’t create any new energy. If they did, the Law of Conservation of Energy would be violated, and would probably get quite upset about being violated in such a rude way. No new energy, simply the same amount aimed where we would like it to be most effective. The universe maintains its balance. No dark matter was harmed in the making of this movie.
Aha, so “gain” is actually a misnomer, isn’t it? There is nothing gained in one place that isn’t also equally lost in some other, thus maintaining nature’s balance. Maybe instead of calling them “gain” antennas, we should start a campaign to call them “redirecting” antennas. Well, maybe it would catch on…
As a final thought, we can make gain antennas either by making them a certain length, or by stacking multiple antennas, or by brute-force reflectors (like a dish). It turns out that certain lengths of antennas have more gain than others. A half-wave has 1.2 dB more gain than a 1/4-wave. A 5/8-wave wire has 3 dB more gain than a 1/4-wave wire. Stacking several 1/2-wave antennas has more gain than them all! The only “trick” to getting all this gain is that you have to make each new length or added antenna happy by matching it to the exact impedance of the rest of the antenna system, and that can get tricky, because it will rarely be 50 ohms. Not super-difficult, just tricky. The reason everyone likes a 1/4-wave antenna is because,given a decent ground plane to work with (the car’s roof, in this case), it happens to come out to a nice, round 50 ohms characteristic impedance. That means we can connect any old 50 ohm coax, and we’re done!
What if we want the gain and directionality of the 1/2-wave antenna? OK, here’s a new challenge: The 1/2-wave antenna, at its two ends, are thousands of ohms in impedance, instead of just 50 ohms (but you can find 50 ohms in the middle of that 1/2 wave antenna–makes sense?). Darn, now what? Well, we have to create a (hopefully simple and small) matching network to transform 50 ohms to several thousand, and connect it up to our half-wave antenna. The trick (and the reason I get paid for this!) is to lose as little energy as possible, and make it for cheap and simple. I can tell you that the matching network will probably involve an inductor (small coil of wire) that has a tap on it about 1/4-1/2 turn up from the bottom), and a capacitor to resonate with the coil near the design frequency. Depending on how big and lossy those parts are, we could easily lose most of the gain. You won the battle, but lost the war, so to speak. (RF engineers’ joke: What do you get if you end up with a poorly-designed matching network? Answer: New antenna company on eBay with offshore mailing address!)
OK, haha, thanks. So adding a bunch of 1/2-wave, 5/8-wave, or even 3/4-wave antennas together will give you more and more gain, offset by whatever losses you have in your matching networks (and there’s always some energy lost). A vertical stack of these antennas to make one high-gain, 20 foot-tall antenna of maybe 9-12 dB is called a collinear array. All the while, that pattern you see above gets flattened more and more out towards the horizon. Is there a practical limit to the size of a collinear antenna? Yep, eventually, the losses in your matching networks at each junction of new antenna elements will add up to give you diminishing returns. You will have less and less of the original energy available down at the antenna connector. The array also becomes more and more mechanically unstable until finally someone shouts “Jenga!” Again, you’re welcome.