Technology is moving so fast these days. It was only a short time ago that the idea of using a satellite was something that television networks, or governments, or the military did, and not people such as you and me. But now this has changed and millions of people around the world have a dish on their roof and receive signal direct to their home. Many are using satellites to navigate with global positioning satellites (GPS) in their cars or in their PDAs.

One piece of this technology, especially with home entertainment satellite reception, which is often overlooked, is the cable. But, you might say "cable is cable." Why do we need to talk about them? We need to talk about these cables because they are as advanced, as technologically improved, as the dish and the receiver to which these are attached.

To say "cable doesn't matter" is like taking your expensive sports car to the cheapest tire dealer in town. Sure, he will find tires that fit. And maybe you could even drive around on them. But take that car to the race track and you will find out very quickly that the tires need to be as high-tech as the car itself. And that's the point. What your satellite dish requires is a race track compared to the television of yesterday. And the cable you choose has to keep up.

So what is the difference between an old and a new cable? Certainly, old and new television cables are coaxial cables (coax). Only coax has the bandwidth necessary to carry multiple channels of video. The difference between past and present-day requirements is the number of channels, and emerging services (such as video-on-demand) that require more and more channels, more and more bandwidth.

So the main difference between old and new cables is bandwidth. Old cable systems covered a range of frequencies up to 750 MHz, or sometimes even less. Today's generic broadband coax handles 1 GHz, which covers 158 channels of video. But cable companies today are demanding more and more channels, with bandwidth on cable out to 3 GHz, and 500 delivered channels. Of course, their whole system has to support that extended-channel delivery and not just the cable to your home.

Satellite downlink cables often carry many channels and these are delivered differently than in regular broadband/CATV systems. Inside your downlink dish, often a very small dish is a low-noise amplifier (LNA). The power for this amplifier comes from the same cable that will carry the signals back to the cable box or satellite receiver. Carrying DC power is a very low frequency. (Actually direct current (DC) is a frequency of zero, so you can't get much lower than that).

Satellites send signals down (downlink) at very high frequencies. Some are C band (4 GHz) while others are Ku band (11 or 12 GHz). While it is relatively easy to pick these signals up with a dish, running the resulting signal down a coaxial cable is extremely difficult. Very few cables can carry these high frequencies. Therefore, especially with DTH dishes, the incoming signal is converted by "heterodyning." Inside the LNA, the received signal from the satellite is mixed with another frequency. The two frequencies beat or "heterodyne" with each other. This results in an "intermediate frequency" (IF) which is lower in frequency than the original signal and therefore easier for the cable to handle. The IF for most satellites extends from 950 MHz to 1450 MHz.

As this is beyond the tested limit of most generic cables, there are special cables for satellite applications. These are typically tested and verified to 2.25 GHz, more than enough to cover existing and proposed satellite applications. Sometimes these are called DTH or direct broadcast satellite (DBS) cables.

At high frequencies beyond 50 MHz, the signal on a cable travels only on the skin of the conductor. This skin effect means the rest of the conductor can be anything, and typically is a steel wire, to add strength, with a layer of copper on top. This works very well for high frequency applications. However, we noted earlier that these cables also carry the DC power to run the LNA in the dish. This is a very low frequency. Low frequencies require a whole low-resistance conductor. Therefore, the copper conductor intended to run both high and low frequencies needs to be all-copper. The DC power runs one way down the whole conductor, and the IF runs the other way on the skin.

Cables intended to run high frequency signals need to be very precise. The impedance of these cables is 75 ohms, which provides the lowest loss for small signal, such as our satellite downlink signal. But the real question might be - is this cable 75 ohms at all frequencies?

If cables vary in their impedance, it can create a reflection on the line, which sends part of the signal the wrong way (back into the LNA). It is quite possible, with only a small amount of reflected signal, to confuse the LNA so much that it can shut down. Therefore, it is important to obtain some assurance from a cable manufacturer that the cable you have chosen is consistently close to 75 ohms.

This test comes in two versions. One is called return loss and the other is called structural return loss. Yes, to the uninitiated, those sound like the same thing. But they are subtly different. In the structural return loss test, the test cable is attached to an analyzer, and the other end of the cable is terminated. The analyzer is then adjusted to match the impedance of the cable, and then the reflected signal is measured. Any large variations in the structure of the cable will show up as reflected signals, and cables can be compared.

But I ran something very fast by you that you might not have noticed. I said "The analyzer is then adjusted to match the impedance of the cable…" so if the cable itself is 72 ohms or 78 ohms, or something other than the ideal 75 ohms, we will null out that variation and only see the large structural variations.

In true return loss, we attach the cable to the analyzer and terminate the other end of the cable. We then adjust the analyzer to 75 ohms. All variations are shown up, including variations in design and manufacturing. So take note when you look at the specifications for cables. Do they measure structural return loss or is it real return loss?

Of course, part of the problem is that large cable users buy this cable by the million feet (or even more). They want to get the lowest possible price, so there is intense and constant price pressure to get the lowest cost. But, opposite to that is the desire for a good performing cable, with a higher bandwidth (for future expansion), with excellent return loss numbers. Structural return loss is just one way they can compromise quality to get the price down. Return loss is also an excellent indicator of cable consistency.

If you are buying this by the million feet, they are most certainly not testing every roll. Most cable is made on a "process reel" (usually around 25,000 feet, 7600 meters in a process roll). They might test the first part of that roll. If they're very good, they might test the end of the roll. You can often tell, if you request return loss (or even structural return loss) data for every roll because a manufacturer will simply photocopy the report from that first roll and stick it on every other roll from that process reel. As you can see, this is not a big help and really tells you nothing about anything except that first roll.

Can you test the cable? Of course you can. And many large companies routinely test the cable they buy. If you buy a lot of cable, you are strongly urged to buy the correct test equipment and do periodic quality checks. This is especially important if you are getting "the deal of the century" and buying really cheap cable. Sometimes cheap is expensive.

And by that I mean that installing cheap cable, that fails often, will cost more in trucks rolls and man-hours than you saved on the cable. The expensive cable that works for a long time, without compromise or failure, and never needs to be repaired or replaced, might just be the cheapest possible solution.