load-coils
load-coils
load-coils
load-coils
load-coils

Load Coils

Dec. 3, 2016
 by: Bob Stoffels (This article originally ran in OSP Magazine) If you don’t know what a load coil is, you’re probably under 50 years of age. Unfortunately, a load coil […]

 by: Bob Stoffels

(This article originally ran in OSP Magazine)

If you don’t know what a load coil is, you’re probably under 50 years of age. Unfortunately, a load coil is a prime example of the principle of unintended consequences.

But let’s start at the beginning.

You know that the original telephone wires, the unshielded twisted pair, or UTP, wasn’t the best transmission medium in the world. Voice and data could leak out, crosstalk could leak in.

Furthermore, there’s always a certain amount of attenuation in a pair of wires, and the longer the wire, and the thinner the wire, the greater the attenuation. Also, because the 2 wires in a transmission circuit (a cable, to be precise) are quite close to each other, there is a certain amount of capacitance involved. That is, the 2 wires act like the 2 plates of a capacitor. There is another characteristic of the transmission line — its inductance — but with a single pair of wires this is not significant.

To review: In any transmission line there are 3 characteristics:
1. resistance,
2. capacitance, and
3. inductance.

With a long transmission line consisting of unshielded twisted pair in a cable, the resistance increases with length of the line, and also as thinner wire is used.

The capacitance gets bigger as the line gets longer, and as the spacing between the two wires gets smaller. Unfortunately, these factors cause distortion of the transmitted signal as it moves down the line. To be more precise, frequencies at the upper end of the voice range might suffer more attenuation than frequencies at the lower end of the range, and the higher frequencies might differ from the lower frequencies in the time taken to reach the receiving point.

By using a lot of fancy mathematics it can be shown that a partial solution to this is to add inductance to the line. This is called loading. In the past this loading was applied by the thousands. So many in fact, that accurate records were not kept.

So, what’s wrong with applying some load coils to improve voice quality? Nothing, if all that particular line is going to be used for is voice. But that, of course, is no longer the case. Data is the name of the game. And the way to transmit more and more data, at high speeds, is to use high frequencies on the telephone line. And therein lies the problem. Loading not only corrects for distortion, but also imposes a sharp limitation on the total range of frequencies that can be transmitted. A line using loading is, essentially, a low pass filter, and the high carrier frequencies used for data transmission simply cannot pass. So, if a particular telephone line is to be used for data (and what line isn’t) then the load coils have to be removed.

Thus we have a solution to the one problem; that is, removal of the load coils permits high speed data (e.g., ADSL) to be used. But it re-creates the problem of distorted voice.

Technology again comes to the rescue.

One solution is to re-design the load coil so that it not only corrects distortion at voice frequencies but also opens up high frequencies for high speed data. What we have is, essentially, a notch filter. It provides internal inductance and capacitance with a resonant point tuned to the dead space between the voice band and the ADSL band. Equalization is provided at voice frequencies, and the high (ADSL) frequencies are left pretty much alone. And thus the load coil lives on.

There is one other problem as regards high speed data transmission on telephone lines that we should mention. Frequently, in the past, telephone lines have had extra bridges applied as they go through a neighborhood. (look at it as future off-ramps). These bridge taps has simply been left un-terminated. Sophisticated equations demonstrate that a signal (especially a high-speed data signal) hitting the un-terminated end of such a bridge tap will reflect some of the signal backward on to the main line. Except it will be a few microseconds or milliseconds delayed. The result is more distortion. As a result all such taps have to be terminated in a particular resistance, called the characteristic impedance. No special gadget will correct this problem.

Transmission over telephone lines isn’t what it used to be. The game has changed. And although load coils played an important, and positive, role in the past, the appearance of high speed data has changed that.

About the Author

ISE Staff