The G-Line Transmission Line
February 1964 Radio-Electronics

February 1964 Radio-Electronics [Table of Contents] Wax nostalgic about and learn from the history of early electronics. See articles from Radio-Electronics, published 1930-1988. All copyrights hereby acknowledged.

In "The G-Line Transmission Line" from a 1964 issue of Radio-Electronics magazine, Owen Patrick recounts his innovative solution to bring distant TV signals to his Carmel Valley home, spurred by his wife's desire for "good channels" enjoyed by neighbors atop a ridge 5 miles away, despite a 100-mile gap from broadcast sources like San Francisco. Facing high-loss coaxial cables and problematic open-wire lines, Patrick opted for a surface waveguide dubbed the G-line - a single insulated wire developed by George Goubau (the "G" in G-line) in 1953 - offering low loss (6 dB/mile), no radiation, and a 300-ohm impedance, ideal for spanning 3,400 feet from Saddle Mountain's 600-foot peak. Using No. 6 aluminum wire with polyethylene insulation and custom launchers made from aluminum tubes and Plexiglass, he matched the G-line to 72-ohm coax, overcoming terrain challenges with minimal cost and maintenance, delivering Perry Mason to his valley-bound TV after six years of reliable service - proving both his ingenuity and the G-line's prowess for electronics hobbyists yearning for distant signals.

The G-Line Transmission Line

By Owen G. Patrick

My wife (bless her un-electronics-oriented little heart) had been chipping away for quite some time with "Why can't we have all those good TV channels the Cinch Joneses get?"

My explanation that we live in a valley, walled in by mountains, while the Joneses, though a mere 5 miles away, are in the open; that the "good channels" are 100 airline miles from the both of us, but they could receive them and we could not, cut little or no ice at all.

"Well, you're in electronics, so do something about it" was her understanding reply.

When neighbor Bill added his "Say, why don't you -" to hers, the ground was freshly plowed and ready to seed. Perhaps I could make like a modern-day Merlin and conjure up those "good channels," thus making Number One happy. I must admit that, in addition to the stimulus of the challenge, I too liked Perry Mason.

There were acceptable signals from San Francisco, San Jose, Sacramento and Oakland atop a 600-foot hill known locally as Saddle Mountain. The big problem was how to transport all that flickering drama over some 3,400 feet of rough terrain to our house on the floor of Carmel Valley.

I began to consider common forms of transmission lines, and found that none were right for us. Coaxial lines, even the more recent types developed for community antenna distribution systems have fairly high losses in the upper vhf. These lines are also expensive and require messenger support, which means frequent poles and special equipment.

The relatively high loss of 3,400 feet of this line would also require an antenna-site amplifier, which means duplexing power and rf on the transmission lines. Open wires are not as lossy as coaxial lines, but they too present problems, not the least of which is radiation.

If lines with spacers are used, there will be losses in the dust and moisture bridges formed at each spacer. Open-wire lines (without spacers) were considered - two wires under tension, separated only at the supporting poles - but this type of line will modulate the signal, causing video flutter somewhat like sound bars. This is due to a rapid impedance variation when the lines "sing" in gentle winds. Also, this type of line is subject to bird-perching, which usually results in the conductors being shorted together.

I finally decided to abandon the usual lines and techniques and to investigate waveguides. Not the rectangular or round "plumbing" usually associated with the term, but the external waveguide, a "surface waveguide," as it was termed by George Goubau, who developed it at Fort Monmouth in 1953.

The G-line, as it came to be called, is just a single insulated wire. When rf power is properly given to and taken away from this single conductor, it responds with some amazing results. Its loss is very low, approximately 6 db/mile over the vhf TV band. This means the output voltage from a 1-mile line is half that of the input. Unlike coax or open-wire line, its loss does not increase with frequency. It does not radiate. Its impedance is approximately 300 ohms, and it is inexpensive.

G-line is not a panacea for all situations. For those of you who live in an area given to long periods of snowfall, read no further, for this and heavy ice loading on the line will increase attenuation severely. Rainfall and fog, however, affect the line very little.

If a situation requires abrupt sharp turns, the line can make it, but the rf cannot, and will spit right off the turns. This doesn't mean you can't make turns with the G-line. It just means that you have to make the turns gently, with a large radius.

The G-line's last shortcoming is that it must be held clear of all objects for a radius of a half wavelength at its lowest operating frequency. This includes buildings, poles, the ground, metals and foliage.

Why - and How - It Works

Sommerfeld, in 1899, suggested the possibility of an external waveguide, but attempts to transmit rf energy in this way failed. Then George Goubau, a German-born physicist, discovered that all the line needed to make it propagate was insulation.

Examining the electric field within a longitudinal section of solid-dielectric coax line, you can better understand the role of the insulation Goubau found so necessary. The electric field lines vary in direction and magnitude with the distance along the line. The entire pattern, maintaining relationships shown in Fig. 1, moves away from the source at nearly the speed of light. Notice particularly that the electric lines are shown beginning on charges on the surface of a conductor and extending to charges on the opposite conductor.

All of this is just another way of saying that there is a difference in potential and that electric lines of force are depicted as beginning and terminating on charged bodies. That is, they begin and terminate on charged bodies if the transmission line is not to radiate. Herein lies the problem, for what we plan to do is to expand the outer conductor gradually, until it becomes so large that, in effect, it doesn't exist electrically.

Normally this would mean that the electric lines of force beyond this point would have only one body on which to begin or end, and hence could detach themselves from the conductor and join heads to tails, forming closed loops that move away from the wire, taking all the energy with them. That is, if the outer conductor is removed, we expect the inner conductor to become an antenna. This is true if the dielectric before and after expanding the outer conductor is the same (air, as an example). However, our example is a solid-dielectric line, and the secret to the operation of the surface waveguide lies in continuing this solid dielectric after the outer conductor is removed.

Where the outer conductor has an increased diameter, the dielectric is no longer all solid; it is part air and part solid. The speed with which the charges move on the surface of conductors de-pends, among other things, on the surrounding dielectric. The charges, to which the electric lines are attached, tend to move with a greater velocity on the surface of the outer conductor than on the inner conductor, thus forcing the electric lines to curve (Fig. 2).

If we continue this process, gradually making the outer conductor larger and larger, the electric lines will bend in an arc so that they begin at a charge of one polarity on the inner conductor and end with an opposite charge on the same inner conductor (Fig. 3). Thus the electric field becomes imprisoned by one conductor and cannot detach itself, whereupon the expanded outer conductor is unnecessary and can be omitted.

In practice, the enlarged portion of the outer conductor is called a launcher. A complete transmission system requires one at each end of the line. The one at the receiving end has been called a catcher.

All this seems to suggest that these launchers should be exponentially shaped. Experiments show that this form is not necessarily best, and in any case, not many of us have the wherewithal to construct such a gadget. I decided to corrupt the optimum in favor of a design that was easy to make and much less expensive. Fig. 4 shows the natural transition from the ideal to the simplified practical. This simplification only adds about 1 db loss overall.

Put One Up

The wire I used in our system is No. 6 aluminum with a black commercial-grade polyethylene jacket (Fig. 5). Aluminum was selected because this system consisted of two 1,700-foot spans and here the weight is important. Any plastic-insulated wire will do, such as common No. 14 or No. 12 solid-copper house wire. Rubber-insulated and so-called "weatherproof" wire should not be used; those dielectrics are lossy and absorb power.

Ideally, the line should be constructed from wire with an overall diameter twice that of the conductor itself. However, variation from this formula does not change the performance greatly.

Fig. 8 - Rear of launcher shows "eye" bent into aluminum wire, and nylon support rope. Box on back of launcher "sandwich" is transformer to match G-line to 72-ohm coax. Short piece of ribbon lead-in goes from 300-ohm side of transformer to end of G-line.

Our launchers were made with four 1/2-inch x 6-foot aluminum tubes located between and along the diagonals of two 12-inch x 0.060 inch aluminum squares (Figs. 6, 7). Two small squares of 1/4-inch Plexiglass or Lucite with holes in their centers to admit the bared end of the wire form the insulator block. A knockout punch cut relief holes in the centers of the two aluminum plates. The entire sandwich was secured by 1-inch 6-32 machine screws.

The size, height, weight and whether or not "deadmen" will be used at the terminals of the G-line will depend on the particular circumstances. Saddle Mountain's two 1,700-foot catenaries were suspended between 6 x 6-inch x 10-foot posts set in the ground 3 feet and tied back to expandable anchors. These supports were later found to be far more rugged than necessary since this line required only a 200-lb stress.

Be careful not to exceed the maximum load rating of the wire under the strongest winds expected in your area.

The tie between the supports and the line is 1/4-inch nylon rope. Since nylon is very slippery and doesn't knot well, it was bound with copper wire. Nylon rope can be kept from unraveling by fusing the ends with a match or cigarette lighter flame.

The bared end of the G-line is passed through the hole in the launcher insulator (Fig. 8). A slotted bolt connector is slipped over the wire at the rear of the launcher and the wire is then formed around a small cable "thimble" and once again passed through the wire connector.

Once the connector is in place, the nylon tether is secured to the thimble and the G-line is ready to raise.

In our system, 72-ohm coaxial feed lines are used to and from the launchers. Weather-tight transformers such as Jerold's model TO-374 or Taco's Magi-Mix No. 1597 match the G-line's 300-ohm impedance to that of the coaxial line.

Even though transformation to an intermediate impedance of 72 ohms between the G-line and the TV set required a transformer at both ends, it was well worth the small additional cost to be able to run the interconnecting line next to water pipes, on the ground and through walls with complete freedom. With 300-ohm line, a transformer should still be used, even though the impedance of the G-line matches that of the ribbon. A 300-ohm line is balanced, and the G-line is inherently unbalanced.

Our system has been in operation more than 6 years. It has survived winds over 80 mph and has withstood the tests of time and weather with complete success. Its entire cost was less than that of the set we couldn't use without it and everybody's happy. Now don't let a little thing like a mountain range stand between you and Perry Mason!

***

Since the name G-Line is a trade mark used by Surface Conduction, Inc., for their products which are also protected by patents, we have asked for and received permission to publish this article under this name. The permission was given in view of the fact that it represented a private study proving both the ingenuity of the author and the reality of the G-Line.