Last month, this column discussed some of the fundamental concepts of selecting coaxial transmission lines. That included low-power flexible lines as well as some issues involving high-power rigid systems. It should be remembered that only the most elemental points could be covered in a short article. The actual selection of a large transmission line system is a topic that should be discussed at length by the station engineering staff with their consulting engineer.
The Dieletric truncated waveguide allows the transition from coaxial cable or rectangular waveguide to be made in the transmitter room.
Look at it this way — the decisions made will directly concern the operating costs of the station for 20 years or more, and have a significant impact on the selection of a transmitter and can well involve an expenditure of as much as $200,000. When performing that evaluation of the transmission line system, the choice may well be not to go with coaxial lines at all but to move on to the more efficient but costly solution of waveguide.
Waveguide offers the advantages of lower loss and extremely high power handling capability. On the negative side, the cost is higher, the installation somewhat more demanding and the wind load on the tower is usually higher than that for coaxial line. Again, the total analysis and selection process must include the impact of the efficiency on the transmitter selection along with tower capability, system cost, system reliability and the real cost of money for the purchase of equipment. Waveguide systems don't use sliding inner connections or wristwatch bands that need to be replaced regularly. Absent external mechanical damage, they will outlast anything else in the transmission chain.
Most station engineers don't really understand exactly how waveguide works. Honestly, it is difficult to explain without digging deeply into some rather rigorous math. In its simplest form, think of a waveguide system as an antenna that is transmitting the signal into good old free space. In this case, the space isn't really free but is bounded by the sides of the waveguide. Rather than being allowed to freely expand in all directions, the electromagnetic field is contained within the waveguide (hence its name) and is required to radiate only in the direction permitted by the boundary of the walls of the waveguide. At the other end, another antenna is used to receive the energy and couple it to the desired load.
Over the years, numerous configurations have been tried for waveguide, each with its own set of advantages and disadvantages. The most common type is rectangular, both in the mathematics and in operation. All waveguide is capable of operating with different modes of electrical and magnetic field distributions inside the boundaries. Each of those modes normally has a maximum and minimum frequency at which the mode can exist. Those are called the cutoff frequencies of the waveguide. That is, operation will not exist with that mode of field distribution above or below those cutoffs. Normally, the cutoff frequencies for various modes overlap somewhat, which can result in the signal changing its characteristics as it transverses the waveguide. That is not usually something to be desired. It can be a real problem when attempting to couple the energy back out of the waveguide. An example of this is found in large coaxial cables. At high frequencies, the cable starts to act like a waveguide and the apparent cable loss increases drastically. That doesn't really mean that more signal is being lost by normal attenuation — simply that the energy can't be properly removed from the cable by a standard connector. In practical terms, the cable becomes unusable at such frequencies, creating a maximum usable frequency in practice.
The solution is normally to operate the waveguide at a frequency where only the simplest of modes can exist. While not necessarily the best in terms of efficiency, such operation is very stable and is least affected by the minor variations that must occur in any waveguide, whether by expansion or minor discontinuities.
Next is the geometry of the waveguide itself. Numerous combinations have been tried and some have special advantages for special needs. The most common is rectangular where the broad dimension is twice the narrow dimension. This has been varied in similar types all the way to fully square with some waveguides having notches in one or more walls. Those special types have little application in broadcasting and essentially none in long vertical runs. Elliptical waveguide has long been popular in semi-flexible systems for microwave use. With either a smooth or grooved wall, these are easy to install, relatively inexpensive and very efficient. The large size needed for UHF television frequencies can create real problems in trying to make semi-flexible elliptical waveguide. Primarily, the mechanical problem of making such a large line both flexible and capable of standing the normally anticipated mechanical loads rules out practical construction. However, Myat is now producing an elliptical rigid waveguide for UHF television.
The other non-rectangular waveguides that are popular are the fully round waveguide from Andrew and the truncated elliptical waveguide from Dielectric. Both of those systems work very well with their own advantages. The fully round Andrew waveguide is normally used only for the straight vertical run. Andrew has made elbows but their difficulty doesn't seem to be worth the trouble. The main run, where the most loss occurs, is the vertical portion. Therefore, Andrew transitions from round to rectangular waveguide at the top and bottom of the vertical run. The rectangular waveguide is then either used to connect directly to the antenna or coupled back to coaxial line to the antenna. At the bottom of the tower, conventional waveguide elbows and sections are used to go into the building to the internal RF plumbing. Based on published data, and theoretically, the totally round waveguide is the most efficient. However, the system efficiency must also include the lesser efficiency of the rectangular waveguide and/or coaxial transmission line at each end. The round vertical run also is a bit better from a wind load consideration than the rectangular waveguide with its flat sides.
In the past, the primary problem with round waveguide has been the rotation of the mode inside the waveguide during the vertical run. As the waveguide expands when heated by the sun, it tends to decouple the transition at the top, resulting in reflections and ghosts. Andrew has solved that problem by inserting pins across the waveguide. Those pins force the fields to maintain the proper orientation and eliminate the mode change problem.
The truncated elliptical waveguide from Dielectric does make full use of elbows. Therefore, the transition from either coaxial cable or rectangular waveguide can be made in the transmitter room. The complete run then is waveguide for both the horizontal run and vertical run. While the truncated elliptical efficiency is less than for the purely circular, the elimination of the rectangular waveguide for the horizontal run compensates for much of the difference. The truncated waveguide is enclosed in a round cover that brings its wind load down to essentially the same as the round waveguide. The shape of the waveguide, as for rectangular, stops any change of the mode with minor expansion.
The round waveguide does not require tuning in the vertical run, although it is necessary at the transitions at each end. For the truncated elliptical waveguide, tuners are inserted periodically to compensate for minor reflections. Tuners are also used when necessary for rectangular waveguide, especially at or near elbows. That brings up a major point. Waveguide is tuned by distorting it. That is, either a reflection is introduced into the waveguide by inserting a tuning probe or the walls are intentionally flexed by applying pressure to specific points. That indicates the necessity of treating the waveguide carefully during installation. If it is dented, it is mistuned. The system must be installed very carefully to prevent any mechanical damage. That is one advantage of the cover over the truncated elliptical waveguide. The cover is not part of the waveguide electrically and shields the inner part of the waveguide from damage.
Another critical part of the waveguide is the assembly at flanges. The flanges on waveguide are very carefully machined to make the junctions as smooth as possible. Therefore, the alignment of the flanges is critical. Normally, the proper alignment is accomplished through the use of pins, which are used to force the correct positioning. For optimum performance, the entire installation has to be done with a great deal of care. A careful measurement of the system performance and final tuning is absolutely necessary for the best possible operation.
Finally, the problem of pressurization. No, you can't just hook up the old nitrogen bottle and dump in five pounds or so. To do so will flex the walls of the waveguide, changing the tuning significantly. The best solution to the problem is to use a dehydrator system specifically designed for waveguide. Such a system only applies a pressure of a fraction of a pound. Then, a dump valve is used to let off pressure when the waveguide is heated by the sun. When it cools off at night, the dehydrator adds air to maintain a steady pressure. Remember, it is the effect of the sun on the waveguide that primarily causes pressure change. The power rating of waveguide is so much higher than TV transmitters will provide that the RF has very little effect on the waveguide temperature.
When properly selected, installed and tuned, waveguide systems provide the best possible efficiency, performance and lifetime. If the suits in the front office get too concerned about the price, tell them it is used to carry the heat from the transmitter up the tower to reduce the air conditioning load and expense. That they can understand.
Don Markley is president of D.L. Markley and Associates, Peoria, IL.
Send questions and comments to:firstname.lastname@example.org