Field tuning ERI (and other slug-tuned) FM broadcast antennas.
Copyright 1997, Chris Scott
This is essentially an expanded abstract. Although incomplete, this note identifies the elements involved in slug tuning well enough that most RF engineers can get through the procedure with some experimentation. A knowledge of Smith chart operations is assumed, along with the companion understanding of transmission lines. Please pardon my unmerciful switching between impedance and admittance units; the operations are calculated in parallel- admittance mode, but the related impedance term may also be used.
ERI (Electronics Research Incorporated) fm antennas arguably have the largest market share in fm radio broadcasting in the Midwestern U.S. In business for many years, ERI's most popular antenna type may be the "rototiller" design, so named due to its appearance.
Circular polarization came into vogue for FM broadcast during the seventies, and is now the standard. Two types of Cpol designs are practical at one-hundred MHz; the classical helical, including the "ring and stub" form, and crossed dipoles either side-mounted or placed in front of a reflector. When constructed using "fat" L/D ratios (diameter of tubing), and properly fed, the crossed dipole form seems to have better ice accumulation VSWR tolerance properties. Indeed ERI's high-power product has earned a respectable reputation in this regard, but particularly with the lower-power versions, radomes should still be considered if serious ice is expected at the installation site.
Most popular FM antennas use a transmission line "backbone" joined with EIA RS-225 flanged junctions and are end or center fed. Element spacing is usually one-wavelength but sometimes is one-half to eliminate the end-fire mode of phase addition, lessening downward radiation. Particularly when end fed, modulation of the beam width and elevation occurs with FM, but is not normally a factor with antennas of six elements or less, due to the very wide vertical pattern. This "squint" effect is due to the phase error induced into the far elements being multiplied by the number of wavelength spacings being imperfect at other than the center carrier frequency. Corporate feeding eliminates this, but is much more complex due to the large number of interconnecting lines.
Commonly, field installation results in VSWR of less than 1.1:1, and is often left untrimmed. Sometimes, due to conversions, repairs, or radome installation the impedance match deteriorates and the antenna needs to be re-tuned. If the transmission line is within specification, a match of better than 1.05:1 can be achieved with slug-tuning.
Individual elements are factory tuned close to resonance by size. A near quarter-wave line-section transformer is formed by the coaxial route between the center of the bent dipole junction and the tee section on the backbone or "inner bay." Center conductor diameters are customized to transform the impedance of the single element so that the parallel in-phase equivalent of the total number of elements is near 50 ohms. Normally this approach by itself will produce a match of 1.5 : 1 VSWR or better. ERI's means of fine tuning is novel and very cheap; take a ceramic doughnut and tape it to the center coaxial conductor below the feed.
This is ERI's implementation of the slug tuner, functioning essentially the same as the common single stub tuner commonly explored in transmission lines 101. The match method consists of just finding a point in the first half wavelength of line section where the complex impedance is 50 ohms or 20 mho, j whatever inductive, and placing a parallel capacitor at that point to cancel the inductive reactance (actually susceptance due to the parallel circuit). Simple eh?
Steatite 410 is a ceramic commonly used for electrical and electronic insulating applications with favorable properties for the task. It has very low water absorption, 220 volts per mil dielectric strength, and near six dielectric constant. A capacitor is in the making. These steatite slugs are available from ERI and must surely be a major profit center for them in judging by price, but they do accept returns on unused slugs. A selection of several slug lengths is needed, since two or more may be used to form the exact value (length) of the capacitor required. Amazingly, they really are taped in place coaxially on the center conductor with 3M/Scotch 471 electrical tape. This type of tape has favorable qualities for its mission, and has actually proven quite reliable when applied liberally.
A means of precisely measuring complex impedance at 100 MHz is required. I use a mechanical instrument called a 1602B admittance meter (bridge) made by General Radio. Various newer forms of network analyzers are available and work just as well.
I developed programs for my HP calculator that automate the transmission
line calculations needed for this job, but the educational method uses the
graphical approach using Phil Smith's wonderful chart. The expanded scale
version with 1.6:1 VSWR at the perimeter is wanted for maximum utility.
The two major questions are:
1) where exactly on the line section does the slug go? And,
2) How many inches of slug length are required?
Assemble the test equipment connected through a low VSWR EIA to "N" adapter (no plate types). The 1602 can either be fastened directly to the adapter connector, (best, but requires an additional line length correction step) or it may be connected through a clean slider-type variable length, constant-impedance line section. This second method allows direct reading of load end admittance.
Break open the flange connector where the antenna feed and the transmission line join. This should be just below the six-foot rigid "matching section." Do not let the center conductor of the six-foot section fall out. Secure each end so both are stable and mechanically clear. Carefully remove the center conductor from the six-foot section. Measure and record the length between the bottom of the center conductor and the center of the existing slug(s). Measure and record the total length of slug(s) used. Mark this initial position with a grease pencil..
Meanwhile, down in the warm transmitter building, measure the long transmission line's open-circuit impedance (susceptance plus line loss). If using a variable length line section, adjust and lock the length to display an open circuit with no reactance. This is done exactly on center carrier frequency. Note that it makes a difference whether or not the center conductor coupler, or "bullet" remains or is removed. If it remains in place, the line will appear longer. The exact correction factors for removed or in place can be found by experiment using a ten-foot section of rigid line in a comfortable electronics lab.
Back up on the tower in the cold wind, remove the slug(s) from the center conductor and re-mate the flanges, being careful not to damage the bullet fingers above the six-foot section by misalignment. The climber must be well outside the antenna aperture, at least ten feet from any bay.
Measure the complex impedance and plot it precisely. Label this spot "uncorrected." Draw a VSWR circle through the point. Find the single point on the circle that intersects the 50 ohm (20 ms) line, on the inductive susceptance side. Label this "slug spot." Draw a line from the chart center through this point extending to the perimeter. Note on the Smith chart the portion of a wavelength counter clockwise toward the load from this perimeter point to the zero susceptance perimeter point. Convert this wavelength portion to inches. I calculate inches per wavelength for pure air dielectric line by dividing 11,760 by the frequency in MHz. Multiply that figure by the wavelength portion to set the distance from the bottom of the center conductor, up to the center of the slug(s).
Slugs are offered in 1/4 inch increments. To predict how much correction is necessary, read the inductive susceptance component shown at the slug spot. The units are either ms or mMho, depending upon your age. Use the following constants to convert to inches of steatite:
3-1/8" line: 4.32 mMho per inch.
1-5/8" line: 4.10 mMho per inch.
If you're really careful, the load now is matched better than 1.05. I have done this on the first try twice. There remains however, a collection of minor error sources that can require the trial and error method, "bracketing" the numbers at each step. I usually tell the tower man to expect ten open - close cycles so he will be pleased with fewer.
No tuning session is complete without a bandwidth "sweep." The 1602B was calibrated to read complex impedances directly only on center carrier frequency, but it still shows values that will reduce to VSWR within plus or minus five-hundred kilohertz span without readjustment. The error created by omitting the susceptance standard re-balancing step is less than .5 percent. Therefore, the drill is to measure spots at plus and minus 50 kilohertz intervals to learn the match center, revealing the true quality of the tune. Because of the generally good bandwidth of the ERI design, it is common practice to intentionally compromise the center match slightly, tuning one-hundred or two-hundred KHz high as a pre-correction to increase ice tolerance.
This process may seem just a bit daunting. It is in fact complicated
enough to require some practice in the lab before committing to doing
it in the field, but really is not too difficult once the chart operations
and test gear are understood. Now if one can just find a 1602B ...
GoTo cs&a HOMEPAGE