A session of this spring’s NAB Show Broadcast Engineering Conference provided an excellent opportunity for engineers to engage the industry’s best-known experts on the proper implementation of the new AM directional antenna modeling and proof rules.
Crawford Broadcasting Director of Engineering Cris Alexander, a Radio World contributor who has been active on this issue, chaired the program, which featured consultants Ben Dawson and Ron Rackley, plus Clear Channel Vice President of AM Engineering John Warner. A panel discussion that included CBS Radio Director of Engineering Glynn Walden followed the presentations.
Building the model
Dawson, president of Hatfield and Dawson Consulting Engineers, helped craft and guide the new Method of Moment modeling rules through to adoption by the FCC. He led off the session by laying out the background of how the MoM model is constructed and why MoM DA proofs offer a much more accurate and predictable method of tuning and proofing most AM arrays.
Dawson explained that the new rulemaking limits the use of MoM for proof purposes to common and straightforward cases. Only series fed elements qualify and must use a calibrated sampling system to drive a calibrated antenna monitor. Unequal height towers do qualify. The model must closely match the measured base impedance matrix.
Certain adjustments to the physical measurements of the towers may be applied in the modeling to accommodate velocity factor effects since RF does not propagate through steel at the speed of light. The equivalent electrical height of a tower can be between 75 and 125 percent of its physical height while the equivalent radius can vary by 80 to 150 percent of the physical radius.
An MoM antenna array model depicts every tower in Cartesian or geographic coordinates (depending on the program requirements) with appropriately assigned length and diameter geometric constants. Loads such as STL dishes, top loading skirts and base reactive elements are also included.
The electrical behavior of each tower in the array is analyzed in segments.
John Warner visits a tuning unit on an inspection trip to Kintronic. Depending on tower height, 10 to 20 segments offers adequate resolution of current/voltage distribution. The moment method program drives each element of the array with a current source and generates the open circuit voltages at the elements, from which an impedance matrix is created.
The resulting data is then used by the program to generate the current distribution and impedance drive requirements when a desired set of far-field parameters is provided as program input. Actual measurements of each tower’s driving point impedance while the other towers are grounded and/or floated then are made with an impedance bridge or network analyzer. Such measurements can be made right at the tower base or at the ATU output (or both). Measurements at the ATU output are desirable if current sample transformers are used to drive the antenna monitor.
The measured results need to agree with the MoM model results to within 2 ohms or 4 percent for both resistance and reactance. When the mathematical moment method model is created, to match the measured data, the rules allow for a 250 pFd maximum of total base shunt capacitance and 10 uH of maximum connecting feed line inductance or an actual measured value.
Dawson illustrated examples of modeled towers showing how small changes in equivalent height impact base impedances much more than changes in the equivalent radius.
Sampling loops need to be placed at the elevation of the current minimum if the tower were to be effectively detuned. That height can be calculated by MoM and is generally at 1/3 the height above ground. Current sampling transformers should be placed at the same physical positions for all towers close to the ATU feed-through insulators. The pattern is then set to the theoretical parameters as read by the antenna monitor with any appropriate offsets as calculated with the MoM model.
Dawson summed up his presentation by citing the various MiniNec and NEC resource tools that can be used to construct the reference model and theoretical solutions for comparison to field measurements.
MiniNec Broadcast Professional for Windows by J.W. Rockway and J.C. Logan is the current MiniNec version used by many practitioners and consultants. Versions of NEC, particularly NEC2, can be used; and the book “Basic NEC” by J.L. Smith is a good reference.
There are a number of other moment method programs that can be use for AM array analysis, but they are generally designed for other purposes and can be quite awkward to implement.
Making the Measurements
Ron Rackley, partner of du Treil, Lundin & Rackley, focused his presentation on how to properly make all of the required field measurements to support an application for license using the MoM modeling rules.
He first outlined the required list of system components that need to be measured. He advised that such measurements are most easily made with a network analyzer but can also be made with a calibrated impedance bridge.
ATU output impedance measurements are made generally at the ATU output port with the other towers’ ATU output ports open circuited or shorted to ground to calibrate the MoM tower model and base conditions. Exact sampling line lengths and characteristic impedances are determined to each tower with the sampling devices disconnected and reference impedance measurements are made with the sampling devices connected. The sampling lines can be used to expedite ATU output impedance measurements at each tower base from the antenna monitor location if a network analyzer is used. Each base sampling device must be measured individually and verified to comply within the manufacturer’s rated tolerance.
He discussed the various real-world influences that can alter actual tower base measurements, including the added shunt reactances of iso-couplers, lighting and sampling isolation coils and Austin ring transformers. The series inductive reactance of each feed line and lightning loop has to be included in the model. Impedance bridge test lead impedances, which can have values like 1 ohm R and 1.5 ohm X sub L with the test clips shorted also have to be accounted for to set the actual zero ohms reference. Dual modeling of the base hookup circuits and the tower base impedances must be done to arrive at the best overall match with the actual measurements using an iterative analysis technique.
Rackley detailed how precise sampling system calibration is done. Each sampling line is swept over a range of spectrum to determine adjacent zero crossing frequencies near the carrier frequency of the station. Impedance zeroes occur at each odd multiple of 90 degree line length. The 90 degree multiple closest to the operating frequency is used for reference in determining the carrier frequency length to compensate for delay distortion effects along the lines at AM frequencies.
Since the measured impedance magnitude of a line will equal its characteristic impedance every odd multiple of 1/8 wavelength (45 degrees), 45 degree offsets are used to calculate the real impedance values of the sampling lines. A carefully constructed sampling system should yield line lengths essentially identical within a few tenths of a degree and a few tenths of an ohm for both resistance and reactance.
Sampling loops all have to be physically identical. When connected to their sampling lines under measurement, reflection coefficients can be calculated and used to determine line length differences. Fortunately the sampling lines do not have to be pulled off the loops for the required biennial certification tests. Sampling toroid transformers can be measured side-by-side and verified in compliance using either a network analyzer or the antenna monitor itself. Both phase and magnitude components must be measured. Measurements must be made of the impedances looking into the antenna monitor ends of the sampling lines at carrier frequency with the sampling devices connected. When toroid current transformers are used, the impedances will be in the neighborhood of 50 ohms resistance if the toroid pickup units are in good condition, as they are internally terminated. Sampling loops terminate the lines with the loop inductance rather than 50 ohms, so significant mismatches will be found looking back into them. Sample loop impedance measurements must be avoided with magnitudes greater than 200 ohms. The frequency for sample loop measurements may be offset from carrier frequency to meet this requirement.
After the sampling and antenna monitoring system components are measured and fully documented, a MoM adjusted pattern must also be measured in the field with a minimal number of actual field intensity readings. Three points must be chosen and measured on each minima and each maxima radial of the pattern.
Rackley clarified that these are not “monitor points” that must be maintained or regularly checked in the traditional sense, but merely markers for future reference should changes or interference issues arise. In practice, the FCC appears to only require such measurements on the minima radials, a.k.a. those that would be MP radials in a conventional proof.
The final measurement needing certification in a MoM proof is the verification of the as-built array physical geometry or tower locations. A land surveyor can make such measurements, typically for about $1,500 to $2,000. No error tolerances are specified in the new rules but preliminary indications are that tower locations within 1.5 electrical degrees of Cartesian coordinate references will be acceptable.
A Struggle in New England
Clear Channel’s John Warner chronicled the challenge of building the country’s first AM DA system completed and licensed using MoM rules this February.
Champion Broadcasting host station WUNR (1600 kHz, 20/5 kW DA-2) collaborated with Clear Channel’s WKOX (1200 kHz, 50 kW DA-2) and Beasley’s WRCA (1330 kHz, 25/17 kW DA-2) to finish this long-awaited undertaking. All three stations are licensed to Boston suburbs and beam their major lobes over the metro area.
The project presented a host of unusual complications faced by few who have ever built an AM DA, let alone a high-power multiplex. The first hurdle was winning local approval to build the five-tower tri-plexed system over the objections of wealthy neighbors. That process consumed almost eight years.
The blue spotted salamander played a major role in this saga. Before building permits could be granted, local environmentalists forced a five-year study of the life cycle of the reptilian rascals, who populated the nearby woods. They wanted to determine if radio tower construction might interfere. Fortunately the little guys didn’t seem to mind.
Another imposition by the locals was the requirement the transmitter building had to be rebuilt to look like a residential house but at the same overall footprint size. This required Warner to design and install a compact two-story Kintronic phasor layout to serve WKOX.
Yet another challenge was dealing with a three-foot water table that partially flooded portions of the ground field and also made construction of new towers rather interesting. Warner was compelled to haul in many tons of rock to build roadbeds to each tower base location. But then he had to remove all of the rock after construction was complete. LTUs and tower base piers all were raised above the statutorily mandated flood plain for flood protection.
The neighborhood on three sides of the site was older mature construction with lots of overhead power and telephone lines. The deep null side of all six patterns was immediately adjacent to a huge swamp with no access. Warner realized that along with battling the wild variations of New England seasonal effect, doing a traditional field proof in this environment would be very problematic and could be extraordinarily costly. Instead, he seized upon the new MoM Rules.
Tuning up the three station DAs and performing all of the required system measurements for the very first MoM proof filing with the FCC turned out to be straightforward, with virtually no complications. To minimize seasonal effect variation, Warner picked his three field point measurements on each minima and maxima radial between 1.5 and 4.5 km from the site.
Warner believes the project saved some $250,000 by not having to do a field proof.
The panel answered questions about MoM proofing and discussed classic cases in which attempts to license DAs in difficult environments proved almost impossible.
Reradiation problems that cannot be resolved or controlled plus surrounding terrain or access complications have always challenged AM DA proofs but they are becoming more and more common. Untold millions have been spent on stations facing these issues to try to convert long-running STAs to final licenses. Many were probably not worth the amount of the investment.
All the panelists agreed that MoM proofing gives most AM broadcasters faced with moving and/or maintaining DAs in difficult environments an additional licensing option.
Rackley observed that consultants cannot even predict or estimate the hard costs that may be needed to complete traditional proofs in such scenarios. The costs of doing a MoM proof, he said, are almost all predictable, and in most cases, much more reasonable that doing traditional field proofs.
The panelists concurred that tuning a DA with MoM instead of traditional field measurements and ground wave analysis will more likely produce a pattern that more closely matches the theoretical design, especially for vertical angle radiation and far field protection considerations.
The author is Radio World technical adviser.