ERI Lone Holdout on FM Directional Antenna Modeling

Broadcast engineers and equipment manufacturers commenting in docket 21-422 all endorsed an FCC proposal to allow the use of computer models to verify the patterns for FM directional antennas. The one exception was antenna manufacturer ERI (Electronics Research Inc). “You’re going to have antennas that don’t work the way they’re represented,” said ERI CEO Tom Silliman in an interview Friday. Said engineering firm Meintel Sgrignoli, echoing most commenters: “It is our experience that various computer modeling software programs do an excellent job of predicting accurate antenna pattern data."

Under current rules, stations verifying the patterns for FM directional antennas must either build a full-size mock-up of the antenna or a scale model. ERI is the only manufacturer with the infrastructure -- a 50-acre testing range -- to test full-size antennas, Silliman said. He said the company isn’t concerned about a loss of business but about antennas that don’t perform as modeled and about potential interference. Full-scale modeling is a relatively small part of its business, Silliman said. Incorrectly modeled antennas could lead to “protracted and continuous interference disputes” between full-power FM stations, ERI said.

“We have the ability to do the computer models and we do them, but it's our starting point, not our finishing point,” said ERI Vice President-Marketing Bill Harland. The petition that prompted the FCC’s proposal was a joint effort from Dielectric, Jampro Antennas, Shively Labs, Radio Frequency Systems and the Educational Media Foundation. The petition argued that computer modeling is less costly (see 2111160039). Silliman said ERI was invited to join on the petition but declined. Engineer Merrill Weiss, who prepared the petition, confirmed in an interview that ERI was invited.

The examples of an FM directional antenna with a pattern verified by software don’t reflect the construction of most FM directional antennas, and ERI has found inaccuracies in such models compared with range testing, Silliman said. Many commenters said the computer models are more accurate than physical tests on a range. “The practicalities of mechanical modeling at full scale and/or quarter-scale generally force human compromises,” said Cesium Communications.

Nearly all commenters disagreed with ERI about computer modeling reliability. The use of computer modeling rather than the physical modeling and measurement for directional antennas is “one of the very best proposals to come out of the FCC in years,” said Cesium. “Computer modeling of DA patterns will definitely help in the overall verification process,” said low-power FM organization REC Networks. Software programs “do an excellent job of predicting antenna pattern data accurately,” said broadcast technical consultant Kevin Fitzgerald.

Many companies said that for models to be accurate, they need to be performed correctly. “Patterns produced by individuals that lack that expertise and intimate product knowledge may not be compliant,” said Shively, one of the petitioners. Computer models should be performed only by antenna manufacturers, Shively said. The manufacturer has historic data to know how a specific radiator behaves in a particular environment,” said Communications Technologies. Shively wants “a period of discovery and study to determine the best methods.” Engineering firm Cohen Dippell said the agency should consider a trial period followed by a Further NPRM to assess how computer modeling has fared.

Commenters disagreed whether the FCC should require specific software for FM directional antenna models. “The concept is to have a standard and avoid the very pitfalls and large variables that today’s 'invent your own range' methodology allows,” said Cesium. REC Networks said allowing multiple software options could mean savings for smaller broadcasters. Requiring specific software would be a barrier to innovation or technological upgrades, said several commenters. “The Commission should not create monopolies,” said Radiotechniques broadcast engineer Edward Schober. “Permitting only specific softwares by individual manufacturers should not be codified in regulation.”