Two years ago, in the Feb. 2, 2005 RF Technology column, I wrote about John Magliacane’s (KD2BD) excellent propagation program SPLAT!
(See “Doug Lung on RF” at www.tvtechnology.com/features/On-RF/).
John noticed the column and asked for my help in adding antenna patterns to the program. I provided some of the basic information and John did the complicated work. Version 1.2.0 of SPLAT! was released in late December 2006 and includes the ability not only to add azimuth and elevation antenna patterns to the analysis, but to include mechanical beam-tilt as well.
STRM DATA
John also added the ability to use SRTM (shuttle radar terrain mapping) mission elevation data. As I pointed out in my July 6, 2005 column, this data allows more accurate Longley-Rice studies by including manmade structures and forests in determining elevation and potential obstructions. That column is also available on the TV Technology Web site.
(click thumbnail)Fig. 1: Path loss from the Richard Tower site at West Orange, N.J., based on the USGS elevation data.Fig. 1 shows the path loss from the Richland Tower site at West Orange, N.J., based on the USGS elevation data.
Fig. 2 shows the same facility, but using the SRTM elevation data, which includes the impact of the New York skyline. East of Manhattan, several areas have much greater attenuation.
If you look closely, you can even see the impact of individual buildings (or clusters of buildings) on the path loss. Richland Tower is planning to install an antenna on the Bloomberg building at 731 Lexington for use with an on-channel synchronized transmitter to fill in coverage in this area.
Download the high-resolution images covering a much wider area at www.xmtr.com/splat/ and switch between them. This will allow you to see the extra loss due to manmade structures and forests that USGS-based Longley-Rice studies don’t include. (Before you get too excited about SRTM, read the caveats in my July 2005 column.)
Fig. 3 shows the topography of the New York City area based on the SRTM elevation data. I plotted this using SPLAT!, which displays topography using a logarithmic scale. Higher elevations show up as lighter colored pixels. As you can see, the building heights appear to drop off more slowly to the west than to the east. While SRTM data may include some errors due to poor radar reflections, the final result should be more accurate than with the standard topographic data alone.
(click thumbnail)Fig. 2: Richland Tower site at West Orange, N.J., using the SRTM elevation data.When creating the SPLAT! terrain files, the program gives you the option to use USGS digital elevation model terrain data when there is a void in the SRTM data. When USGS DEM terrain data isn’t included for a location, SPLAT! uses the average of surrounding sites to generate it. That’s how the maps here were generated.
While obvious in hindsight, I didn’t realize until I compared field test data with signal levels predicted using SRTM elevations that if the elevation of a location is based on roof-top height and, for example, a 30-foot antenna height is used, the receive antenna will be considered to be 30 feet above the roof top—the high point of the radar reflection! SPLAT! allows you to enter elevations that override the SRTM (or USGS) data, which should help correct for that.
There are a few other things to keep in mind when using SPLAT! Since the terrain data is based on WGS-84 or NAD83 datum, the NAD27 coordinates used by the FCC for broadcast facilities have to be converted to NAD83. See www.ngs.noaa.gov/TOOLS/Nadcon/Nadcon.html for a program and online converter, or you can look at the FCC tower registration, which has the NAD83 coordinates. While the difference has little impact on flat terrain, in mountainous areas, the error can put the transmitter on the wrong side of a ridge!
Terrain database files are required for the area you are studying. They can be downloaded at no charge. The SPLAT! Web site at www.qsl.net/kd2bd/splat.html shows where to obtain this data for almost any location on the Earth. If you run the program without the terrain data, it will list the areas needed. If terrain data isn’t available, it uses sea level as the elevation.
(click thumbnail)Fig. 3: Topography of the New York City area based on the SRTM elevation data.If you want USGS data to fill in holes in the SRTM data or for comparisons, it may not be obvious which files you need due to the way the files on the USGS site are named. Visit store.usgs.gov/ for a catalog. Click on “Enter the store” to load a Web site that allows browsing map listings. Click on 1:250,000 in the column on the left. When you go to the ftp site listed on the SPLAT! page, note there are “east” and “west” sections for most maps.
COMPENSATING WITH TILT
Until now, without expensive propagation programs, it was difficult to analyze the impact of electrical and mechanical beam-tilt (EBT and MBT) on coverage. Tilt is important at high elevation sites, where the desired service area is likely to be well below the radio horizon.
I’ve found several cases where stations were not putting the strongest signals possible over the areas with the greatest concentration of population due to less than optimum amounts and direction of beam-tilt.
SPLAT! uses simple text files consisting of the angle and the relative field (zero to 1). Both the Dielectric and ERI antenna system planning software can generate the data, although you may need to run it through a spreadsheet (using text import/export) to remove unwanted columns and format it for SPLAT!
While SPLAT! won’t substitute for a program that allows interference analysis and coverage population counts, it can give a quick indication of whether coverage is optimized. Fig. 4 shows the Longley-Rice path loss from the KVEA Channel 52 auxiliary broadcast facility on Mount Wilson.
The tilt had to be adjusted to stay within the contour of the main facility on Mount Harvard. Notice the reduction in path loss around Montclair (increased signal strength), where the main elevation beam of the antenna hits the ground. Extra tilt at 130 degrees was necessary to protect Channel 52 DTV in Palm Springs, Calif. This map was based on a standard Dielectric TFU-24-DSB-M antenna with 2.0 degrees EBT and 1.0 degree MBT at 130 degrees rather than the slightly different custom antenna actually used.
(click thumbnail)Fig. 4: Longley-Rice path loss from the KVEA Channel 52 auxiliary broadcast facility on Mount Wilson, north of Los Angeles.One of the features John has added in SPLAT Version 1.2 is the ability to output path loss data to a text file. This file contains a listing of each point covered in the study by latitude and longitude. The data for each point includes the azimuth angle and elevation angle to the first obstruction and the path loss without considering the antenna pattern.
Since it can take awhile to run a large coverage study, this file can be loaded into SPLAT! to avoid the need to recalculate the Longley-Rice path loss when studying the same location with a different antenna pattern, orientation or mechanical tilt. This should make it easier to zero in on the optimum antenna configuration.
What if you want to calculate field strength rather than path loss? John has added the ability to include field strength, with the antenna pattern taken into account, in the output data file. This will be a big file!
He’s working on adding the ability to plot field strength on the map. When John updates the program to add this feature, I’ll include a note about it in my weekly RF Report newsletter, available at www.tvtechnology.com.
John continues to update the program to improve accuracy and add features. You will need to have Linux on your computer to run this software. It’s not as difficult to add as you may think if you have a few extra gigabytes on your hard drive. Try MEPIS Linux at www.mepis.org for its combination of features, ease of installation and ease of customizing or one of the Ubuntu (www.ubuntu.com) variants.
NAB RF EXPECTATIONS
For an idea of what I’ll be looking for at NAB2007, read last year’s pre-NAB column at www.tvtechnology.com/features/On-RF/. Not much has changed!
One new area I expect will attract a lot of attention is mobile ATSC TV. Samsung demonstrated A-VSB at CES in January. It will be interesting to see if they and Rohde & Schwarz show anything different at NAB. Harris will be showing its mobile ATSC DTV solution as well. Harris has released little technical data on its system.
In the microwave area, we will finally see a demonstration of digital return link for ENG at the ATSC DTV Hot Spot.
As usual, I’ll be looking for useful, inexpensive test equipment. While it’s probably too early to expect to see test equipment for the proposed mobile ATSC standards, test equipment for distributed transmission systems will be shown.
Look for a report on what I found in my June RF Technology column!
