This month I'll finish the overview of UHF antennas I started in December's column. In addition to the overview, I'll describe how Andrew Corporation tests their UHF TV broadcast antennas without a test range and offer some observations on beam steering and differential gain in antennas. Wireless cable is a rapidly growing area of TV broadcast transmission. Phone companies (PacTel, Nynex and Bell Atlantic to name three) are purchasing wireless cable systems to offer video programming to the home. The 1995 Private & Wireless show in Miami showed how far this service has come. I'll update you on the latest developments.
Last month I discussed UHF panel antennas and described the trade-offs involved in using them. Top mounted UHF slot antennas are much less complex to install with less chance for error. They can also provide excellent omni-directional coverage. Based on my observations, I'd guess that top mounted slot antennas are the most common full power UHF antennas in use in the U.S. today. Many UHF stations started the same way KVEA (then KBSC) did: with a General Electric "Zig- Zag" top mounted panel array then, later, a top mounted RCA "TFU" slot antenna as transmitter power levels increased. The G.E. "Zig-Zags" were not noted for their coverage -- I heard one story where a station got better coverage from a single dipole emergency antenna after their "Zig-Zag" burned up. Perhaps other experiences like this are why panel antennas, in spite of their features, have not been as widely used in the United States as in other countries.
Much of what I have to say about top mounted UHF slot antennas also applies to side mounted slot antennas. I'll cover these common areas first. Unlike panel antennas, which can be designed to cover the entire UHF band, UHF slot antennas are restricted to a much narrower frequency band. Panel antennas can be made of broad-band dipoles or rings. Slots, on the other hand, are much more constrained physically and it just isn't possible to provide a low return loss over several channels using a slot.
At one time conventional end fed slot array antennas were accused of "beam steering". That is, the elevation pattern of the antenna varied with frequency, causing the ratios between the visual carrier, chroma subcarrier and aural carrier to vary depending on how far you were from the antenna. While this complaint may have been true for some very high gain slot array antennas, it isn't true today. When working with Geza Dienes (now retired) from Andrew Corporation designing the WSCV antenna for Miami I asked him how he avoided "beam steering" with the slot antennas he designed. After showing me a few computer screens with various formulas and data, I understood he slightly offset the impedance of the individual slots so when combined the pattern remained stable across the channel bandwidth.
While I was at Andrew checking out the test results of KVEA's antenna, I remembered my conversation with Geza. I asked Kerry Cozad if he could show me how much differential gain it had. He and Steve (the fellow working on the antenna) quickly took a set of readings from each of the slots and fed them into Andrew's VAX computer. (More on this later.) After a few tries the numbers came back. The worse case differential gain was 4.3 dB. It occurred above the radio horizon at a null in the pattern. Between 1 and 10 degrees below horizontal (the area that counts) the differential gain never exceeded 1.6 dB. After thinking about how the elevation pattern was often neglected in antenna designs five or ten years ago, I realized what most likely led to the worst "beam steering" complaints. If you look at the elevation plot of a slot antenna, you'll notice that without null fill the nulls are quite deep and quite sharp. Even a slight angular change with frequency in the elevation antenna pattern would cause large changes in the visual to chroma and aural carrier ratios. With heavy null fill, such as we employed with KVEA's (and WSCV's) antenna, even a relatively large angular change with frequency wouldn't make a significant difference in the carrier ratios in the portion of the elevation pattern below the horizon.
I mentioned earlier one of the major advantages of top mounted slot antennas was the ease of installation. Set it on top the tower and bolt it down. Are there any disadvantages to a top mount slot antenna? The answer, of course, is yes. In order to be self supporting, the antenna must be made of galvanized steel. For a large antenna, the only way to make the slot is by flame cutting. This makes it much more difficult to cut the slot to precise dimensions. Once it is cut, the coat of galvanizing required to protect the steel reduces the precision even more. Before I get a lot of angry e-mail, please note I'm not saying it isn't possible to make a good top mounted slot antenna. The number of such antennas in successful use around the country today prove it is. One final point is worth consideration. A top mounted antenna of any sort places a huge load on the top of the tower. Seeing bolts loosen on the pipe mount supporting an antenna on top a 150' tower in high wind many years ago gave me a lot of respect for that force! The same forces put a lot of stress on the lower portion of the antenna, at a location where slots in end fed antennas have to take the most power. Again, none of these factors have to be a problem. Just make sure the structural engineer designing the tower and evaluating the antenna design is familiar with the conditions at your tower site.
If top mount antennas work so well, why would anyone consider a side mount antenna? First, if you are sharing a tower with other stations, you may not have a choice. Second, if you don't need omni-directional coverage a side mount antenna can offer more options for pattern shaping. Third, side mount antennas can be constructed using less material and the material doesn't have to be as heavy as steel. Andrew uses brass for their side mounted TRASAR (tm) antennas. They look quite nice after cleaning. This lighter material can be machined, allowing for tighter slot tolerances and more stable parameters through the manufacturing process. While I don't believe its wise to sacrifice performance for cost, if a directional pattern can be used the side mounted antenna will not only be less expensive than an equivalent top mounted antenna, it will weight less and the required tower strength and cost may also be less.
When I was considering an antenna for KVEA, I added another reason to this list. California is known for its earthquakes. I've never been on a tower during an earthquake, but I have been on the ground talking on the radio to someone who was during small quakes and I heard, in somewhat more explicit language, the amount of motion at the top of a 150' tower is significant! Combined with the high winds that can occur during Santa Ana weather conditions, I wanted a solid, rigid mount for KVEA's antenna, especially with the precise mechanical beam tilt we planned to use. A five foot triangular section behind the antenna provided that support. The area behind the tower has high mountains which block the signal, so a directional antenna made sense. It would focus the signal where it would do the most good and keep energy off the northern mountains which could cause ghosting. In this case, a side mount antenna saved money and allowed for a more stable mount which will help maintain signal quality throughout the life of the antenna.
As you can see, there is a place for side mounted slot antennas, top mounted slot antennas and panel antennas at UHF. Which one will work best for your station depends on the tower available, the pattern needed and the possibility of using the same antenna on multiple channels, whether NTSC or HDTV.
I alluded to Andrew's method of testing UHF TV antennas earlier. Before I leave this discussion, I think you will find it interesting. The conventional method of verifying antenna patterns is to haul the antenna out to a test range, put it on a turntable, and measure the pattern. I mentioned Harris' excellent antenna test range in my WME/SMPTE/SBE conference report two months ago. Andrew doesn't have such a test range at their facility, so another method had to be found to check the patterns. The azimuth pattern for a slot antenna can be determined with one slot or one set of slots for the antenna. At this size, an anechoic ("without echoes") test chamber can be used. If tower sections or other structures have to be included, scale models will work.
The elevation pattern is more difficult to measure. A typical UHF slot antenna may be over fifty feet long! Furthermore, the fields from the individual slots don't combine until they are some distance from the antenna. Without a test range like the one at Harris or the one at Dielectric Communication's facility in New Jersey, Andrew had to find a way to verify their pattern.
Their solution was to measure the individual phase and amplitude from each slot of the antenna. KVEA's antenna had thirty slots. Even though the vertical gain was only 25, the extra slots were needed because of the heavy null fill. A sophisticated computer program is able to synthesize the elevation pattern at any desired frequency from the samples taken at each port. As a result, not only is the over-all pattern known, but if any one slot is out of adjustment, it shows up quickly. The photo shows the probe used to make these measurements and how it is placed next to the antenna.
Does it work? The laws of physics and electromagnetics say it must. Just to be sure, Andrew told me they had verified the performance of their software with other antennas at Harris' test range in the Mississippi flood plain near Harris' Quincy Illinois plant.
Now that ATV is on the horizon and all but a few stations will be transmitting on UHF channels, I suspect there will be more interest in UHF antennas. Over the next few months I'll be taking a closer look at the types of UHF transmitting antennas and describing in more technical detail how they work.
While I was at Andrew Corporation checking out KVEA's antenna, I saw a lot of 2.5 GHz. antennas, both under construction and new designs in R&D. Kerry Cozad told me Andrew sold 350 of these antennas in the past year, mostly overseas, for wireless cable applications and interest was growing. A few years ago wireless cable was an industry with no clear future. Now, as I mentioned at the start of this column, telco's are purchasing wireless cable licenses on both coasts for video distribution in competition with wired cable. I spent an afternoon at the 1995 Private and Wireless show in Miami Beach looking at the technology behind this industry.
Wireless cable usually refers to systems operating in the 2.5 GHz. spectrum using conventional AM VSB transmission, the same as broadcast TV transmitters. A scrambling system is usually included to inhibit piracy. Recently some operators in rural areas have combined several LPTV channels in the TV broadcast band for wireless cable operations. One of the biggest will be in Alaska, using equipment supplied by Acrodyne Industries.
At the Private and Wireless Show I heard a speaker (I don't remember his name) announce that the first over the air commercial use of digital television wouldn't be by TV stations, it would be by wireless cable operators. This, for me, set the theme of the show. Digital compression can put six or more digital TV channels in one 6 MHz. wireless cable segment. This makes it possible for wireless operators with a limited amount of wireless spectrum to match the channel capacity of wired cable operators' analog plants. A recent press release from PacTel referred to a study that showed viewers preferred MPEG compressed video to VHS recorded video even at rates as low at 1.5 Megabits per second. At this rate a single wireless cable channel (between 20 and 27 Mbs. capacity depending on modulation method used) could carry over a dozen channels, even at the slowest rate. During the show, the data rates suggested for video ranged from 1.7 to 4.4 Mbs. for quality exceeding that available from VCRs.
When I wrote this column, the F.C.C. had not approved the use of digital modulation for wireless cable systems. At the show I didn't detect any concern that it wouldn't be approved. Issues such as power levels, technical specifications including out of band emissions and type acceptance requirements had to be worked out. Over all, it was agreed that digital would offer better coverage than analog at equal power levels. However, Bob Unetich of ITS pointed out that older transmitters would have to run at lower levels, perhaps a tenth of their NTSC power, to meet requirements for out of band products and intermodulation. I was disappointed to see that no standard modulation scheme was proposed. Some manufacturers were touting 16QAM or 64QAM, while others supported 2VSB, 4VSB or 8 VSB, the modulation used in the Grand Alliance ATV system. The main reason for not using 8VSB appeared to be that no royalties had to be paid to use 16 or 64QAM, while Zenith collected royalties on 8VSB based systems.
A group of non-competing manufacturers of products for wireless cable joined together to form the "Wireless Cable Digital Alliance". The manufacturers include companies familiar to TV broadcasters -- Andrew Corporation, California Amplifier, Emcee, Microwave Filter Company and Zenith. They did some testing at sites owned by another member, American Telecasting, to see how well digital transmission would work with a "real world" system. At the "Digital Update" session Pat McConnell from American Telecasting talked about the results. The tests used 2VSB and 8VSB modulation with Zenith receivers. At seven "challenge" sites where foliage blocked the signal, they had a 43 percent completion rate with the antenna at 50 feet using NTSC. With 8VSB digital, they were able to achieve an 87 percent completion rate with the antenna at 37 feet. Dropping the data rate and substituting 2VSB digital modulation allowed a 100% completion rate with an antenna height of only 24 feet. These tests were done at 2.5 GHz.
Broadcasters' should be able to learn from the efforts of wireless cable operators as they move to digital transmission. I plan to cover developments in wireless cable in this column as the technology develops. The next technical event for wireless cable operators is the 2nd Annual Wireless Cable Technical Symposium, February 3-5th, at the San Antonio Marriott Riverwalk in San Antonio, Texas. If you are interested in it, contact WCA Convention Services at (202) 452-7823. fax (202) 452-0041 or e-mail email@example.com. The Wireless Cable Association also has a web page at http://www.cais.com/wca.
Here are the addresses and info you'll need to contact me. The RF Page is at http://www.transmitter.com and contains, among other things, a complete listing and full text (some with graphics) of all my TV Technology RF columns since January 1993. The same articles are available from my FTP site at ftp.transmitter.com/pub/. You can E-mail me at firstname.lastname@example.org . You can also fax me at 305-884-9661 or phone me after 6 PM eastern time (when things quiet down a bit) at 305-884- 9664. Both numbers are at the Miami Telemundo office, so expect a delay in a response if I'm traveling. My mail service address is 2265 Westwood Blvd., Suite 553, Los Angeles, CA 90064. Because I'm often traveling, if time is critical (response needed in less then ten weeks) please contact me for a local address before sending items by mail. Also, remember that I'm no longer able to offer the software programs I've written and mentioned in these columns by disk. They are available for downloading at by FTP site or through the R.F. Page, my Web site. As before, the programs are free.
Return to Doug Lung's TV Technology RF Columns 1996
or The RF Page @ www.transmitter.com
Copyright ©1995, 1996, 1997 H. Douglas Lung. All rights reserved