If you were to consider for a moment the question: "What is the most important part of the transmission system?" likely the first answer you'd come up with is "the transmitter." That's not really true, though, is it? The STL, the transmitter, transmission lines, and the antenna are all equally important. With the failure of any one of those, your station will be off the air. I'll focus on the FM antenna in this installment, and in doing so I'll consider three separate scenarios that would prompt you to buy a new one: the installation of a new main antenna, the installation of a broad-band antenna as part of a back-up system, and a low-power booster (or translator) antenna.
Photo by Martyn Gregory, Shively Labs.
There are many questions to be answered before purchasing an antenna; some are the same no matter which of the three scenarios are being considered, and some are of course specific to the scenario. Here are some common questions:
■ What is the required effective radiated power (ERP)?
■ What is the available transmitter power output (TPO)?
■ What are the necessary downward radiation characteristics?
■ What is the necessary polarity (i.e., circular, vertical, horizontal)?
■ What is the available space on the structure?
■ How high will the antenna be above the ground?
■ Will the structure have the strength to hold the antenna in all conditions?
So, as you can see, some amount of research needs to be done even before you issue any kind of a request for quotations. Let's consider these questions in more detail.
Replacing an old antenna
The ERP vs. TPO question is easy to answer in the event that you are simply replacing an old antenna, and not changing transmitters. As an example: If you needed 17.5kW TPO for 50kW of ERP (with the old antenna power gain of three, and accounting for transmission line loss) then likely you'll be getting a new antenna with the same amount of power gain and similar physical characteristics, such as the amount of space needed on the tower. In this event you would still want to make sure that the weight and wind-load characteristics of the new antenna are acceptable for the tower, and that the downward radiation characteristics are the same or better than the old antenna.
- continued on page 2
Acquiring a new antenna
The problem at hand is much more involved if you are buying a new antenna for a new FM installation. All seven of the previous questions need to be answered, and perhaps even more. There's also a philosophical question regarding the high TPO/low antenna gain versus the high antenna gain/low TPO question. Some examples will illustrate my point.
In San Francisco there were two FM stations that transmitted from nearly the same location on Sutro tower. One used a single-bay antenna with around 20kW of TPO; the other used a two-bay antenna with half of the TPO. On the Empire State Building, there are two master antennas; the one known as the mini-master has half the gain of the master FM antenna, and thus requires more TPO. (I'm sure there are many other examples like this scattered around the country.) I have never seen any studies, technical or based on ratings/revenue, that show that stations transmitting in the middle of town (such as those at Sutro Tower or the Empire State Building) benefit from the lower gain/higher TPO model. On the other hand, if your new FM station is to be located outside of town (as most are) and the majority of the audience is effectively at the horizon (plus or minus a few degrees) it doesn't seem logical to use the lower gain/higher TPO model, since a lot of power will be sent at low (and high) angles of elevation, likely in areas that have no potential listeners.
Another important consideration for a new antenna installation is its downward radiation characteristic -- meaning directly below, on the ground. If the available antenna location is somewhat low, you may opt for a 1/2-wave spaced antenna (as opposed to a full-wave spaced antenna). The former has less gain than does the latter, which would compel you to go with either more elements (to recover the power gain) or to use a higher TPO.
During my career, circular polarization has been the norm, and there is one compelling reason for it: Receive antennas come in all flavors and polarities. It would seem that most cars have vertical antennas but there are many that have horizontal antennas embedded in glass; and there are plenty of opportunities for user-installed antennas to be horizontal by chance (think of a folded-dipole sitting behind someone's desk). It's not surprising that circular polarization is used by just about every main antenna I've ever encountered. It's the best way to provide optimum reception in the vast majority of receive antennas in the field. There are some instances where vertical polarity seems like a reasonable compromise though.
The available space on a tower, and its height above ground, are of course related questions. These are more of a consideration when building a new FM site. Clearly the space available on a tower at a potential site will dictate many of the characteristics of the antenna that ends up being chosen.
Finally, it makes sense to engage the services of a structural engineer (who is a registered professional engineer) when determining if a tower will safely hold whatever antenna it is you end up choosing. Just about any site lessor is going to make you do this prior to the installation of said antenna; and if he doesn't, you should certainly be concerned about the mechanical viability of the tower in question.
- continued on page 3
Hypothetical but typical scenarios
So now that we've discussed the generic questions regarding the purchase of a new antenna, let's look at a couple of hypothetical installations. First, let's say you are participating in the construction of a brand new FM station. The allocation aspects are outside the scope of this article; so let's just say that you've found a potential site (a mountain top outside of town) with two potential tower locations. At one, your ERP will be 55kW; at the other (which has a slightly lower height above average terrain, or HAAT) your ERP will be 57kW. At the higher tower location, 70' are potentially available for your antenna; at the lower location, 63' are potentially available.
I'll present multiple antenna products in this article, and I'll start with ERI. When visiting the website you'll notice an antenna size calculator as well as a chart of antenna gain figures listed by the number of elements and antenna spacing (full or half-wave) for the various antenna types. Starting with the 55kW number, you decide (for starters) to look at a six-bay, full-wave spaced antenna with circular polarization. From the website, you get the antenna power gain figure (3.3028). Divide the ERP by the power gain: 55/3.3028 = 16.7kW (rounding up). The SHP-6 will give a large margin of safety on the input power (16.7kW compared to the 39kW input power rating). 50Ω 3" Heliax will also work in this application; in our hypothetical situation we find we need 225' of the line to reach the transmitter. According to ERI, the efficiency of this line length (at 93.3MHz, as an example) is 93.2 percent; so, excluding any loss attributed to connectors, our TPO is: 16.7/.932 = 17.92kW (again rounding up). Clearly a new 20kW transmitter will be adequate in this case. Also you can see on the website that this antenna requires an aperture of 65', and fortunately that's available on the higher antenna location.
However, it turns out that the lessor for the lower location will charge less, so it makes sense to see if that location will work. As I mentioned earlier, only 63' are available, so the full-wave spaced antenna won't fit. Referring back to the online calculator, we see that the SHP-10AC-HW will give us close to the same power gain. Doing our TPO calculation again: 57/3.126 = 18.3kW (rounding up) is needed at the antenna input. We'll use the same transmission line type and length, so our necessary TPO is: 18.3/.932 = 19.65kW (again rounding up). The aperture necessary for this antenna is 60', so it will fit.
At this point you'll need to decide if the extra height is worth the extra monthly expense. You'll have to consider the cost of the proposed 1/2-wave-spaced antenna for the lower location as well -- it'll likely be more money (since it's a lot more material and there's more labor involved in making it). After deciding on the preferred location, have your FCC consultant calculate the RF levels directly below the antenna to make sure they do not exceed the occupational limits (referring to OET bulletin 65) and have your structural engineering consultant determine whether or not the tower can safely hold the antenna.
Jampro offers the JHPC line of FM antennas, and that is obviously worth looking at in this same application. Its website gives antenna gain figures, so you can do the same set of calculations as shown above. Likewise, Shively Labs offers the 6810 high-power series, which can be seen in detail on its website. Shively offers all the details about this antenna series -- the gain figures, along with size and weight, to help you do all the same calculations that were described earlier. Dielectric also offers its DCR-C line.
continued on page 4
Let's look at our second hypothetical scenario. In this case our project is to come up with an antenna that will be used to combine four FM stations in to a common antenna, for purposes of a backup transmitter site. This is a low-to-mid power application; all four stations will have an ERP of around 7kW.
Shively 6832. Photo: Mike Fitzpatrick/NECRAT.US
In this application I'll use the Shively 6832 series. The specs from the website show that the six-bay array has a power gain of 3.024 at 98MHz. (I spoke with Shively regarding this application and was told that the power needs to be limited to 2.5kW per station.) The antenna input is an EIA 1-5/8" flange. A quick check of Andrew HJ7-50A (1-5/8" coax) indicates that it would handle this power at a single frequency. I would seek advice from the antenna manufacturer in this case to be sure that the connector size and the coax can handle the four combined RF signals. Research indicates that the coax length in this case is 150', so again referring to the HJ7-50 data, the efficiency of this line is 93.3 percent at 98MHz. So, after running the same set of calculations we did earlier, we find that we'll need about 2.5kW of TPO. (Since this is a wide-band system, the power we need will vary slightly with frequency.) Again, using data supplied on the website, you can see the necessary aperture is 40.8' and that the recommended space needed is 60.8'. You can also take the weight and wind-load figures to help your structural engineering consultant determine whether or not the tower will safely hold this antenna.
Clearly if you were to plan a system such as this you would need a four-port combiner. One possibility is the Shively 2640-04-1/1. 5kW per input (max) and 15kW through the output (max) seems appropriate in this scenario.
I've highlighted Shively in this hypothetical scenario, but there are other options, of course. As one example, you could opt for the Jampro JPCB in a multiple-element array; or (although it would be overkill for this power level) you could consider a Dielectric DCR-S series wide-band antenna array, or the ERI Rototiller Axiom series.
Jampro makes combiners as well, such as the RCCC-X9X for this power level. ERI is also very well known for its combiners. Again, at this power level, you could opt for the 955-6 (if stations are spaced by at least 1MHz) or the 955-8 if the spacing is down to 800kHz. Options with Dielectric include the DFC14005CIF (for 800kHz spacing). Myat makes combiners as well; at this power level and with 800kHz spacing, an option would be the CSFMBP8400CZ, which is a four-cavity filter (per station) constant-impedance combiner, for 8kW per station.
- continued on page 5
Our third scenario involves a different (though still common) type of antenna -- the log-periodic array. This type of antenna is an option in (Part 74) booster (or translator) applications, not only because of the gain, but the directivity as well. Often times in booster applications, you will seek to pinpoint the RF power in the desired direction, while keeping it from going where you don't want it to go (to the extent possible). Probably the most well known manufacturer of this antenna type is Scala (a division of Kathrein) and its version is known as the CL-FM. You can mount the CL-FM in either polarity easily enough, and perhaps even more importantly, you can design multiple CL-FM antenna arrays (with the help of Scala Engineering) for custom antenna patterns (varying the power between elements, as well as the phase angle between elements). The CL-FM also comes in a 1kW input version. Scala will also provide power dividers for multi-element arrays.
Just as with the other antennas we've talked about, you'll need to perform the same set of calculations before using the CL-FM. Power gain is five, and Scala's specifications give the weight and wind-load.
As you might expect, Scala is not the only manufacturer of log-periodic antennas: Shively offers its 6025 which can actually handle up to 5kW (7/8" EIA flange on the input); and, Jampro makes a log-periodic that it calls JAVA, also with 7/8" EIA flange, rated at 5kW max.
I've pointed out example products from the big four antenna manufacturers but it's important to remember that there are even more makes, especially in the lower-power categories. While pricing antennas, take a look at these other manufacturers as well: Propagation Systems Inc., Systems With Reliability, Armstrong, OMB, and Scala's parent company, Kathrein.
It's difficult to over-emphasize the importance of the passive elements of a transmission system. There can be many variables that need to be considered when deciding upon the correct one, as we have seen. The antenna is the final, vital link between you and your audience and as such, you should put as much attention and financial resources that you can towards its selection, purchase and installation.
Irwin is RF engineer/project manager for Clear Channel Los Angeles. Contact him at firstname.lastname@example.org.