The Search for STL Bandwidth
Have you tried to get a new frequency in the 944–952 MHz aural STL band lately? How’d that work out for you?
Certainly there are places out in Middle America where frequencies are still available, but get anywhere near a city of any size and you can just about forget it.
By my count, there are 16 500 kHz channels available in that band, and the grid dish antennas we use for those frequencies aren’t all that tightly directional. Add to that all the reflections and scatter that you can get at 950 MHz and it’s not too easy to reuse frequencies. If a market has an “antenna farm” where many of its stations are located, you’ve got a single point of convergence and you’ll run out of frequencies in a hurry.
Now factor in HD Radio. Ideally you will have an Ethernet path to your transmitter site with enough bandwidth to carry the output of the exporter. Modern transmitter sites also have need of additional Ethernet bandwidth. Many transmitters these days have a built-in GUI/Web interface, remote control units operate over TCP/IP connections and more and more transmitter sites are being equipped with DVR/video surveillance systems that need an Ethernet connection back to the studio. It’s hard to do all that on a 500 kHz channel in the 950 MHz aural STL band.
Seven or eight years ago, I began looking for alternatives to traditional 950 MHz aural STL links. Because all our stations are in large markets, we have always struggled to find frequencies to operate on and, in many cases, have resorted to landline options such as T1. Those aren’t cheap, however, and aging local telco infrastructures plus increasing copper theft combined to reduce reliability. I had to find another way.
Wireless Internet was just getting off the ground in those days, and distribution/backhaul equipment was not hard to come by. I found that we could get a point-to-point Motorola Canopy BH system for about $2,400, including antennas. Such an unlicensed 5.7 GHz system would work well over a 20-mile path. We deployed quite a number of them, using Harris Intraplex multiplexers to transmit audio over the provided Ethernet bandwidth.
We used these systems for years, but as wireless Internet proliferated, so did the number of signals on the air. As that happened, interference levels rose and throughput suffered, eventually to the point that we didn’t have enough throughput for the Intraplex to reliably convey audio.
We changed from the Canopy BH systems to the Motorola PTP-400 and got by for a while with that frequency-hopping, dual-polarity system, but the interference eventually caught up with us. The handwriting was on the wall: We had to have an interference-protected, licensed link.
So it was that I began looking hard at licensed 11 and 18 GHz systems last year. We found that Dragonwave and Trango were two manufacturers of this gear, and it was reasonably priced. I was already familiar with the PCN frequency coordination process required by Part 101 of the FCC rules, so finding frequencies and filing applications was no problem. Or so I thought.
WAIT, GIVE IT BACK
In our Denver market, I ran head-on into the Department of Defense and the Table Mountain radio telescope on 18 GHz.
I had done all the coordination and received no objections from these government agencies, but eight months later the FCC called and told me that they were going to have to rescind the 18 GHz license grants I had already received because of late objections. Evidently one of the frequencies I had chosen was too close to or coincident with something the DOD was using in the area (of course they won’t tell you what frequencies are in use), and the others might be used by E.T. to phone home. So it was back to the drawing board.
NOT THE LAST LINK
I should have started out on 11 GHz, because I didn’t have any trouble finding frequencies and neither the DOD nor the E.T. seekers care a whit about that band. I filed new FCC apps and got grants quickly. It didn’t take long to get the equipment in, and soon we had an operating link to one of our transmitter sites.
It should be noted that these links operate under Part 101 as fixed microwave links (“FXO” in FCC terminology). There is a prohibition in Part 101 on using FXOs as “the last RF link” to a broadcast transmitter. When the rules for this section were written, evidently they wanted broadcasters to stick to the BAS bands and not soak up a bunch of fixed microwave frequencies. So we had to find a way around this rule.
One option was to request a waiver. The FCC has, from time to time, granted such waivers upon a showing that (a) there are no Part 74 (BAS) frequencies available for the desired path, and (b) that the bandwidth required for HD Radio operation is not available with a traditional aural STL.
I did not request a waiver, however, because I found another workaround that is actually based in practicality. The Trango Apex microwave radios mount right on the back of the antenna. Power, control and through-data are supplied via a couple of Ethernet (Cat-5) cables. The practical problem is that the antenna is mounted 400 feet up a tower, which is itself 600 feet from the transmitter building. With the Ethernet limit for Cat-5 cable around 300 feet, this wasn’t going to work.
So instead of using a couple of boxes of Cat-5e and several switches along the way, I opted to use an unlicensed 5.7 GHz link between the tower and the transmitter building. I chose the Ubiquiti NanoBridge M5, which comes with a 1-foot dish antenna on each end and offers 100+ Mbps of throughput.
The NanoBridge was connected to the Apex up on the tower, and the other end of the NanoBridge was installed on the roof of the transmitter building. The distance between NanoBridge antennas is about 700 feet, meaning that we’re not worried about signal strength or interference on that short-hop link.
With this arrangement, the 5.7 GHz NanoBridge and not the 11 GHz Apex is the “last RF link” to the transmitter, thus complying with the Part 101 rule. To keep everything on the up-and-up, I included a statement explaining this in the Form 630 application.
The author assists the contractor with installation of the 3-foot studio antenna.
Installation was mechanically challenging — the 3- and 4-foot radome-equipped antennas are heavy! — but otherwise straightforward.
Hang the antenna on each end and attach the radio itself to the back of the antenna with the supplied fasteners. The circular waveguide input on the antenna mates right up to the waveguide output of the radio with just an O-ring to install on the connection. Connect the through-data and management Ethernet ports and fire it up.
There is a large red two-digit LED numeric display on the back of each radio that displays the RSSI (receive signal strength indication) in dBm (although a positive number is indicated, the display actually indicates a negative number). Adjust the elevation and azimuth of the antenna for the lowest number and the path is aligned.
On our link, the calculated receive signal was –40 dBm (EIRP + antenna gain – path loss + antenna gain), and that’s very close to the “40” that we saw on the display. The link has been in operation for several months now and the signal strength varies between –42 and –39 dBm.
This provides us with about 40 Mbps of throughput using a 10 MHz channel, plenty for our application. We continue to use Harris Intraplex multiplexers to transport audio over the Apex system. In addition to 12 channels of audio, we’re also transporting Internet/e-mail, webcam, Amb-OS files, remote control, HD Radio program service data (PSD) and more. The effect is the same as if the transmitter site were connected right into the LAN subnet switch at the studio. I guess in a way it is.
The tower crew aligns the antenna on the far end of the path.
I have filed and received grants for two more 11 GHz links and have one of the systems ordered and on the way. By the time you read this I will likely have both these systems online and operating.
One more thing I should mention: The antennas for these units are heavy and have significant wind loading. A structural analysis is a must before hanging one of these things on your tower.
In one of our installations, I am going to have to increase the size of the guy wire at one level to maintain a proper safety factor. Don’t assume that the tower can handle it and hoist the thing up. Make certain.
As inadequate bandwidths and scarce frequencies continue to be the rule on the Part 74 aural STL BAS band, radio engineers are going to have to find other ways to get the signals from the studio to the transmitter site and back. Fixed microwave links aren’t right for every application, but they do offer an alternative in some cases.
Cris Alexander is the director of engineering at Crawford Broadcasting and a recent recipient of SBE’s Broadcast Engineer of the Year award.