If the hopes and expectations of iBiquity Digital Corp. come to pass, engineers will be scrambling this time next year to prepare their facilities for the in-band, on-channel digital broadcasting revolution.
Before we get to that point, a little planning is in order.
Information for this article came primarily from presentation materials provided by Jeff Detweiler of iBiquity, at a meeting of the Washington, D.C., chapter of the Society of Broadcast Engineers.
Additional information came from experience gained by participating in DAB tests at WHFS(FM) in Annapolis, Md., and related discussions with several transmitter manufacturers, including Nautel, QEI and Harris. Pat Malley, iBiquity’s primary installer for the DAB test sites, gave useful insights as well.
Station engineers looking to prepare their facilities for DAB should keep these factors in mind:
- 44.1 KHz sample rate should be the plant standard;
- Air control rooms need to be able to monitor "pre-delay" audio;
- UPS/power protection is required for CPU equipment in the air chain;
- The air chain audio processor must be moved to the transmitter;
- The STL may require an upgrade to pass AES signal;
- Separate processing for digital and analog audio is required;
- The existing transmitter needs about 10 percent more output to compensate for combiner insertion loss (0.46 dB);
- The transmitter site will consume more power;
- The second transmitter may require more floor space;
- Cooling and air handling loads are increased significantly;
- Most AMs will have to upgrade STL to stereo or AES;
- AMs will have to upgrade TX to solid state;
- AMs will have to be flat to +/- 5 kHz bandwidth.
Digital broadcasting will introduce some unexpected changes for some stations.
At the studio, the biggest will be the nearly 4.5-second delay between program origination and audio coming out of the receiver. At present, the design has 4.5 second DSP delay and 4.5 seconds of diversity delay. The diversity delay will remain at 4.5 seconds; however DSP delay may be reduced in commercial implementation. The standard analog portion of the radio signal also is delayed; it will be time-aligned to coincide with the digital audio.
This means changes in air control rooms, where most stations monitor directly off air. This will also require a second audio processing chain just for the headphone feed for the DJs. No self-respecting DJ enjoys listening to himself or herself in unprocessed glory off the Program Bus. At the transmitter, the STL will feed an AES signal into the 4.5-second delay, which will feed AES 44.1 kHz into your analog audio processing. (Some early digital audio processors will require an upgrade to handle 44.1 kHz; analog processors will require an A/D conversion.)
The DAB signal will have its own processor to optimize it for the data-reduced signal. This represents a change for most, because many STL links have the processing at the studio, and the processed composite signal is sent off to the transmitter. Now, the STL will send stereo AES into the processing.
Sample rate DAB will be a 44.1 kHz sample rate signal. Besides pre-delay monitoring, the prime digital studio consideration is in the sample rate standard. With DAT tapes at 48 kHz and CDs at 44.1, and standards in the digital audio processing ranging from 32 to 48 kHz, the best you can do is reduce the number of sample rate changes before transmission.
The same goes for data reduction schemes. Everyone will probably tire of the familiar "cascading algorithms" issue. The transmitter site raises further issues. There are two ways to prepare an FM site for DAB: low-level or high-level combining. Before you discount the low-level method, read on.
Consider a site that requires less than 10 kW. For various reasons, 10 kW rigs are currently the upper limit for solid-state transmitters (although higher-power models are in the works).
A 10 kW transmitter linearized for IBOC will be able to deliver between 7,500 and 8,500 watts of the analog RF signal, plus the digital signal. In the case of a site with an analog TPO of 7,500 watts, the actual digital signal requirement is only about 75 watts (20 dB down from carrier).
By comparison, in a high-level separate amplification approach, the same transmitter site would require the original transmitter, a combiner, plus a linearized solid-state transmitter capable of 3,000 watts peak. These are worst-case digital power figures.
The high-level combiner requires some attention. It is a small device, only about three feet long, and a little bigger around than a 3-1/8-inch line section, with two extra ports. The design consideration was to optimize the combiner such that the impact to existing plants will be minimized. By unbalancing the combiner, the final design loses only 10 percent from the analog signal, but 90 percent from the digital signal. This will allow many stations to insert the combiner without having to upgrade or purchase new for their legacy facility.
With the introduction of the DAB transmitter and combiner, there also comes a dummy load to handle reject power. Reject power from the combiner is a sum of the 10 percent lost from the analog signal, plus 90 percent of the digital signal.
In our example of 7,500 watts output, 750 watts will come from the analog transmitter, and 675 watts (average) will come from the digital transmitter. We derive the 675 by taking 750 watts and subtracting 75 watts forward power.
The reject load will have to handle 1,425 watts full time.
Naturally, we must deal with all the extra heat. We now have twice the heat load compared to a standard transmitter site. While a computer room-quality cooling system is not necessary, consideration should be given to the transfer of heat out of the transmitter building.
Solid-state FM transmitters are susceptible to heat rise considerations. Most of the new equipment is CPU-based, and excessive heat buildup will only shorten the life of your investment. Check equipment manufacturer’s specifications for recommendations.
Turn on the power
Power considerations include providing for an extra 500 watts of DAB equipment at 120 VAC, any extra cooling required, and the transmitter. Single-phase and three-phase transmitters are available. Check for breaker panel space and get breaker sizing based on manufacturer’s recommendations. A typical three-phase 10 kW solid-state transmitter will require a breaker in the 60 to 70 Amp range. Do not ignore main breaker capacity and service entrance capacity.
Some transmitter sites already have uninterruptible power supplies because of the newer STL equipment, and digital exciters and processors. They may need to be upgraded or supplemented to handle more equipment.
The rack of digital equipment will consume less than 500 watts unless you get a solid-state transmitter that requires a linear amplifier in front of it. The drive even from a low-level combined RF signal is under 1 watt. The digital RF signal is on the order of 1 milliwatt.
If you have a backup generator, check to see if it can handle both transmitters at the same time. If not, you will have the choice of upgrading the generator, or running without one transmitter in case of power outages.
The typical transmitter dimensions range from 24 to 32 inches wide, and anywhere from 24 to 50 inches deep, depending on manufacturer. Some transmitters require an external bandpass filter, which typically can be hung or left on the floor. An extra equipment rack may be needed for the DAB exciter, audio processing, remote control and UPS.
There is also the reject load. Figure on about one cubic yard (3 x 3 x 3 feet) to account for needed free space around the load. These can be hung from the rafters as well. A single-phase transmitter may require more floor space (extra cabinetry). And don’t forget room to swing transmitter doors.
Wiring considerations include interlocks from coax switches, combiner and reject/dummy loads. The additional remote control channels can be bare bones (three control, four metering, three status), up to extravagant.
Thankfully, the typical FM antenna handles the digital signal as well as the analog signal, so the antenna is not going to be a concern. FM conversion costs range from $48,000 to $235,000 for digital equipment purchases and other upgrades. If you already have a solid-state transmitter, check with the manufacturer to see if an upgrade kit will be available. Naturally, these are ballpark numbers, and each site will have to be surveyed for current capacity and needs.
The AM transmitter story is somewhat less dramatic. Many AM stations already have solid-state transmitters; and with the possible exception of very early PDM models, many solid-state transmitters can be upgraded with a relatively low-cost modification. The four-phase and higher designs will handle the DAB signal best. AM costs ranged from $27,000 to $185,000 depending largely on how much work has already been done to the site for such things as AM stereo, broadbanding and pattern optimization. The STL will have to be upgraded for most stations, again with the goal being to get an unprocessed stereo AES signal into the transmitter site.
This is an exciting time in the history of broadcasting. A generation of equipment is coming which will give rise to many late-night stories over the workbench that start out something like, "I remember when transmitters ran non-linear, and had little ‘fire-bottles’ all through them …"
For the past 10 years, we have witnessed the development of several variations on the digital broadcasting theme. It was a sad but necessary step to watch various parts of the world take the digital plunge, only to realize that some facet of the system was not yet ready for the masses.
IBOC has always held out the promise, and now appears to be ready for implementation. It is stirring excitement even in Canada and Europe, with its promise of low cost receivers and a relatively "soft" conversion.
RW welcomes opinions from readers and manufacturers about the implementation of IBOC.