Split-level combining

July 1, 2004

The excitement of HD Radio is rapidly building momentum. Some of this has been fueled by the promise of new features, such as the additional audio channels of Tomorrow Radio, 5.1 surround sound and the fact that radios have finally become available for sale in the retail stores. Even still, radio stations are hesitating to convert to IBOC operations because of the cost to install and operate the HD Radio transmission equipment. These issues center on inefficiencies in the current combining scheme. This inefficiency is responsible for an increase in operating costs, the need for more physical space in the transmitter building, possibly being forced to replace existing transmitters and increased HVAC needs.

The WPYO split-level combining prototype using a Harris HT-5 and a Z-16HD transmitter. Above the center rack is the 3dB hybrid used for the combining of the outputs of the two transmitters.

These are major hurdles for some station owners to jump and it's causing them to sit back and wait for a better way. The recent introduction of the dual-antenna method has caught people's attention, and is an efficient method for stations that already have a licensed auxiliary antenna to get on the air fast. Stations without this luxury are faced with the high cost of purchasing, installing and licensing a new antenna and transmission line, plus they may be faced with the cost of tower loading studies and additional monthly tower rental costs. These costs could easily exceed the savings realized from the improved efficiency. The dual-antenna method also produces variances in the digital and analog signal propagation, requires an STA from the FCC and cannot be used for directional FM stations.

Offered solutions

The new split-level combining method addresses all of these issues and can finally make HD Radio affordable. Invented and developed by Steve Fluker, director of engineering for Cox Radio in Orlando, and George Cabrara, Harris RF design engineer, this patent-pending technology is now available for delivery.

Figure 1: A typical high-level combining system.

To see how split-level combining works, first examine the high-level combining system shown in Figure 1. Until recently, this has been the more popular method of transmitting the IBOC signal. In this example we have a radio station operating with a transmitter power output (TPO) of 14,000W. To convert this station to IBOC, a new digital carrier with a power of 140W must be added. While this sounds easy, there are challenges around every corner. Traditional tube type transmitter amplifiers operate in a Class C mode and are not able to carry both signals simultaneously due to linearity issues. A second transmitter is needed to transmit this digital signal, which must be a solid-state transmitter, set up to operate in a linear Class AB mode. The outputs of the two transmitters are then combined through a 10dB hybrid to produce the final mixed signal to the antenna.

This method is inefficient because the outputs of the two transmitters are so different in level and difficult to combine. Because of this mismatch of the signals, 10 percent of the analog and 90 percent of the digital output signals never make it to the antenna and are instead routed to and absorbed by a reject load. Because of this loss, the digital transmitter output must be increased to 1,400W, just to get 140W of the power to the antenna. Something else to notice in this method is the need to increase the power of the analog transmitter as well. In this example, the power level must be increased from 14,000W to 15,500W just to maintain the licensed power. If the analog transmitter has sufficient headroom, this can be done, but in many cases, the existing transmitter is already pushed to the limit, which may force a perfectly good transmitter to be replaced. As you can see, the wasted power of these two transmitters adds up to 2,810W, which is converted to heat. This additional heat places a new burden on the HVAC system in the building.

Locating the reject load outside is a good idea to help reduce the heating problems, but this will incur even more costs for the parts and labor. Because the load can run extremely hot, it's also advisable to protect it from people accidentally touching it, and to keep animals away from it. Weather protection would also be advisable.

Figure 2: The split-level combining system.

Figure 2 shows how the split-level combining system resolves these issues. Notice the addition of an analog component injected from the digital transmitter. In this example, the IBOC transmitter provides 7,000W, or 50 percent, of the total analog power. This eliminates the need to increase the output of the old analog transmitter, and in fact, the power is reduced, which will increase tube life and further lowers operating costs. Also notice that analog outputs of both transmitters are now equally matched, allowing the signals to be combined with no losses. As for the digital signal, because the transmitter must be linear, the output of the digital exciter must be fed into the digital transmitter only. This creates a mismatch at the outputs causing 50 percent of the digital carrier to be lost, but in this example you can see that we've reduced the power to the reject load from 2,810W down to only 140W. This power loss can easily be absorbed into a reject load with virtually no heating inside the building.

At first glance it would appear that the power savings would be equal to the 2,670W, however, keep in mind that the digital transmitter is operating in class AB mode, which is not as efficient as a standard Class C transmitter. Even taking this into account, the station will still be looking at an expense savings in the area of $300 per month, depending on the cost of electricity in the area. Keep in mind that this amount of savings is compared to the operating cost of the equivalent high-level combined method. The overall power bill will still increase when the IBOC signal is added, just not by as much. can be used for virtually any TPO level and can yield a cost savings between 5 and 25 percent over the high-level combining method.

Space saver

Another concern with implementing HD Radio is a lack of physical space in the transmitter room. Everyone has seen the typical transmitter where the back of the transmitter is accessible only by squeezing between the wall and equipment rack. It's no secret that many transmitter buildings were built to fit a main and back-up transmitter only, with no room to spare. Both the high-level and dual-antenna IBOC methods require the addition of yet another transmitter and equipment rack. If the room isn't big enough there is no choice but to add on to the building. This obviously adds a significant amount to the cost of conversion. Worse yet, what if the building cannot be expanded? A choice between a back-up transmitter and IBOC must be made. The split-level combined system addresses this issue.

With both transmitters carrying an analog component, either transmitter can be used individually in an emergency to keep the station on the air. If space is an issue, remove the old auxiliary transmitter and put the new digital transmitter in its place. It might also be possible to convert an old dummy load into the needed reject load for the combined system, which not only saves space, but also saves more money.

The exciter rack with a Harris Digit exciter, Dexstar IBOC exciter and Epal audio interface.

Testing in Orlando

was put to the test in April on Cox Radio's Power 95.3 in Orlando. A recent class upgrade from A to C3 required the station to purchase a new transmitter, creating the perfect opportunity to test this new technology. The main transmitter was a 5kW tube transmitter, which could not make the new 7.3kW TPO, so a new Harris Z-HD transmitter was purchased. Both transmitters were set at 3,650W of analog power, and the digital signal was adjusted to 146W. In a standard high-level combined mode, the reject power level would have been about 1,560W. With the two transmitters properly tuned and phased together, the reject load power dropped to under 73W. This reflects the 50 percent loss of the digital signal. The transmitters were then switched into test loads and both transmitters were turned up to 5,000W. The forward power increased to simulate a TPO of 10,000W, and the reject power was just below 100W, as expected. High-level combining would have rejected more than 2,000W in this case.

In both of these experiments the power output of each transmitter was matched. It is not always possible to match the power levels, nor is it necessary or even always desirable. There is a lot of flexibility in this combining method, which can be customized to fit the needs of a facility. In some cases it may make more sense to give up some of the gained efficiency to save tens of thousands of dollars in installation costs.

Installation costs

The HD Radio/analog low-level combiner section to feed the analog and digital exciters into the hybrid-mode digital transmitter.

In this example, let's assume a radio station with a TPO of 36,000W. Most 35,000W transmitters can actually be pushed to make the extra 1,000W, but not much more. If this station wants to convert using the high-level system, the analog TPO must be increased to just over 40,000W. This transmitter cannot produce this power. The most economical way to achieve this extra power is to purchase another high-power transmitter to combine with the existing transmitter to be able to reach the new power needs.

On the digital side, the power output to the antenna must be 360W, but with a 90 percent loss, the transmitter must be able to produce 3,600W. To achieve this power level, a dual-cabinet transmitter is required. The building must be large enough to house a new analog transmitter, a dual-cabinet digital transmitter, plus additional rack space and a new reject load. All of this adds up to a lot of floor space, which may not be available.

The prototype 3dB high-level hybrid used for split-level combining the two transmitters in the Orlando test. The digital Thru-line sections for the wattmeters are used to accurately measure the true average power of the IBOC signal.

In this arrangement, 7,240W of power will be lost into the reject load. This is not a peak power reached occasionally, but a constant power 24 hours a day. It is unlikely that the building air conditioning system will be able to handle this increased heat load. The options are to add additional cooling, or to move the reject load outside.

The same station using split-level combining will have several configuration options to choose from. To gain the most efficiency, the station may still choose to purchase a dual-cabinet digital/analog transmitter that can produce about 14,000W of analog power along with the digital signal. The old analog transmitter's power will now actually be reduced to about 22,000W and will not need to be replaced or upgraded. Tube life will also be extended because it's not being taxed to the limits. In this case the installation costs could be reduced by as much as $65,000. Because the digital transmitter carries a significant amount of analog signal, it can replace the older auxiliary transmitter, so this installation might fit in the same or only slightly larger footprint as the old transmitters.

Read a white paper that further details split-level combining at this link.

A second option to save even more money up front would be to use a single-cabinet digital transmitter. The efficiency will not improve nearly as much, but the installation costs could be lowered by as much as $150,000. This may also be an attractive option when floor space is restricted, as it may actually consume less space than the existing configuration. The key is in the planning.

Editor's note: has several patents pending. At this time, the technology is sold exclusively through Harris.

Fluker is director of engineering for Cox Radio Orlando.

Receive regular news and technology updates. Sign up for our free newsletter here.