Alan White, Senior Project Engineer
The emergence of digital audio for radio has evolved from over a decade of overcoming technical challenges. The enhanced user experience now available with HD RadioTM technology (also know as IBOC) is not without challenges. More stringent requirements on transmission equipment have led to the development of a variety of methods of implementing IBOC technology, often requiring new transmission equipment. This paper will address IBOC in general; methods of broadcasting both the analog and digital signal; and in particular the high power low-level combined transmitter developed by Continental Electronics.
The IBOC signal is a digital multi-carrier OFDM signal that shares the same channel allocation as the analog FM signal and allows both digital and traditional analog FM (Analog + HD) to be broadcast. The digital signal is broadcast as a dual sideband to the analog FM, but 20 dB lower in average power and is within the FCC spectral emissions mask for FM broadcast systems. However, to prevent interference to other channels Ibiquity has specified a new spectral emissions mask that is more stringent then the currently accepted FCC mask. Considering that a typical analog FM transmitter is not capable of meeting the new spectral emissions mask while passing the new IBOC digital signal, external equipment such as high-level and mid-level combiners, or separate radiation systems needed to be employed, or new transmitter designs needed to be developed.
The amplification of the digital signal requires a linearized transmitter to keep intermodulation (IM) products, also know as spectral re-growth, from causing interference and distortion problems. The nature of a multi-carrier digital system is to have a high peak-to-average power ratio, as high as 10 dB. This high peak power can drive a power amplifier into compression, which is a non-linear region of operation. This leads to spectral re-growth and distortion. The peak-to-average power ratio used in HD only transmitters is ~6 dB. The peak-to-average ratio used in Analog + HD transmitters is ~1.3 dB. Trying to accommodate the peak power and operate the transmitter linearly reduce the average power output of a given device and reduce the efficiency. Several methods of achieving the combined Analog + HD signal have been derived to handle the challenges of digital radio. Each of these methods has their own advantages and disadvantages. The following is a brief overview of the prevalent methods of implementing HD Radio.
Low-level combining is by far the simplest and most elegant solution to generating the IBOC signal. It is a technique where the analog and HD signals are combined at the exciter level, thus requiring one transmitter. The same amplification path is used for both signals. An expensive high power combiner, additional antenna or additional transmission line will not be needed. A tower crew will not be needed to perform the installation. Additional floor space may also not be needed. The cooling and AC power requirements are only marginally affected. Historically this has required a solid state transmitter that is linearized and has reduced power output to accommodate the 1.3 dB peak-to-average power ratio of the Analog + HD signal. This means that a typical 10 KW solid-state transmitter is limited to about 7 KW of Analog + HD power. Using a solid-state transmitter for this approach is not as efficient as high-level or split-level combining. However, using an 816HD series single tube transmitter for high power situations, there is a significant improvement in overall efficiency over solid-state due the inherently higher efficiency of the 816HD transmitter and the lack of a combining system required in order to increase a solid-state transmitter to high power levels. As recently demonstrated at the 2005 NAB show, power levels in a low level combined system can reach up to 17.5 KW in a single box solution using tube technology.
Figure 1
Low-Level Combining
High-level and mid-level combining techniques are similar and suffer similar disadvantages, with the mid-level combining scheme offering an improvement in efficiency over high-level combining. However, mid-level combining is far more complex to install correctly. System block diagrams for both techniques are given in figures 2 and 3. Both techniques require a high power analog transmitter and a medium power HD transmitter. The two transmitters are then combined into a common transmission line with a 3 dB, 6 dB or 10 dB coupler. A circulator is often required on the HD transmitter due to inadequate isolation from the analog transmitter. No additional antenna is required. In the case of high-level combining 10% of analog power and 90% of digital power is lost as heat in an RF reject load. The analog transmitter therefore needs to operate at 11% higher power to make up for this lost power. This additional power may not be available in the customers existing analog transmitter, requiring the purchase of a new analog transmitter. The power lost into the reject load creates a significant amount of heat in the equipment room that must be dealt with, often by increasing the cooling capacity, and additional AC capacity of the equipment room may be needed to power the new equipment. Finally, the additional floor space required in the equipment room is often problematic. Mid-level combining is an improvement over high-level combining but with the added complexity of phase matching two transmitters and similar problems with heat, AC capacity and floor space.
Figure 2
High-Level Combining
Figure 3
Mid-Level Combining
Separate radiation systems require the use of two RF paths, one for analog and one digital. This solution also requires two transmitters. However, the HD transmitter power requirement is only 1% of the analog power requirement. This may provide the best overall system efficiency but has numerous other drawbacks. The requirement of an additional antenna can often be very expensive due to the cost of the antenna, transmission line and installation. In addition, replicating the coverage area of the analog signal with the digital signal may be very difficult.
Figure 4
Separate Radiation Systems
Dual input antenna systems are available that offer similar efficiency as separate antennas without the difficulty of the digital signal replicating the coverage area of the analog signal. However, a good deal of costs associated with this solution such as a separate low power HD transmitter, separate feed line, a new dual input antenna, and a tower crew for the installation may drive the costs to a level approaching that of other more complex solutions.
If there were one "right" solution for transmitting IBOC then obviously there would not be so many alternatives. At Continental, we looked at the various methods of IBOC implementation and decided to look at the problem from the requirements side rather than the solution side. A requirements based process involves listing the requirements such as complexity, purchase and operating costs, space required, conversion logistics and technical support. After a careful analysis we determined that the better solution for many higher power stations would be low-level combining because of the advantages over alternative solutions. Low-level combining offers many advantages: a.. Simplicity - A single transmitter does it all b.. Space - Generally smaller than separate amplification systems c.. Cost - Generally lower cost than separate amplifiers because of fewer components d.. Logistics - Almost never requires tower work to go digital
Even though these are compelling advantages, the low-level combined system has not been used extensively in high power IBOC systems. Until this new breakthrough there have been significant disadvantages in size, cost and efficiency. In order to produce the complete IBOC signal a transmitter must produce about 35% more peak power that its normal analog FM power. Solid-state amplifiers require a certain number of output devices based on the peak power rating of the transmitter. Factoring in that power supply and cooling are generally sized relative to average power a low-level combined solid-state IBOC transmitter would likely be more than 25% larger than the equivalent analog FM transmitter. Analog solid-state transmitters are already at a size and cost disadvantage over high power single tube designs and adding IBOC to the equation makes the disparity even greater but a 20kW tube transmitter using the modern 4CX20,000E tube can produce the extra peak power required by IBOC without increasing the size.
The primary technical hurdle to overcome with any digital transmitter whether tube or solid-state is linearity. Non-linearities in the transmitter produce intermodulation products that cause out of band spectra and self-interference between the digital and analog signals. Continental has researched these requirements and identified several design changes that will allow a single tube transmitter to meet all performance requirements for IBOC. Some of the changes include new operating points for the tube and a redesigned RF drive chain.
Convincing a tube to perform as a linear amplifier requires design elements often not considered in FM transmitters. Nearly all high power FM amplifiers operate in a high efficiency mode. The most common is the so-called class C mode. Class C amplifiers are inherently non-linear but this is not important for constant envelope signals such as FM. Something that is usually not a priority in the design of class C amplifiers is the small signal gain and stability. I'm sure many readers can recite interesting examples of FM amplifiers that produce eye opening (and ear deafening) responses when RF drive is accidentally removed. This is indicative of small signal instability. But a very important requirement for any IBOC amplifier is that it must be linear and therefore must be stable in the presence of a varying input signal. This requirement is already met by the final amplifier in the Continental 816-C series that utilizes the Eimac 4CX20,000E tetrode.
The main reasons that the 4CX20,000E tube was chosen as the basis for IBOC development is its high plate dissipation rating and long tube life owing to its large cathode structure. The optimum configuration for this tube was developed through months of research to arrive at a combination of linearity, efficiency and output power. There are changes to both screen and bias operating points. The output cavity is essentially the same as the current 816-C series but the grid RF input has seen significant revisions to improve bandwidth, group delay and linearity.
Once the final amplifier met the linearity requirements additional tests were also performed on the tube to determine the effects aging will have on tube performance, particularly pertaining to linearity. As a tube ages the free electrons available from the cathode diminish. This is due to the depletion of the thoria at the cathode surface. Typically this is observed as reduced gain and output power in analog FM transmitters. Tests were conducted by lowering the filament voltage of the tube below the nominal operating point of 10.0 V to as low as 7.0 V to quantify the effect of having reduced the number of free electrons available from the thoriated tungsten filament to simulate the effects of tube aging. The results showed that significant power loss on the order of 25% or more was observed prior to any noticeable effect on the IBOC emission mask. Therefore, the tube would most likely be replaced due to power loss before a significant degradation of linearity occurs. Results are shown in Figure 5.
Figure 5.
Filament test results.
There were significant design changes required in the input-matching network of the 816-C series to maintain linearity. But an equally big challenge involved the RF driver. The 816-C was designed with a high efficiency class C solid-state driver. This driver is compact and requires few if any operator adjustments. However a new IPA was required to meet the stringent IBOC mask specifications. The new IPA consists of four push-pull VDMOS RF power transistors with a total P1dB of +59 dBm (800 Watts). This ensuresthat the drive to the grid circuit of the tube is clean.
Another concern with low level combing is the analog performance of the transmitter. Measurements show that critical analog specifications such as THD+N, IMD, stereo separation, etc. are not adversely affected by the presence of the IBOC digital carriers. Even after all these changes the overall efficiency is every bit as good or better than a solid-state transmitter while at the same time costing less and taking up less space.
Continued testing and modifications to the tube and drive circuits of the 816R series transmitter yielded a rugged transmitter that exceeds previously accepted levels of output power for a low-level combined system and meets the NRSC recommendations on spectral emissions mask without the need for pre-correction, see Figure 6. The power dissipation in the tube is well under the manufacturers specification with a manageable increase in operating temperature over the same tube in our analog transmitter. Based on test results a 30 KW analog transmitter can deliver up to 20 KW of Analog + HD power.Higher levels of output power can be achieved through combining of two 816HD transmitters to achieve power levels up to 35KW.
Figure 6
Spectral plot of 816HD at 18 KW TPO with IBOC mask overlay.
The various methods of generating the IBOC signal have been discussed and it has been shown that the optimum solution for high power IBOC is a low-level combined solution. All competing techniques for generating a high-power IBOC signal suffer from a common significant drawback. They require two transmitters, an analog transmitter and a digital transmitter. The power requirement for the digital transmitter will vary depending on the specific combining technique. All competing techniques are more complex and more expensive at the system level. Low-level combining is by far the simplest and most cost effective solution for most applications. At lower power levels this can be accomplished using solid-state transmitters, but at high power levels a tube transmitter is much more efficient and appropriate. A single tube IBOC transmitter has been developed and demonstrated to be superior over competing solutions offering a simple, elegant and cost effective method of generating a high power IBOC signal.
