The transmitter is the final leg of the radio broadcast chain. No matter how good the chain is up to that point, failure at the transmitter will render the whole effort for naught.
But changes are coming, disrupting big iron. Radio World talked to some of the leading transmitter manufacturers about what is on their radar these days.
Joining us in separate Q&As that are integrated below were Scott Incz, managing director, BW Broadcast; Rich Redmond, chief product officer, GatesAir; Chuck Kelly, director of sales, Nautel; Thorsten Becher, vice director sales, Transradio; and Eric Pere, broadcast project manager, WorldCast.
RW: What is the hottest thing in transmitters these days?
BW Broadcast Chuck Kelly: I’d propose that there are two themes driving innovation in our industry today, intelligence and integration; and they are interrelated. By intelligence I mean harnessing IP, local remote data gathering, processing power and connectivity. Mixing these elements has created new options for automatic audio backup and local playout, unprecedented control and opportunities for a true digital path right through to the exciter for optimum loudness and sound. Integration, as a byproduct of harnessing the intelligence in sensible ways, helps engineers reduce the component count in their facilities and do more with less in a positive way. When fewer boxes or components are needed, fewer things can fail and there can be less cost by integrating features in the transmitter. Examples of these two themes working together include integrated audio processing, remote control, codecs, fail-safe systems, etc.
Scott Incz: The advancement in transistor technology, which enables transmitters to be more compact, run cooler and have longer lifespans, especially with the introduction of LDMOS devices. Additionally, one of the hottest things is IP connectivity — not only for remote control monitoring but also for the delivery of audio to the transmitter via network protocol, such as AES67 and others.
Rich Redmond: Transmitters are going in two directions. They either seem to be built for higher and higher power capabilities, often liquid-cooled in design, to deliver HD Radio and other digital formats; or they are getting simpler and simpler for lower power. If you think of translators, certain Class A stations at very high elevations, educational FMs and single-frequency boosters, there is a lot of activity in relatively low-power levels under 1 kW. Over time, those transmitters have had more capabilities added, but they get much simpler to operate. And both low- and high-power transmitters have also grown far more compact.
Thorsten Becher: We like to talk about long- and medium-wave broadcast transmitters, which we feel are transmitters offering a DRM power equal to typically 80 percent (or even more) of the AM carrier power with MER >30 dB.
Eric Pere: It makes me smile to hear of “hot topics” in the FM transmitter market given that it has been around (at least in stereo form) since the 1960s. It is truly amazing to think that it is still so prevalent and has not undergone any major technological changes in that time — just adapting modern technologies to work with old and existing FM systems is in itself a major task.
Not that there have not been any changes but, when it comes to transmitters, engineers are notoriously conservative. Most people would never contemplate buying a CRT TV over a flat screen just because it’s “proven technology” but the critical role of the transmitter coupled with the expense involved means that change will come much slower than in other areas. Personally, I think that the time has come to make the definitive switch to digital audio and perhaps even digital MPX. Digitizing an analog MPX signal may not be ideal but we can multiplex various signals in a digital stream in a much more efficient way.
RW: Liquid cooling would appear to be a winner for closed-loop cooling systems and climates with sites that need almost constant AC like desert or tropical regions. Energy savings should pay for the transmitter in a few years. Few offer this option so far. Is the market just not big enough or what are the reasons it has not gained more traction?
TransradioRich Redmond: GatesAir has been in the liquid-cooled transmitter business for nearly 50 years, including high-power AM and TV transmitters. While liquid-cooling is relatively new to FM, the technology and its benefits are not new to GatesAir.
It does require a certain skillset to operate that not all engineers possess. These systems are a little more complex, and will require some training to learn if the customer does not have prior experience. GatesAir provides that training for customers both new to, and experienced with, liquid-cooled transmitters.
In some places, the climate doesn’t necessarily call for liquid-cooling. In warmer climates, where irrespective of the temperature you want to run a closed-loop system with air conditioning, there are distinct operating advantages with liquid cooling. But they require more attention than an air-cooled transmitter, where you are simply dumping air into a room for cooling. Each station needs to decide what is most appropriate for their situation, and where they feel comfortable. But we are fans of liquid-cooling, and it offers a very attractive ROI for sites that have significant cooling needs and costs. It is a technology that we are comfortable with, and we have led the industry on the liquid-cooling innovation front for many years.
Thorsten Becher: Both techniques, air-cooling and liquid-cooling, have their particular pros and cons. From our point of view, the major drawback of liquid-cooling is potential leakage, either during normal operation or, particularly, in the course of maintenance work. There is no means for hermetically sealing a liquid-cooling system, thus there remains some risk of water drops affecting sensitive transmitter electronics, or even worse, provoking injury to people due to water contacting mains wires.
Eric Pere: Generally, FM transmitters operate far more efficiently than DTV or DAB transmitters so, the advantages of water cooling aren’t as significant in the FM sphere.
If a manufacturer does wish to produce a water-cooled FM system, they can opt for one of two approaches. The first option requires a significant initial investment by the customer but delivers a highly-efficient and effective performance. This is the route which we at WorldCast Systems chose when, in 2004, we developed a liquid-cooled FM transmitter under our Ecreso brand. For us at that time, it was less about operating at high ambient temperatures and more about improving the reliability of the system. We utilized technologies initially created for the railway industry to reduce the temperature of the power-related components such as the power supplies, RF, coupler etc. We also deployed DC/DC conversion techniques used in distributed computer architectures for greater overall efficiency. The result was a reduction to the total cost of ownership and to the carbon footprint of the system but customers have to be prepared to make a significant capex spend to acquire it.
The other approach that manufacturers can adopt is much more affordable for the customer but, in general, offers little or no advantage compared to an air-cooled system. In order to get any appreciable benefit, you need to have a large number of FM transmitters using a single water-cooling system.
Many also tout the maintenance advantages of a water-cooled system as no air cleaning is required. However, when maintenance is required on the water-cooling system, it requires a complete shutdown of all transmitters — a point often forgotten. Given these reasons, we can see why the market is limited for these systems.
Chuck Kelly: Let me start by taking a look at TV broadcasting, where liquid-cooling has been more commonplace. Liquid-cooling as an alternate to a pure air system has historically been very attractive to larger TV sites where high amounts of waste heat are generated by powerful, 20–40 percent low-efficiency TV transmitters. It is hard to beat the thermal properties of a fluid to move large amounts of thermal energy. If there is a negative related to liquid-cooling in these large installations, it is typically related to the additional effort need for plumbing, approvals and exterior air exchange units. Typically the advantages aren’t compelling enough at low- to mid-power and so in the TV industry, liquid-cooling is largely reserved for higher power transmitters.
When we look at the radio broadcast industry we find dramatically more efficient transmitters, typically in the 60–75 percent range, so radio engineers normally find heat extraction much easier to manage. While there may be some broadcasters with very high power needs and liquid-cooling-friendly sites, the majority can manage waste heat extraction quite effectively using an all-air cooling system. So, this becomes a “your mileage may vary” situation depending on many factors including TPO power, transmitter efficiency, existing air investment versus greenfield site, floor space costs, site readiness for liquid-cooling and local power rates. For new installations, where the cooling infrastructure is not already in place, and where high-power and warm climates incentivize its consideration, a case can be made for the total cost of ownership of a liquid-cooled system. However, the station must be aware that initial cost and installation costs will be higher and the time required for installation will be days instead of hours. Further, only about two-thirds (but not all) heat is transferred to the heat exchangers, so nominal conventional cooling is still required for power supply dissipation. Factory training on the technology is essential, and it may not lend itself to being maintained by nonregular contract engineers making only occasional visits.
Similar to the TV broadcast situation, for transmitters with lower waste heat generation, air-cooling will continue to be the de facto standard. The vast majority of broadcasters have air-cooling-compatible sites today, so it will be interesting to watch the ongoing trend in preference for air-cooling versus liquid for radio broadcasters as radio transmitters become more efficient and waste heat requirements decrease.
Scott Incz: With efficiencies of modern devices sometimes exceeding 90 percent, the requirement for liquid-cooling is less necessary in FM transmitters, than say, TV transmitters. TV transmitters’ digital waveforms mean lower efficiencies are obtained from power amplifiers, creating heat issues. Liquid-cooling should really be unnecessary and would only increase the complexity of design and build for the manufacturer, and maintenance cost for the customer. We would rather spend time increasing our transmitters efficiency and thus alleviating the need for complex cooling systems. Less maintenance, increased reliability and reduced operating costs. It’s a win-win for manufacturers and customers alike.
RW: LDMOSFET has been the winning SS RF device technology choice for a number of years. Has that technology been maxxed out for efficiency and performance? If not, what other improvements are in the pipeline and if so, are there any new technologies being developed to replace LDMOSFET?
WorldCast Eric Pere: Today the performance of MOSFET amplifiers is close to the theoretical limit. Those final few percentage points will be the most difficult to obtain, but development continues. As a comparison, 90 percent efficiency was a good performance for switching power supplies a few years ago. Now, with some fine tuning on the DSP, we can obtain upwards of 95 percent.
LDMOSFET’s most likely successor waiting in the wings is GaN or gallium nitride. GaN offers low intrinsic losses to achieve high efficiency as well as very large bandwidth with a flat gain over frequency. It is currently more expensive compared to LDMOS and will likely only become an affordable option for FM transmitters following larger scale deployment in the TV market.
Scott Incz: The latest technology of LDMOSFET devices has certainly not “maxxed” out. Our own design engineers are starting to see efficiency close to 90 percent and we believe we can obtain even higher with some of the design concepts we are working on at the moment. This is still work in progress but we hope to reveal more soon.
Rich Redmond: LDMOS is the standard today, and the most beneficial innovations in recent years was the development of 50 V LDMOS. This really ushered in the convergence of robust, cost-effective power supplies from the IT and telecom industry; along with very robust and high-power RF devices. The transmitters can now take advantage of some of the innovations in very-high-efficiency power supplies that come from these other industries.
What we continue to see is increased power density in the device. A few years ago, a 500 W device was a fairly large device to have. Today, there are 1500–1600 W LDMOS devices. The technology to cool and handle that amount of wattage in a small footprint continues to evolve. The advances continue, and we’ll continue to see the evolution of LDMOS both in power handling capability, the ability to cool it and energy efficiency. Some of these innovations come from mobile base station market developments, so in broadcast the convergence of 50 V LDMOS really introduces a lot of technologies that broadcasters can take advantage of for very reliable, highly efficient and cost-effective transmitter solutions.
Thorsten Becher: The latest versions of LDMOSFETs seem to have reached some maximum of efficiency and performance limits. Nevertheless, innovation in the semiconductor market never comes to a stop. Remember the microprocessors used in PCs of the 1980s, and compare to today’s high-speed multicore chips used in any cheap smartphone. We are confident that ongoing development in new material compounds, further minimization of the active element size and application of new structural designs will without doubt lead to further improvement until reaching the limits given by physics.
Chuck Kelly:While LDMOS and alternative chip technologies still hold potential for performance gains, unfortunately most chip manufacturers are directing their investment towards the high-frequency or pulse-power devices needed in lucrative, high-growth sectors such as LED lighting, high-power ISM, “big science,” cellular, satellite, military and space. The result is that broadcast transmitter designers today are faced with the prospect of squeezing out incremental performance gains. That being said, our industry is populated with many creative engineers who have proven their ability to leverage chips optimized for other technology sectors and create performance advancements in our own industry.
RW: Transmitters used to last 20 and even 30 years before short lifecycles for chip versions and computer control entered this space. The same is true for consoles and studio gear. Manufacturers are now telling us 10 years is about the most we can expect. Replacing large capital items like transmitters at that cycle is not something most station owners can easily budget for and accommodate. What are transmitter manufacturers doing to alleviate that pain?
Nautel Eric Pere: Granted, new transmitters nowadays may not last 20 or 30 years but, in general, they are more affordable than they have ever been. This is especially true when you consider the fact that they have (or at least should have!) a built-in digital exciter and a much-improved total cost of ownership as well as being more compact than previously.
And of course, computer control also brings some great advantages such as the ability to monitor every single part of your transmitter in great detail. We are very fortunate at WorldCast to have great experience in monitoring and control through our Audemat-branded products and we apply this expertise and technology throughout our Ecreso transmitter range. This has resulted in features such as Expert Maintenance Reporting (EMR), which sends the user regular reports on the status of key parameters as well as information on the performance and lifespan of components. We also have a new SNMP management software suite called WorldCast Manager which provides performance and alarm information on not just the transmitter but all devices across a network.
Chuck Kelly: At Nautel we talk a lot about the long view. We know our customers buy equipment for the long haul and we do our best to design our whole offering: transmitters, reliability, serviceability and support in a way that supports that long vendor relationship. In fact, even the ownership structure of our company is supportive of the long view approach. Our experience is that we are still seeing many customers maintaining our systems for multiple decades, but we also do see the impact of an increasing pace of technological innovation that causes some customers to replace systems earlier to gain new capabilities or efficiencies.
For instance our higher efficiency transmitters sometimes justify an earlier technology upgrade to attain lower operating costs. Another example would be a customer who can change the way they monitor their facilities thanks to modern sophisticated control capabilities afforded by Nautel’s Advanced User Interface. Digital radio components also tend to have a shorter replacement cycle given the pace of change in the digital broadcasting space. So while we still see and believe in the long term view, there are situations where customers are choosing to change out equipment at a faster pace than may actually be required through the lifecycle of the product.
Thorsten Becher: A good work rule in the early design phase is to select and use well-proven components known for their long-lasting availability in the market. More complex devices such as embedded computers or touchscreens should be, both electronically and mechanically, of an easily replaced design. What really counts is energy costs, reliability, ease of maintenance and availability of spares for a long period of time. Following this rule, Transradio manufactures transmitters with minimum maintenance requirements, and can provide spare parts for typically 15 years or more.
Scott Incz: The advances made in transmitter technologies have actually led to an increase in reliability, cost and lifespan of transmitters. The reason many stations may look to upgrade transmitters after 10 years is to take advantage of features that the newer transmitters will offer them. Each generation of transmitters is more efficient, more reliable and offer stations flexibility, sometimes integrating other products to create a “one box” solution. Stations may actually find it more cost-effective to upgrade to a greener, more reliable, less maintenance heavy transmitter, than continue with their existing one. The cost of transmitters is actually far less than it was 20 years ago, so even smaller stations can take advantage of upgrading to keep up with the larger networks.
Rich Redmond:There was a period of time when computers lasted for a very long cycle. Transmitter designs, like computers, have changed over time. Part of it is the sophistication: Transmitters that lasted 20-to-30 years had relay-type control and/or very large transistor logic. They had a single tube and a blower inside to cool them. They also weren’t incredibly efficient like today’s transmitters, which have the benefits of soft failure, solid-state and high-efficiency advances. Often, the replacement cycle is not because the product wears out; it’s that certain devices within that transmitter cease to be available.
If you look at the price of transmitters now compared to then, they are much more cost-effective than they were a number of years ago. In the early 1990s, a 1 kW FM transmitter might cost you $16K. Today, a 1 kW is generally under $8K. The prices have dropped, and they are easier to repair and maintain over time. Engineering resources are also more efficiently put to use; weekly trips to the site, and consistent tube cleanings, are eliminated. Lighter weights from today’s more compact transmitter footprints often allow single-engineer maintenance. So overall, the capital investment of buying, installing and maintaining a transmitter is substantially less expensive, even when considering the quicker replacement cycle.
RW: In general, what are the most common failure modes of modern designed transmitters manufacturers are seeing in the field? Lack of cleaning and maintenance plus improper grounding and lightning protection are certainly the major ones, but what are the others?
GatesAir Chuck Kelly:The vast majority of failures can be traced to lack of maintenance, improper grounding or poor lightning protection (ties to grounding). It’s hard to say there’s another common failure mode beyond those — short of the ones that result due to customers not keeping up with software updates. It’s important to remember that we frequently use software updates to perform hardware changes (adjusting amplifier bias levels, power supply output voltages, etc.). Therefore, what we used to do by sending out a bag of parts and an instruction sheet we now do with a software update and not implementing these updates can have a direct effect on equipment reliability.
Eric Pere: Generally, poor or unstable mains supply is one of the most common causes of failure observed. In many awkward transmitter locations, mains electricity is subject to large variations, short dips and interruptions which means that voltage regulators, inverters or generators are often necessary. However, these can also generate some unexpected variations when switching or adjusting the voltage. Most modern transmitters use switching power supplies which can easily manage voltage changes but several, quick variations generate large current peaks which, in turn, generates a lot of stress for their active parts. This was not an issue for the old linear-style power supplies but their poor efficiency and noisy AC behavior rendered these obsolete.
Scott Incz: Lightning protection is incredibly important and is responsible for some in-field failures, but we see many more power supply failures due to AC main surges or dirty power, this is especially a problem in emerging markets.
Thorsten Becher:From our experience, the predominant cause for failure is an inadequate cleaning of air filters by maintenance staff. In few cases, also high spikes on the mains supply lines have been encountered.
RW: How much longer will high-power tube transmitters be relevant in our business? Many engineers worry that if companies that make and rebuild tubes like Econco/Eimac and Richardson decide to stop producing or close their doors, that will mark the end of this era.
Rich Redmond: High-power tubes will remain relevant for the foreseeable future for a certain customer base. This will continue to dwindle as the engineering base shifts to a younger, more IT-savvy generation. Fewer engineers entering the market today do so as RF specialists. More often, RF is a trade they learn, or outsource to firms with experience. However, today’s solid-state transmitters are often sensible for younger engineers that have IT experience, and there is a quicker learning curve when it comes to maintaining the transmitters thanks to design benefits such as hot-swappable PA module and power supply replacement. But the change will continue over time, and there is still a need for vendors that make tube transmitters, and make and/or rebuild the actual tube components.
Chuck Kelly: Tube technology is still the most efficient, cost-effective approach to high-power shortwave transmission where quick frequency changes are made over a wide frequency range. However, in AM and FM systems over all power ranges, solid-state has become the preferred technology as it has evolved to be more cost-effective, more fault-tolerant and offers higher performance than tubes. And yes, the writing seems to be on the wall for tube suppliers/rebuilders. In AM and FM broadcast at least, it’s hard to imagine a scenario where it would be wise to specify a tube in a new design or purchase.
Thorsten Becher:Today, all broadcast bands ranging from longwave up to band IV/V for DVB-T(2) and, literally, each RF power in the respective band can be served with all solid-state transmitters. Thus, we consider tubes no longer relevant for new transmitter investments. But agreed — as soon the few remaining companies in this business such as Eimac or Richardson close down production or refurbishment of power tubes, the tube transmitter era will come to its final end.
Scott Incz: Not so relevant because long-life solid-state amplifier parts, including the latest LDMOS technology, are producing longer lifetime, more compact levels of high-power RF. It is probably accurate to say that the tubes won’t be around for much longer, at least for broadcast applications.
RW: Put on your science fiction cap. What fantasy yet possibly-doable-someday transmitter technology do you wish you could invent, perfect, bring to the market at an affordable price?
Thorsten Becher: Just my personal fantasy — making a transmitter offering the same RF power in a cabinet of half the size and weight than today, saving production costs and simplifying shipment in the entire world.
Rich Redmond: Imagine a high-power transmitter that is 100 percent efficient and the size of a college dorm fridge. Once we have reached that point — and the day will eventually come — there’s a high chance that there won’t be much more room for innovation. In the meantime, we will continue to work toward that goal.
Scott Incz: A supercool, super-efficient, super-reliable transmitter, in an unbelievably small box!
It would need to have superconductivity in the power supplies and RF amplifier stages, an RF amplifier output requiring no cooling and minimal real estate. It would need increased high-power FET devices that need little matching to the antenna stage, as well as a single-chip IP audio to on-channel modulation capabilities. It’s not here yet, but we are getting there.
Eric Pere: I can imagine some kind of high power RF DAC to achieve direct to RF output modulation. On the other hand, it is more likely that, after being around for a century, FM broadcasting will be replaced by a new way of listening to radio. I still dream that it could be possible to achieve highly-effective digital radio with affordable receivers and good coverage but if we continue to pretend that it already exists, it will probably never actually happen.
Chuck Kelly: Well, with ATSC 3.0 we may see a lot of broadcasters implementing SFNs … we should watch that space. As a company we’ve explored some work on making portions of the radio broadcasting spectrum all-digital and yet live comfortably within the existing analog and hybrid digital spectrum allocation. We still see generous amounts of room for dreamers and innovators in radio broadcasting. We’ve been fortunate to be able to hire a lot of fresh designers over the past few years and we look forward to the contributions they will make in the coming years.
Looking even farther out, we envision that all broadcast content will be data, reconstituted at the receiver to be separated out and readied for consumption. Broadcast AM/FM/TV will continue to be relevant as long as the cost to distribute that data is less than other methods.
Comment on this or any story firstname.lastname@example.org “Letter to the Editor” in the subject field.
Read our recent roundtable about trends in processors at radioworld.com/processors.