Appendix

Real-life Experiences with HD Radio

The first part of this book deals with helping you get your HD Radio signal on the air. Hopefully, it has helped ease some of the confusion you may have been feeling and, if you are putting an FM HD Radio station on the air, has helped you sort out the various methods of combining the analog and digital signals.

The first part of this book is intended as an educational aid, and presents the information in factual form.

This part of the book discusses my personal experiences with operating a hybrid AM HD Radio station since October 2002. I have many observations to share with you with the hopes that you can get an idea of what to expect with the signal you put on the air. I also have experience with the FM HD Radio system, though not as much as with the AM system, so my observations will lean toward the AM HD Radio system.

I will tell you that being a pioneer with a new technology is not an easy job. Since signing WOR in New York on as New York’s first AM HD Radio station on October 11, 2002, I have been:

   Called just about every name in the book, including some combinations I didn’t think possible.

   Called “iBiquity’s mouthpiece”.

   Threatened, not only with physical violence but also with being taken in front of the Federal Communications Commission (FCC), to “be exposed for what I am”.

   Harassed and stalked.

   Accused of being highly paid by iBiquity for my views on HD Radio and on my willingness to experiment and report my findings.

   Called “the killer of AM Radio”.

   Accused of “starting this mess” and was told that I had “better fix it”.

My parents being married at the time of my birth has been questioned.

My technical abilities and knowledge have been questioned.

All of this has been uncalled for. Moreover, all of this has been done by a relatively small group of people who, in my opinion, wish to continue living in the dark ages. If you have been in the business for any length of time, you know very well that our industry has been struggling for a while, as new technologies and gizmos have been luring the audience away from radio.

My basic belief is that radio needs to do something and come into modern times to survive. The first obvious area that comes to mind is programming, particularly with the advent of the iPod and satellite radio. I’m not a programmer and don’t play one on television, but it doesn’t take a rocket scientist to figure out that if someone can essentially program their own “radio station” with their own music collection on their iPod, or if they can hear music or other programs on satellite or Internet radio, they won’t listen to us.

That being said, if a potential listener has the choice of listening to essentially the same program on a digital source or analog radio, I highly doubt the choice will be analog radio. In order for radio to remain competitive, it needs to evolve to digital.

It could have been relatively easy, had the FCC made additional spectrum available for a digital radio band then gradually easing out analog stations. This, in its basic form, is what the FCC did for HDTV – it assigned unused frequencies, essentially wasted spectrum, to digital television stations and set a sunset date for the analog stations.

Because the FCC did not make additional spectrum available for digital radio, it became necessary to adopt a hybrid analog/digital system, inband on-channel (IBOC), or HD Radio. As with any hybrid system, there are compromises that must be made for both the digital and analog signals to coexist. Part of this is potential interference to adjacent channels in both the AM and FM bands. It requires a different way of thinking about AM and FM station coverage to ease the transition to digital. Incidentally, if you read the FCC’s Report and Order authorizing the IBOC rules carefully, it is quite obvious that the FCC is looking toward a digital future for radio. Some think HD Radio is a fad and will soon go away. It will be with us for a long time based on statements made by the Commission.

So, in the spirit of full disclosure, I would like to address the previous bullet points:

   My vocabulary of foul language was becoming quite repetitive and boring before I became an HD Radio pioneer. Thanks to all who brought different combinations of words to my attention, thus broadening my horizons.

   I have never been a mouthpiece for anyone, and my statements and observations are strictly my own.

   Threats don’t faze me; I have nothing to be concerned with regarding the FCC and, in fact, the FCC has quoted me in its proceedings; and I have nothing to hide – I am a broadcast engineer who stays current with the technology.

   Those who have harassed and stalked me have police folders with their names on them. I don’t mess around.

   I have never been paid even a dime by iBiquity Digital Corporation (hereafter iBiquity). The only thing I have received from iBiquity, besides knowledge, a hearty pat on the back and a huge “thank you,” is one of the first HD Radios when they became available – so I could listen to my station. Incidentally, WOR is held up as the gold standard of AM HD Radio digital audio, something I am extremely proud of.

   I have not killed AM radio. AM radio has been dying a slow death on its own over the years. If anything I did, I have helped slow or stop its death by putting an AM HD Radio signal on the air.

   I did not “start the mess.” and have no intention of “fixing” what some perceive as problems with the hybrid HD Radio systems. I was one of the first to state that the HD Radio systems are not perfect. My job is to make the HD Radio systems we have perform to the best of their abilities. If I were not doing my job, I would not have been selected to write this book.

My technical abilities and knowledge are a matter of record with the FCC, The Society of Broadcast Engineers, the National Association of Broadcasters, the New Jersey Broadcasters Association, the New York Broadcasters Association, the Connecticut Broadcasters Association, and many broadcast engineers around the United States. I do not have to defend them.

Oh! And my parents were married almost three years when I was born.

I can also say that I have told the truth and will continue to tell the truth in regards to my HD Radio experiences, with my experiences being both good and bad. Some people may not like hearing the truth, and that is simply too bad. I will not bend my observations just because someone doesn’t like what I have to say. While we may not always agree, it doesn’t mean I will lead you astray. My opinion is my opinion and I am entitled to it, just as you are to yours. I back my opinion with not only 5 years of operating a hybrid AM HD Radio station under my belt, but also 30+ years of broadcast engineering.

Now that my introduction is out of the way, I will present several different stories to you, which will take you through my adventures with the HD Radio technology. It has been a fun, entertaining, and interesting learning experience. If you are reading this book, most likely you are either contemplating the conversion of a facility to HD Radio or have been told to make it happen. Regardless, I hope you find not only the first part of this book helpful, but also the following stories.

A.1  The beginning

In the spring of 2002, I was approached by iBiquityDigital Corporation. It was looking for a high-power AM broadcast facility in a large city for experimentation, the idea was to see how the hybrid AM HD Radio signal would propagate and survive in a “concrete canyon.” New York City (NYC) was the perfect testing ground, with not only many canyons of concrete, but also electrical noise galore. iBiquity was also looking for a facility that wasn’t necessarily perfect. WOR fits the bill nicely, as the Common Point graph of figure A-1 shows.

FIGURE A-1

The common point graph from the former WOR, Lyndhurst, New Jersey, transmitter facility. It was far from perfect, yet passed the HD Radio signal with no problem whatsoever.

A discussion ensued with Rick Buckley, President of Buckley Broadcasting Corporation, WOR’s parent, and Bob Bruno, Vice President/General Manager of WOR Radio. Would Buckley be interested in, essentially, lending WOR to iBiquity for the good of the industry? We had numerous questions for iBiquity, the first and foremost being what would happen if, in our determination, the WOR signal were damaged by the HD Radio signal.

iBiquity stated that, if we determine that WOR is suffering because of the HD Radio testing, we have the right to shut off the HD Radio signal and disconnect the equipment from the WOR main transmitter with little, if any, advance notice. With that question answered to our satisfaction, WOR embarked on being a pioneer in the next phase of broadcast technology for radio.

FIGURE A-2

This is the original HD Radio exciter installed at WOR by iBiquity Digital Corporation in 2002.

WOR, with the help of engineers from iBiquity, installed an iBiquity test exciter (figure A-2) on WOR’s main transmitter, at that time a Harris DX-50. The exciter itself was a computer running the Linux operating system (which is like present-day exciters), with a ½ Video Graphics Array (VGA) touch screen installed into the front of the computer case.

FIGURE A-3

The EASU, the companion to the HD Radio exciter and its interface to the outside world.

With the exciter was an EASU (figure A-3), the Exciter Auxiliary Services Unit. This unit interfaced the exciter to the equipment installed in the Harris DX-50, and would force the system to switch to standard monaural analog modulation and the transmitter’s internal oscillator in the event of a failure of the exciter.

Sitting on top of the Harris DX-50 transmitter was a small, gray box that was the interface between the iBiquity exciter and the transmitter (figure A-4). Both the RF and audio (or magnitude) signals from the exciter were sent to the transmitter through shielded Category-6 computer networking cable. The interface box did several things. First, it converted the RF signal sent from the exciter to an unbalanced BNC connection to plug into the external oscillator port on the transmitter.

FIGURE A-4

The interface box sat on top of the Harris DX-50 transmitter. This box converted the phase signal from the exciter to a standard RF signal to drive the transmitter, and switched between the magnitude signal from the exciter and standard analog audio to feed the transmitter in the event of exciter failure. It also controlled a relay in the transmitter that would switch between the transmitter’s internal oscillator and the RF from the exciter.

Next, it converted the magnitude signal from the exciter to balanced audio, which would connect to the transmitter’s audio input.

The box also took the standard analog audio from the audio processor and put it through a relay. If the exciter failed, the relay would open putting the normal audio into the transmitter’s audio input.

The last thing the box did was control a relay that was placed in the transmitter (figure A-5). This relay was switched between the internal oscillator in the transmitter and the RF signal supplied by the exciter. If the exciter failed, the box would switch the audio input to the transmitter back to the normal source, the output of the audio processor. It would also switch the RF input of the transmitter back to its internal oscillator in the event of exciter failure. In this way, the station would remain on the air in mono mode should the exciter fail.

FIGURE A-5

This relay mounted in the DX-50 switched between the internal oscillator in the transmitter and the HD Radio exciter. It was controlled by the interface box sitting on top of the transmitter.

The transmitter tuned up nicely into the dummy load at 10,000 watts. We tried the transmitter at 50,000 watts and it ran just fine. We then put the transmitter on the air at 10,000 watts and fine-tuned the spectral regrowth. Finally, we put the transmitter on the air at 50,000 watts. It was happy for, oh, about five minutes when the transmitter started to glitch. It was acting as if it were seeing problems with the antenna, and muting its RF output momentarily in an attempt to correct the problem.

This is where I first learned that the output monitor circuitry in the transmitter could be too sensitive with the HD Radio signal present. While the transmitter was not being overmodulated in the negative direction, the HD Radio subcarriers add significant power to the sidebands of the transmitter that a normally adjusted output monitor would not expect to see. The output monitor was acting as if it were seeing a “zero crossing,” and muting the transmitter to protect itself.

FIGURE A-6

The rear of the iBiquity test exciter. Note the RJ-45 plugs connecting Cat-6 cable to the interface box on top of the transmitter.

Zero crossing. What the heck is that? If you observe the output of an AM transmitter on an oscilloscope, you will be observing it in a time domain. As you watch the RF envelope, you will see the RF carrier, and will normally set the level of this carrier on the screen as your reference. When modulation is applied, you see the amplitude of this waveform increases and decreases: an increase of the 2X unmodulated carrier voltage is 100 percent positive modulation, the carrier being pinched off at zero volts is 100 percent negative modulation.

If you are using a tone to modulate the carrier and then increase the level of that tone, you will overmodulate the transmitter in the negative direction. This causes two things to happen. First, the audio signal will clip in the negative direction as, in the time domain, you cannot have more than 100 percent negative modulation. Second, this clipping will cause spurious RF signals, or splatter, to be generated by the transmitter.

FIGURE A-7

Before a procedure was approved by Harris to desensitize the output monitor of the DX-50 transmitter, a temporary measure was approved. It involved grounding one of the fault inputs on the output monitor board to prevent false zero crossing trips.

With complex audio, you will observe something interesting on the scope. You will see what appears to be the negative peak crossing over the zero line. This is called a zero crossing. Most modern AM transmitters, if they detect a zero crossing (or, for that matter, a pinched off carrier), will briefly kill the output amplifier’s output for protection. Metal Oxide Semiconductor Field Effect Transistor (MOSFET) amplifiers do not like to be underdriven, as this will destroy the transistors. A zero crossing appears to the transmitter that it has lost RF drive, and it will react in self-defense.

In the frequency domain, on a spectrum analyzer, you know that the AM carrier does not go away. But the effects of negative overmodulation can be seen as spurious emissions of intermodulation products produced by the transmitter. This is not a desirable condition and is, in fact, illegal to operate this way.

The brief (we’re talking milliseconds) carrier clip off will cause pops in the demodulated analog audio. Because the carrier is literally going away when the transmitter protects itself, an HD Radio receiver will lose its reference and will not be able to lock onto the HD Radio subcarriers. It, therefore, will not be able to produce a digital audio output.

We double-checked the modulation of the transmitter with an oscilloscope. We also double-checked the modulation of the internal reference carrier on the exciter. No overmodulation in the negative direction was observed. Harris had us retune and desensitize the output monitor on the DX-50, and the transmitter then ran fine. I do know that the protection afforded the DX-50 by the output monitor was not compromised, as we took several lightning strikes and did, in fact, have an antenna system problem while this transmitter and exciter combination was on the air, and the transmitter protected itself just as it was supposed to.

FIGURE A-8

The spectrum of the iBiquity test exciter installation at the old WOR Lyndhurst facility.

At that time, WOR did not have stereo capability in its studios. Therefore, our audio was monaural, feeding both the left and right channels of the exciter with the same material.

The original audio codec was well, it was OK. It wasn’t anything to write home about, but it did sound better than standard analog AM radio.

I went to the transmitter site several times in the evenings, after the HD Radio signal was shut off (per FCC regulations at that time), and fed the digital audio processor with stereo music from a CD player, using a test radio from iBiquity to monitor the sample output of the exciter. The audio was well, it was OK. Once again, nothing to write home about, but it was definitely a different sound than analog AM radio. It was during this experimentation that I discovered a big problem with the first codec.

FIGURE A-9

When WOR put its HD Radio signal on the air with the iBiquity test exciter, the station had old Pacific Recorders and Engineering System 1 consoles – and they were mono. But, the output was clean and provided good test audio.

I like to push systems for all they are worth. So the music I was testing with had great stereo imaging. And, I brought along things like Beatles songs where the vocal is on one channel while the instrumentation is on the other. These songs just didn’t sound right on the digital channel.

Suspecting leakage between left and right channels, I put WOR’s audio back on the exciter. It sounded fine in digital. Then, I removed one channel of audio, so the exciter was being fed left channel only with nothing on the right channel. Bingo! The left channel sounded OK. The right channel, which should have been dead, had what sounded like splatter from the left channel in it, and this splatter was 26 dB from program level. As long as both channels were being fed the same material, the digital audio sounded fine. As soon as there was a significant difference in what each channel was being fed, the codec was producing artifacts. This was unacceptable. It should be noted, however, that the analog audio on WOR was unaffected by this.

FIGURE A-10

Orban audio processors were used in the original HD Radio installation at WOR. The top unit is an Orban 6200, used to process the digital audio signal. The bottom unit is an Orban 9200, used to process the analog signal.

Several weeks after informing iBiquity of this phenomenon, and demonstrating it for them, they brought out a new load of software for the exciter. This load contained the HDC codec that is presently being used. It was night and day over the original codec, and it eliminated the artifact production with single channel audio.

Over the course of operation of WOR with the iBiquity test exciter, several things were observed which resulted in changes to the system for commercial production. One day, the DX-50 popped off the air, and we could not restart it. We put the auxiliary transmitter on the air and went to the transmitter site.

There, we found the exciter running – fat, dumb, and happy. Or so it appeared. The DX-50, however, was showing that it had lost its RF input. Checking the output of the exciter showed that it was not putting out RF. In theory, if the exciter failed, the interface box on top of the DX-50 was supposed to switch the transmitter back to analog-only mode. But, as it turned out, the box needed to be told that the exciter was in trouble by the exciter itself. Unfortunately, the exciter didn’t know it was in trouble, so the box never received a trouble signal from the exciter and therefore never switched. This resulted in the addition of RF sensing in the exciter monitoring protocol. If the RF output of the exciter quits, regardless of what the exciter actually thinks it is doing at any given time, the transmitter will now revert back to analog-only operation immediately.

I mentioned the iBiquity test radio in a previous paragraph. For the first two years of operation, the only way we could listen to the digital audio on WOR was with an iBiquity test radio. We had one at the transmitter for monitoring, and one at the studio for monitoring. This test radio was also a computer running the Linux operating system. It weighed about 15–20 pounds, and was in a sizeable case. It was not what you would call portable, and required a connection to an AC power outlet to operate.

For the most part, this exciter ran well. It was on the air at WOR from October 11, 2002, until the day we took the Lyndhurst, New Jersey, facility off the air in favor of a new transmitter facility built in Rutherford, New Jersey, on September 8, 2006. This exciter would usually need to be rebooted roughly every six months.

If we did not reboot the exciter roughly every six months, it would slowly start to fail, eventually just locking up, forcing a switch back to analog-only operation. We found we could tell when the exciter was starting its spiral downward by using the graphical user interface, or GUI, to check on any alarms in the system. If the GUI was sluggish, it was time for a reboot. WOR is now using Harris DexStar™ exciters. They tend not to exhibit this problem.

One time we had the exciter start to fail, and it locked up on the analog side. This lock up caused an interesting thing to happen. The analog audio was repeating the 8.4 seconds of audio that aired right before the exciter locked up. Over and over and over. We were able to force the system to analog-only mode with the remote control to restore normal audio to the air. It should be noted that this particular phenomenon happened only once.

A.2  Signal coverage

During the first two years of operation with an HD Radio signal on WOR, we, of course, were curious as to the coverage we were getting with the digital channel. IBiquity, obviously, was interested as well.

FIGURE A-11

WOR’s new facility uses Harris 3DX-50 transmitters and Harris DexStar exciters. The exciters are in the rack between the transmitters.

Without a radio, however, it made my attempt at discovering our digital coverage a tad difficult. I did have the iBiquity test radio, but that would have required either a generator or inverter to tote around in the car, and that wouldn’t work well.

IBiquity had a van outfitted for monitoring. This van contained an iBiquity test radio that was set up to output data to a mapping computer. The mapping computer kept data on the position of the van through GPS. It also had a data input from a spectrum analyzer, and would log spectrum shots along with information such as data errors, actual digital capture by the radio, and received signal strength. So I had no doubt that the information iBiquity would provide would be fairly accurate.

iBiquity ran test after test, and finally gave me a map showing our digital coverage in several major directions. We had coverage in our secondary lobe to the city line of Philadelphia (and, in fact, we listened to WOR’s digital audio at the NAB Radio Show in Philadelphia). We had coverage up to almost Allentown, Pennsylvania, to the west, which skirts the side of the secondary lobe, but isn’t quite in a null. We had coverage to three-fourth the way out on Long Island, which is good, as the ground conductivity over Long Island is awful. The map showed coverage to Waterbury, Connecticut, along I-84 to the north, and also showed coverage just beyond the New York/New Jersey border on route 17. Of course, the naysayers said that the map was fabricated by iBiquity.

FIGURE A-12

This map of New York and the surrounding area was produced by iBiquity Digital Corporation. It shows the digital coverage of WOR Radio along various routes.

When I received my first HD Radio, I took a drive along the same routes iBiquity used to determine the digital coverage of WOR. The radio is a Kenwood and includes the HD Radio module. It was installed in my Ford Explorer and used the stock radio antenna with which the vehicle was equipped.

The map provided by iBiquity turned out to be pretty darn accurate. Where the map said the radio would blend to analog, it did. Where the map said the radio would have digital coverage, it did. Where the map said digital coverage would stop, it did.

FIGURE A-13

The Kenwood HD Radio car receiver used to check coverage of the WOR signal.

I brought along a Potomac Instruments field intensity meter with me on my drive, and purposely made a signal level measurement when I finally lost digital coverage. Digital coverage was lost between 0.5 and 0.7 mV/m in every direction. At this point, the WOR analog signal was fairly noisy. This, basically, is at a point where I think most listeners would consider the signal too noisy to listen to in analog. I consider the coverage to be good.

In NYC, as you can imagine, there are many areas where an analog AM signal is noisy, at best. There are many places where the WOR signal is barely listenable.

The coverage in the City for the WOR HD Radio signal is rather impressive. I recall the moment when I was taking a reporter for a car magazine around for a driving test of both FM and AM HD Radio signals. We took a path for both signals that took us up the West Side Highway, getting off at the 72nd Street exit and driving through Central Park behind a rock formation, then down 5th Avenue to 34th Street, then heading west to the West Side Highway again. We drove this path four times: once listening to the analog FM signal, once listening to the digital FM signal, once listening to the WOR analog signal, and one last time listening to the WOR digital signal.

FIGURE A-14

The WOR digital audio dropped out right before this toll booth along the Pennsylvania Turnpike, I–78 in Allentown, Pennsylvania.

While you might not think it, FM signals in NYC tend to suffer from severe multipath distortion. The reason is all the reflections between the buildings. On the West Side Highway, the Empire State Building, home to most of the FM transmitters in NYC, is shielded from view by buildings. You also have reflections off the Hudson River to contend with.

On the analog FM drive, it was obvious up the West Side Highway that the FM signal was annoying to listen to with all the picket-fencing going on. It would clear up slightly until we hit the rock formation in Central Park, when the multipath would start up again, then it would clear up driving down 5th Avenue and on 34th Street.

FIGURE A-15

A map showing where the WOR digital signal dropped, just outside of Allentown, PA, 72 miles from the WOR transmitter site.

The digital drive produced a perfect audio signal. No multipath. No noise of any kind.

With the AM signal, the same path would find the analog signal noisy on 34th Street, good on the West Side Highway, sketchy along 72nd Street and through Central Park, and OK along 5th Avenue. The digital drive produced great audio. No dropouts, no noise, and no degradation of any type.

After we finished, the reporter asked me if I would drive him back to his office. We headed east on 40th Street to pick up 2nd Avenue and head north. As we approached Lexington Avenue, I almost panicked. The corner of Lexington and 40th Street has a large building on each corner, which forms a box. I recalled driving with iBiquity and observing on a spectrum analyzer that the entire AM broadcast band would disappear into the noise and, on the night of testing with iBiquity, we would lose the digital signal and end up listening to a noisy analog AM signal.

As we approached Lexington, the light turned red and I was forced to stop in the box formed by the buildings. Interestingly, the radio was still producing a digital signal from WOR. The reporter asked me what the analog signal sounded like there, as he noticed we were boxed in. So, I forced the radio to analog mode.

FIGURE A-16

There are several tunnels connecting Manhattan to the outside world. The Lincoln Tunnel has a ‘leaky coax’ system, which allows motorists to hear AM signals while in the tunnel under the Hudson River. The system does not pass an HD Radio signal, but the change from the coax system to the HD Radio signal when you emerge from the tunnel is dramatic.

We were greeted with a lot of noise, with barely any audio. Frankly, we could not understand the analog signal. I reset the radio for digital mode and – much to my surprise – the digital signal kicked right in. Needless to say, this was an impressive test for AM HD Radio.

A.3  Nighttime AM HD Radio testing

In December 2002, I participated in nighttime testing of AM IBOC with iBiquity on the signals of WOR, New York, and WLW, Cincinnati.

Both WOR and WLW obtained Special Temporary Authority from the FCC to operate our HD Radio signals at night during this test period. For these tests, iBiquity installed a special load of their software on the exciters at both WOR and WLW. The software was set up to turn the HD Radio subcarriers on for 10 minutes, then turn them off for 10 minutes. The software was synchronized between the stations so that, if the WOR HD Radio signal was on, the WLW signal would be off, and vice versa.

FIGURE A-17

This audio spectral graph was made from a recording I had made of the WOR signal in the Lincoln Tunnel. It shows the frequency response differences in the received audio on the leaky coax system, the off air analog signal, and the off air digital signal.

I wrote an article for Radio Magazine titled “Three Nights in an iBiquity Van.” The unedited version is given in the following paragraphs.

WOR Radio, the 50,000 watt monster in NYC, adopted the iBiquity HD Radio (IBOC) system, and officially became New York’s first digital AM radio station at 9 AM, October 11, 2002. WOR was initially chosen as a test station for IBOC, as iBiquity was looking for a blowtorch of a signal near a large city for testing of IBOC coverage and compatibility. Buckley Broadcasting feels that IBOC can only help AM radio, and formally adopted IBOC transmission on WOR the day after it was approved by the FCC.

Thus far, there have been many positives in our transition to IBOC. First and foremost is our daytime digital coverage. WOR has been purposely running our IBOC signal –6 dB from where it should be at the suggestion of iBiquity. The secondary lobe of our directional pattern puts 85,000 watts toward Philadelphia from our transmitter site in Lyndhurst, New Jersey, 95 miles away. Running at –6 dB, our digital coverage, solid, goes as far as the Philadelphia city line, our ½ mV/m boundary. But the $64,000 question is: what would our nighttime IBOC coverage be like, and what would the sideband interference be like?

On the evenings of December 2–5, 2002, I found out while riding around the New York metropolitan area in an iBiquity test van with Russ Mundschenk of iBiquity. There were three nights of testing to perform. The first night was for analog compatibility. The second night was for digital coverage with an IBOC interfering station. The final night was for nighttime digital coverage without an IBOC interferer. The station which would play the role of interferer was WLW, first adjacent station to WOR on 700 kHz, with WOR at 710 kHz.

For compatibility, we drove west out I-78 to a point 51.7 miles from the WOR transmitter site. It was at this point where WLW’s skywave, as referenced to WOR’s carrier, was –10 dB on an average. This was measured with an HP spectrum analyzer mounted in the iBiquity van, connected to a 31-inch whip antenna on the roof. The location we chose was the parking area of a state highway maintenance area, away from the streetlights. This location proved to be very noise free, and the skywave conditions couldn’t be better at 10 PM. The spectrum analyzer, set to show stations from 660 kHz to 770 kHz, showed carriers neatly spaced every 10 kHz. What was really amazing was seeing the skywave phenomenon happen on the spectrum analyzer. You always hear it on a radio while listening to one station, but to actually see only one station that can be affected by atmospherics while the others remain the same is fascinating. The signal from WOR was measured with a Potomac FIM-41, and found to be approximately 0.75 mV/m, roughly WOR’s 0.5 mV contour, which was outside the secondary lobe but not yet in the null at 342 degrees. The WOR signal was fairly constant, indicating a ground wave signal. The WLW signal would vary greatly at times, indicating it was sky wave.

The IBOC exciters at WLW and WOR were set so that, for a 10-minute span, WLW’s IBOC carriers were on for 1 minute, off for 1 minute, and in this 10-minute time period we monitored WOR. We then changed our monitoring to WLW while WOR’s IBOC carriers were toggled on and off in the same sequence. We used six radios to monitor these signals. Five were standard consumer models that you would find at Best Buy, and included a home Technics tuner and a GE SuperRadio III. The sixth was an iBiquity test receiver.

It was my observation that when the WLW carrier was –10 dB as referenced to WOR, and WLW’s IBOC carriers were on, there was minor noise under the WOR signal: minor to the effect that the radios needed to be turned up to hear it. If WLW’s carrier met or exceeded WOR’s carrier level (and WLW exceeded WOR by +10 dB at times), the noise level came up under WOR’s signal, but it was far from objectionable. Clearly, the WOR analog signal within our 0.5-mV contour would be useful.

Conversely, when WOR’s IBOC carriers were on, and WLW’s carrier was –10 dB, the WLW signal was rather noisy. Not to the point of tuning them out, particularly if you were a fan of the basketball team’s game being broadcast, but it was annoying nonetheless. If WLW’s signal decreased beyond about –15 dB, their signal was pretty much unlistenable. If, however, WLW’s signal level were equal to or exceeded WOR’s signal level, the noise was only slightly audible and the signal was very useable. It should be noted that when the WLW signal was unlistenable and WOR’s IBOC carriers went off, the WLW signal was still unlistenable due to either a station at 700 kHz coming in under them, or sidebands from a station at 690 kHz splashing them. As a generality, I would say that it was not strictly the IBOC carriers that made the signal unlistenable.

After an hour, we packed up and drove to a location about 72 miles from the WOR transmitter, near Bethlehem, Pennsylvania. At this point, the WOR and WLW signals were pretty equal, and judging from the signal variations on the spectrum analyzer, they were both predominantly sky wave. It was found that when both signals were sky wave, the noise level on the desired signal was somewhat annoying when the interfering carrier was –10 dB. The noise level increased if the interferer increased equal to or exceeding the desired carrier, but I considered the signals to be listenable. Annoying, but listenable.

The following night, we took four routes: one took us out I-78, another took us south down the Garden State Parkway, a third took us through Manhattan, through the Queens Midtown Tunnel, and east out the Long Island Expressway, and a fourth took us up WOR’s null north on New Jersey Route 17.

The digital signal held out and finally fell apart 50 miles to the west, 52 miles to the south (which was in WOR’s minor null), 53 miles to the east on Long Island (which is in the major lobe, but the ground conductivity on Long Island leaves much to be desired), and 20 miles up through our null. It was interesting to note that, driving through our null, the farther we were from the transmitter, the less upper sideband WOR had. Yet the digital signal still decoded, and in general fell apart close to the 0.5-mV/m contour.

In Manhattan, we drove straight down 40th Street. There are several areas on 40th Street before Broadway where the WOR analog signal is fairly noisy. The digital held up. The only place we had an issue was at the corner of 40th Street and Lexington. There are several large buildings here, and it must be a combination of reflections and shielding, but the entire AM band from 660 to 770 kHz as shown on the spectrum analyzer all but disappeared. Of course, no signal, no digital or analog. Coming out from this area, the digital locked right in again.

The Lincoln Tunnel and Queens Midtown Tunnel employ a leaky coax system where the Metropolitan Transportation Authority (MTA) can insert traffic advisories on all NYC AM stations. This system completely stripped off the IBOC sidebands. However, sitting under the roof at the toll plazas at both tunnels, the digital kept cranking.

The last night, testing was done with only WOR’s IBOC on. Coverage was substantially the same as with WLW as an IBOC interferer, leading one to believe that this interference does not affect an IBOC signal.

Quite frankly, listening to the digital on the road was quite amazing. Joey Reynolds had a guest singing live Christmas Carols with an acoustic guitar, and it almost sounded like the guy were in the van with us, it was that good. It was also amazing, realizing this was an AM station, and it wasn’t fuzzing up under bridges and overpasses.

My opinion is that, if the IBOC carriers were backed down from –3 to –6 dB at night, the digital coverage would be less, but there would be less impact on analog signals. Certainly, digital coverage is adequate at night on the WOR signal.

Another thought is this. If you put up a Class-B FM station, you expect to lose signal once you reach about 50 miles out, depending on terrain. No one bats an eye. Maybe we should start considering AM in this same vein. The audio quality of the digital signal, in my opinion, far surpasses the analog quality. Is IBOC the savior AM radio has been in search of? I don’t know. But I do know that, personally, I would rather have the digital audio to listen to rather than the bandwidth-restricted scratchy analog. But that’s not up to me to decide. And that’s why WOR is testing, and proud to be HD.

A.4  Nighttime hybrid AM HD Radio operation

On September 14, 2007, it became legal to operate the HD Radio signal at night on AM radio stations in the United States. WOR turned its HD Radio signal on at 12:01 AM on the morning of September 14, the time it became legal to do so.

The decision to allow nighttime AM HD Radio operation by the FCC was a bold move – and fit in with the wording of their Report and Order on the creation of IBOC rules. The reason it was a bold move was that, while many studies of nighttime AM HD Radio operation have been done (the WOR/WLW tests being just one of them), exactly what would happen when many AM HD Radio signals were lit up at night was somewhat of an unknown.

The hybrid AM HD Radio signal is designed to provide digital coverage out to a station’s 0.5-mV/m contour daytime – and at night if conditions allow. It is designed so that, with the station spacing and regulations in the United States, it should not infringe upon the 2-mV/m nighttime interference-free contour of a neighboring adjacent station. There are, of course, some exceptions to this, but they are relatively few.

And herein lies the controversy with nighttime AM HD Radio operation. Many people mistakenly think that an AM radio station’s coverage is wherever you can hear it. That is wrong. The FCC rules define the coverage contours of AM radio stations. They also define contours where you can expect to receive and not receive interference to your signal. And it is important to note that these contours are based on carrier frequency – not on modulation. And on the basis of this information, it becomes clear that these coverage contours are used to determine spacing between stations.

Section 73.44 of the FCC rules defines your occupied spectrum – in other words, how much space your modulated signal can take up. Under Section 73.44, the hybrid AM HD Radio signal is perfectly legal. In actuality, you could modulate the AM transmitter with broadband noise, tailored by an equalizer to fit under the mask defined in Section 73.44 – and your station would be legal. Adjacent stations may not be happy with you, but you would be legal under the FCC regulations. Contours and occupied bandwidth are two things that some persons simply do not understand.

That being said, on the morning when nighttime HD Radio operation was legal, many of us were greeted by complaint e-mails. These e-mails were generated by frustrated DXers who, they claim, could no longer listen to certain stations because of an adjacent HD Radio station.

I received one from a guy in Maryland, and was actually going to take this guy seriously until I hit the anti-IBOC rhetoric in his note. He claimed that the WOR signal was wiping out reception of both WLW and WGN at his home.

I called a friend in Maryland who happened to live two towns over from where this guy stated he lived. My friend had been up at 1 AM – and purposely listened to 700 kHz, 710 kHz, and 720 kHz. He said he heard no problems and had no trouble at all listening to WLW or WGN at his home. I would tend to believe my friend.

I live near Newburgh, New York, about 50 miles due north of NYC. I have listened on my car radio, on a C. Crane CCRadio Plus using its internal antenna, and on my Yaesu FT-857D Ham rig using a 110-foot dipole antenna. I am specifically interested in how the analog signals of adjacent stations are affected by IBOC operation. In the following paragraphs, my observations of nighttime HD Radio operation are given, with the HD Radio stations in italic type.

700 WLW, 710 WOR, 720 WGN – I live in WOR’s null, and WOR’s signal is normally very noisy at my home. It did not get noisier. The signal still sounds the same, with the exception of about an hour before sunrise when the signal does get noisier. It should be noted that WOR’s signal level in my driveway is 0.3 mV/m. I have no trouble listening to WLW or WGN, and have listened to them closer in to WOR’s transmitter. When the WOR signal is over 2 mV/m, WLW and WGN do get noisy. But, this is outside their 2-mV/m contours.

800 CKLW, 810 WGY, 820 Station not identified – I live about 100 miles south of WGY’s Schenectady, New York, transmitter site. CKLW comes screaming in here like a local at night and is perfectly listenable. WGY sounds fine. I can hear a station on 820, noisy but not IBOC noise, but it is there.

1530 WCKY, 1540 WDCD, 1550 Station not identified – I live about 100 miles south of WDCD’s Albany, New York, transmitter site, and I am in somewhat of a null of their signal. I have no problems hearing WCKY, and no problems hearing one of the several stations on 1550.

1070 CBA, 1080 WTIC, 1090 WBAL – I live about 100 miles from WTIC’s transmitter. WTIC is allowed to remain 50,000 Watts, nondirectional until two hours past sunset, so after sunset, WTIC is still pumping a full 50 kW in my direction. On their night pattern, I would receive considerably less signal. WTIC comes in like a local. CBA, while weak, is there and perfectly listenable. WBAL just screams up into this area, and actually shows more signal on my Ham rig than WTIC does, and is perfectly listenable.

1100 WTAM, 1110 WBT, 1120 KMOX – I can hear all three stations with no problems.

760 WJR, 770 WABC, 780 WBBM – I am about 46 miles due north of the WABC transmitter in Lodi, New Jersy. WABC puts greater than 2 mV/m at my home. The WABC signal was clear. WJR was difficult to listen to, as it was very noisy, but it could be heard. The same case is with WBBM.

It should be noted that, to my knowledge, none of these stations is experiencing a problem with their signal inside their 2-mV/m contours.

While some distant listeners are experiencing problems with their hobby, I do not believe that most listeners have noticed any difference in the signals of their favorite stations since the inception of nighttime AM HD Radio operation. There are, of course, exceptions. But overall, I do not think there is a major problem. The nighttime AM radio landscape has begun to change.

A.5  Practical considerations for the implementation of AM IBOC

The text in the next paragraph in this section is a paper I prepared and delivered for the NAB show in 2003. It is basically the opening story of this chapter as it was presented to National Association of Broadcasters (NAB).

INTRODUCTION: After many years of development, the time to begin implementation of IBOC digital transmission on the AM band has arrived. For some, it will be an expensive proposition.

WOR-AM in NYC was approached by iBiquity Digital Corporation in regards to becoming a test station for IBOC-AM. We agreed and, while the final software load for the exciter was not yet ready, WOR turned IBOC on and committed to it on a permanent basis on October 11, 2002, the day after the FCC approved the use of IBOC. On this day, WOR became New York’s first digital AM radio station.

This paper serves to inform you of some of the practical considerations for implementing AM IBOC at your facility. We will discuss what WOR did and did not do to implement IBOC, and we will discuss some of our findings that you might not have thought of.

WHAT YOUR STATION NEEDS TO IMPLEMENT IBOC: Unfortunately, there has not been a plethora of information available on the implementation of IBOC. If you go by the information in ads in the trade publications, it appears that you need to have a digital studio sources feeding a digital console, which feeds a digital STL, a digital processor, and an IBOC exciter and your transmitter. About the only thing I haven’t seen in the ads is that you should also have a digital announcer.

When WOR implemented IBOC, we were using original, 1978 vintage Pacific Recorders and Engineering System One consoles. They were mono and they still sounded pretty good. While 2003 was going to herald a reconfiguration of the WOR facility, WOR was still using the System One consoles in October 2002. And if you look around the typical WOR control room, you find a rather eclectic mix of older technology and new technology. Everything from turntables, which are used during the Joe Franklin program on Saturday night (and Joe still brings 78-RPM records in to play), to an ENCO system. I would think that the WOR studio situation is fairly typical of most stations in the United States.

To sum up our studio situation with the implementation of IBOC, WOR feeds analog audio into an older analog console. The consoles were in reasonable shape and had reasonable specifications that were still being met. There was no immediate need to go out and immediately purchase five brand new completely digital consoles. If your consoles are still in reasonable shape and sound good, there should be no need to replace your consoles to implement IBOC. The reason WOR is reconfiguring our studios, which will include new consoles, is that most of the semiconductors and switches used in these consoles are no longer available. If they were not next to impossible to repair, we would most likely be keeping the PR&E consoles for another few years.

Our Master Control room contains passive relay studio switchers, which are stereo. Presently, the right channel is used to switch paths for the cue tones used for the WOR Radio Networks. As part of taking WOR stereo over the next few months, these switchers will be added onto so that we have a different path that will follow the main switchers for the cue tones. At the present time, there is no need to change a set of studio switchers that still work fine and we can still obtain parts for.

One reason that the reconfiguration of the WOR facility will maintain analog audio is our top-hour time tone. I have yet to find anyone who makes a small, rack-mountable, two-channel stereo mixer so that we can maintain the automatic insertion of our time tone. Deleting the time tone is not an option, as it has been on WOR for decades, and is actually constructed using a middle-C oscillator module from a Hammond organ. Many times during the day WOR is automated, sometimes we are live, and automatic insertion of the time tone is a must. So for this aspect, WOR must maintain analog audio through the studio switchers.

Once the audio passes through an Orban 8200ST processor, it is fed through a distribution amplifier to feed our STL paths. WOR maintains five STL paths to our transmitter in New Jersey. The first is an Intraplex STL/TSL unit that transmits bidirectionally on a Harris Aurora 5.8-GHz spread spectrum radio. This allows us to bring back our satellite channels from Jersey to Manhattan, gives us a 15-kHz nondata reduced path to the transmitter in stereo, plus allows for five data channels and two telco paths between the transmitter and studio. Because the Intraplex rack is utilizing all the time slots on the T1 created with the Aurora, we could not put Intraplex’s “IBOC” card into the unit and maintain our data and telco channels. This means that the WOR audio goes into and comes out of the Intraplex analog, and is 15-kHz bandwidth. We did not see a reason to increase the bandwidth to 20 kHz, as the IBOC will pass up to 15 kHz.

Our second STL is a 950-band Moseley DSP-6000, which is stereo. While this uses data reduction, it is perfectly adequate as a second stereo backup for WOR.

Our third STL is a 950-band mono Moseley PCL-6010 transmitter and 6020 receiver. Once again, as a third backup, this STL is perfectly adequate and did not need to be replaced.

Our fourth path is an 8-kHz equalized phone line which, at times, has been known to hum or whistle quietly. Once again, if we need to resort to the phone line, it is better than the alternatives (i.e., the sounds of silence), and this STL path will be adequate even with IBOC.

Our final STL path is a G.722 ISDN codec, which is intended for extreme emergency use, such as if we needed to abandon the Manhattan studios and broadcast from elsewhere. Under these circumstances, this path would be more than adequate even with IBOC.

Bottom line, I saw no good reason to replace our studio switchers or STLs with a completely digital path. What we are using sounds good, is reliable, and is perfectly adequate with our IBOC installation. If your analog audio path to the transmitter sounds good and is reliable, there should be no immediate need to go through the expense of replacing your STLs. And while stereo would be nice, if they’re mono, so what? Your present signal is mono. Aim for stereo STLs sometime down the road.

At the transmitter, by necessity due to our need to feed analog audio to our auxiliary transmitter chain, the audio that comes out of our various STL paths is analog, and goes through a passive Broadcast Tools switcher. The WOR transmitter site was rewired in 1997, and I wired it with stereo in mind. The output of the switcher feeds a distribution amplifier, which feeds the auxiliary transmitter chain, still feeds the main chain with an analog signal, and now also feeds an analog-to-digital (A/D) converter. The IBOC equipment requires an AES digital signal. The A/D converter takes the analog audio from the STL switcher and converts it to an AES signal before the ESU unit of the IBOC exciter. If you intend to keep your STL path analog, you will need to purchase an A/D converter. Ours is from Radio Systems.

In addition to certain control aspects it has over the IBOC exciter, the ESU also acts as an AES distribution amplifier. It feeds our Optimod 9200 an AES signal for the analog processing of our signal, and feeds our Optimod 6200 an AES signal for the IBOC portion of our signal. The processor for your analog signal should have AES in and out, as the IBOC exciter accepts AES in only for either the analog or digital signals. And you will need to purchase a separate processor for the IBOC audio. The processor you choose should have digital broadcasting in mind, though an FM processor with the pre-emphasis defeated should work fine. This processor also needs to be AES in and out.

There are a few paths to take when connecting the IBOC exciter to your transmitter. One is to simply connect the exciter and turn it on. The one chosen by WOR was to connect a relay to the IBOC transmitter to exciter interface unit, in addition to our analog output of the Optimod 9200. If the IBOC exciter were to fail, the relay drops out and switches the RF input to our DX-50 transmitter back to the internal oscillator. Having our analog Optimod output available through the interface box (which contains an audio relay) would cause the transmitter’s audio input to change from the IBOC exciter’s output to the standard output from the Optimod. This would keep WOR on the air in the event of an exciter failure, and it proved to be the correct decision one morning when, after running a special program on the exciter to switch the IBOC carriers on and off for testing, the exciter spontaneously rebooted when I attempted to kill the toggling program. The DX-50 simply dumped when the RF input switched, and came right back up after about one second. No issues, no lost time.

Now that you know that the exciter is a computer, running the IBOC program under Linux, how is the power at your site? You should put the exciter, the D/A converter, and the processors, as a minimum, on a good uninterruptable power supply that is the type which is always on line and has surge protection. This will help prevent the Linux computer from locking up with power hits.

Another consideration with the exciter is the interconnection to the transmitter. The RF and audio or, as they are known in IBOC lingo, the phase and magnitude signals exit the exciter on RJ-45 connectors. The connection between the exciter and the interface box in your transmitter should be shielded Cat-5 cable. The connection between the interface box and the transmitter should be an RG-58 cable terminated in BNC connectors for the RF, and regular single-pair shielded audio cable for the magnitude connection, and terminated in spade lugs for the transmitter, and bare wires for the Phoenix-type connectors on the interface box. You will need to determine how your transmitter will accept the external RF input and plan a relay accordingly if you intend to add the fail-safe switching described earlier. In the WOR installation, we discovered during installation that our supplier shipped nonshielded RJ-45 connectors in a package marked that they were shielded. We simply took the shield wires on the Cat-5, crimped spade lugs on the ends, and attached them to the nearest screw with a ground connection. Whatever works.

Your transmitter is going to be a big consideration. IBOC will not work with a plate-modulated transmitter. It is questionable if it will work with a tube pulse duration modulation (PDM) transmitter, and will not work with a Doherty modulated transmitter. The Harris DX series of transmitter, of which WOR has a DX-50, is pretty much plug and play for IBOC. Broadcast Electronics and Nautel transmitters are likewise.

If you have a PDM transmitter that is solid state, you will need to check with the manufacturer to see if the transmitter will pass IBOC. Most transmitters have a modification kit available, as the audio passband must be 50 kHz through the PDM modulator. Most likely the PDM frequency and the PDM filtering sections will need to be changed out. Your transmitter needs to have a pretty flat audio response that is 50-kHz wide, and the RF section needs to have a fairly low amount of phase noise, or, as we know it from AM stereo days, Incidental Quadrature Modulation (IQM) or incidental phase modulation (IPM). You should check with the manufacturer and make a decision as to whether you should purchase a new transmitter.

Your antenna is the other big consideration. WOR employs a three-tower dogleg directional array. There are detuning skirts, two on each tower, for 1010 (WINS) and 1190 (WLIB) to electrically shorten the WOR towers at these frequencies. The tuning houses have detuning networks for 1010, 1190, and 620 (WJWR), in addition to traps for 1010, 1190, and 620. To say that our array is “challenging” for a fairly flat path at and about 710 kHz is an understatement.

We had Tom Jones from the firm Carl T. Jones Corporation redesign our coupling networks and phasor in 1997. By putting line stretchers in each tuning house, making all the tuning networks phase lead (there had been one phase lag previously), and changing a few components in the phasor, as well as changing the common point from 75 ohms to 50 ohms, the WOR common point is now fairly flat, both resistance and reactance, from 690 to 730, ideal for IBOC. The lower sideband is favored slightly, and this can be seen in the lower IBOC sideband being about 1.5 dB higher than the upper IBOC sideband.

The load presented to your transmitter needs to be fairly flat ±15 kHz from carrier in both resistance and reactance. At the very least, if asymmetry is present, the impedance needs to be asymmetric equally and opposite on both sides of carrier. You should sweep your system to see what the transmitter is actually looking into. You may need to hire a consultant to flatten things out. If WOR’s system could be flattened, there is indeed hope for your system.

OTHER CONSIDERATIONS: In addition to the technical considerations mentioned previously, there are a few other considerations that we learned with our foray into IBOC. The first revolves around the inherent approximate 8.5-second delay introduced into the analog audio once you have converted to IBOC. The reason for this delay is that the radios are designed to blend back to analog should the digital signal encounter a problem. This prevents the audio from going away rather abruptly. Obviously, if you do not delay the analog 8.5 seconds, this blend will be choppy at best.

WOR is a talk station and, consequently, operates with a profanity delay most of the time. The exceptions are the top-of-the-hour newscasts and portions of The WOR Morning Show. Obviously, the talent cannot monitor air when air is delayed by 8.5 seconds. To counter this, we changed the air-monitoring position in all studios to feed off the directional antenna (DA) that feeds the STLs. This gave the talent a real time audio feed of the audio before it left for Jersey.

The problem with this is that the air talent were used to hearing the signal all “pumped up” and heavily processed. Some had trouble adjusting to the cleaner, flatter signal. To correct this, we took an old CRL AM system that had occupied rack space at the transmitter and inserted it into the feed the studios were getting. This gave the talent their pumped up audio and made them happy.

The next problem to resolve was: if you can’t listen to air (even in the Master Control room), how do you know if you’re on the air? Oh, you’ll know if the transmitter is on the air. WOR has computer monitors in each control room that displays the transmitter status at a glance. But what about an audio failure?

We installed two silence sensors. One on the feed into the studio to transmitter link (STL) transmitters, the other on the analog modulation monitor. They are set for 10 seconds. If either the feed to the STL quits (i.e., somebody switches to a dead studio) or the transmitter audio quits (i.e., STL failure or processor failure), the thing screams. We should bless Radio Shack for having loud, obnoxious piezo beepers available. There are also silence sensors at the transmitter site feeding status lights on the remote control. If the silence sensor goes off, a simple glance at the transmitter screen will show if it’s at the transmitter site or studio. If the problem is at the transmitter site, the screen will show if it is the active STL receiver or the transmitter where the audio has quit.

Another issue: program on hold on the office phone system. It had been fed off the same DA that feeds the studio monitors. Rick Buckley, our President, observed that the hold audio was 8.5 seconds ahead of what he was hearing on his office radio in Greenwich, Connecticut, and thought it was rather confusing. The program on hold feed now comes from the modulation monitor directly.

What about the time tone? We have people who literally set their watch by the WOR time tone. The switchboard floods with complaints if we patch the time tone out briefly for maintenance. I could just imagine what would happen if it were 8.5 seconds off. Can you say complaints to the FCC for giving erroneous time information? WOR utilizes a programmable timer made by ESE to fire off various things, like the backup feed mini disk machines for The WOR Radio Networks. We programmed a position on this timer to fire the time tone at 59:51.5, so it hits air at exactly straight up on the hour. We have had no complaints.

Live sports? This is an issue that is being worked on by iBiquity. Your station can ramp in and out of the 8.5-second delay but, as mentioned previously, this may cause blend problems between analog and digital signals on radios. iBiquity has told us they are working on setting a bit in the data stream to inform listeners that the analog feed is real time for a live sporting event, and the listener can choose to listen to the digital or analog feed. This will allow persons listening in the stands or watching the game on TV and listening to WOR to hear the game in real time.

One thing that no one considered at WOR was the effect the 8.5-second delay would have on our top hour newscasts. We are formatted to hit the ID/news sounder on top of the hour. At the time IBOC was implemented, the ID cut was 24 seconds long. We normally have a 10 second “brought to you by” announcement at the top of the cast, meaning that we didn’t get to the actual lead story until 34 seconds past the hour. Add the 8.5-second delay to the mix, and we now did not get into the lead story until almost 43 seconds past the hour. While WOR is not a news station, this problem came to light on one particularly busy news day, when every station in NYC had already hit the lead story, gone to the outside reporter, and was into the second story before our announcer was even out of the sponsor announcement.

Our solution was to shorten the ID to under 15 seconds, and advance the clocks in the studio complex so that everything would happen 8.5 seconds early. If we were 8.5 seconds early, it would hit air on time. But there were two problems. First, the digital clocks were locked to our GPS receiver and that couldn’t be changed. Our ENCO system was also locked to the GPS receiver, meaning that all automation-timed events would need to be reprogrammed. Finally, what would we do about The WOR Radio Networks? They need to feed on a real-time basis. If we advanced the clocks, and an operator made a mistake, we would go up at 06:32 rather than 06:40, or come out at 58:42 rather than 58:50, fouling up hundreds of stations around the country.

The solution came in a product manufactured by Symetrix under the AirTools name. The AirTools digital delay units delay audio, TC-89 time code, and contact closures! By setting an 8.5-second delay on the AirTools unit, the TC-89 time code would advance by a corresponding 8.5 seconds. This would make the digital clocks and the ENCO system correspond to the 8.5 second advance of the analog clocks. This meant that, in real time, the network would start at 06:32, but hit the satellite channel at 06:40 as it should. The only time WOR hits an outside source is with Mutual News from 9 PM through 4 AM. Only a handful of automation commands would need to be changed. None of the automated records for the network would need to be changed. And WOR would now be on time.

HARRASSMENT: One of the things that no one considered was the reaction of a certain small segment of the community. These people are DXers. Some belong to an AM Stereo club.

My Chief Engineer, Kerry Richards, and myself were verbally attacked on a personal level on at least two very public message boards. We have been harassed via e-mail and on the telephone. I have had people trying to get me to “admit” that I am operating WOR illegally. I am not. WOR’s operation is perfectly legal under the FCC regulations. I have been threatened with “I’m going to write a complaint to the FCC regarding your illegal operation.” My response has been, “Go right ahead. WOR is operating legally and the FCC is welcome to inspect us at any time. And I would suggest that you should have your facts and documented proof of our illegal operation. The FCC won’t follow up unless you have credible evidence.”

We have had to have the phone number changed at the transmitter site. If we turned the IBOC on after sunset, even after midnight, the phone had rung and we had met with a stream of obscenities and hung up on. The caller ID was blocked and I had no idea how someone got a nonlisted nonpublished phone number. It’s nice to know that some people have nothing better to do than call radio station transmitter sites … at all hours of the day and night … and harass the engineers.

I have been accused of splattering the AM band from 540 kHz to 1030 kHz. I have had people who tell me that I must protect skywave signals coming into the New York metropolitan area at night: incidentally, there is nothing in the FCC regulations requiring me to do so once outside protected contours.

I haven’t been physically threatened at this point, nor has anyone attempted to gain access to the WOR facility. If that day comes, the person or persons involved will find out what happens to them, as the WOR transmitter site plays a big part in New Jersey’s Homeland Security disaster plans. The feds will most likely not take too kindly to the WOR facility or engineers being physically threatened.

In short, your station should brace not only for the positive aspects of what IBOC will bring you, but also for the negativity. It’s amazing how such a small group can have such big mouths. Your promotion department, if you have one, should have a way to handle these situations. Bottom line, in my opinion, is that IBOC helps AM, and that is why you are considering installing IBOC on your AM facility.

CONCLUSION: These are the things that we at WOR needed to take under consideration in our facility when we committed to IBOC. While these items are by no means a complete list, it should help you to see items that will need consideration before you take your AM station IBOC. Each station and set of circumstances is different, and only you can decide what is the best way to handle the issues that may crop up with IBOC implementation.

With well-thought-out implementation, and the ability to react to issues as they pop up, your transition to IBOC will be smooth and fairly uneventful.

A.6  Real-world AM IBOC coverage using a consumer IBOC radio

The next article below was written in 2004 for presentation at the NAB show, and describes my testing with the Kenwood HD Radio.

INTRODUCTION: WOR-AM radio in NYC became an IBOC test station for iBiquity Digital Corporation in October 2002. WOR-HD, as we call our IBOC operation, also became New York’s first digital AM radio station.

Until December of 2003, the only listening that could be done to WOR’s IBOC signal was on test radios provided by iBiquity, one at the transmitter facility in Lyndhurst, New Jersey, the other in the Master Control Room of WOR on the 23^rd floor of 1440 Broadway in NYC. Taking either of these radios for a test drive was out of the question, as they both required AC power, contained hard disk drives, and would also require an analog-to-digital converter along with outboard amplifier and speakers.

The WOR digital coverage was determined by iBiquity using its test vans. We needed to take its word as to what was our digital coverage, as we had no way to measure it.

AND THEN ALONG COMES A PRODUCTION MODEL RADIO: In December 2003, WOR received an actual production model IBOC-capable radio from Kenwood. The radio consisted of the Kenwood KDC-722 head unit, with the Kenwood KTC-HR100 HD Radio module.

This Kenwood radio was installed into an average 2003 model year Ford Explorer, utilizing an installation kit from Crutchfield Electronics.

The HD Radio module was installed in the glove box of the Explorer utilizing hook and loop fasteners. While there was plenty of room behind the glove box inside the dash board to install this module, it was placed in the glove box to offer ease of access to the RCA output jacks located on this unit for audio recording purposes.

Care was taken to make sure the unit was properly grounded to the frame of the vehicle to minimize any ignition or electrical noise entering the radio and HD module.

While the factory antenna cable was capable of reaching the HD module without a problem, it was decided to install the HD module using the included antenna extension cable. This cable was approximately 12–15 feet in length, and would most likely be used in any consumer installation. We wanted this installation to be most like one the average consumer would undergo at his local auto stereo shop.

LET THE TESTING BEGIN: Testing was to be done along a radial used by iBiquity during their coverage tests. The route chosen was directly out I-78 from Newark, New Jersey. This Interstate runs in almost a straight line at a bearing of 258 degrees true from the WOR transmitter facility in Lyndhurst, New Jersey.

Before beginning the test drive, it was decided to place a magnetic mount whip antenna on top of the vehicle connected to the external antenna input of a Potomac FIM-41 Field Intensity Meter. The WOR monitor points were used to calibrate the antenna with a second FIM-41. Once the FIM connected to the magnetic mount was grounded to the vehicle chassis to provide a ground point, the signal proved stable and remained calibrated for the purposes of this testing at each of the four WOR monitoring point locations.

Leaving the WOR transmitter site showed that the radio’s stereo light was on, meaning that the radio was locked on the enhanced IBOC carrier. The route to I-78 went East on Route 3, South on I-95, then heading West on I-78.

The majority of the ride out I–78 found the Kenwood radio locked onto the enhanced mode. There was no blending to analog mode under bridges, though there was a quick blend to analog at one point where I–78 passes through a “concrete canyon,” two high concrete walls located on each side of the Interstate. The analog signal at this point was noisy, so it made sense that the radio would blend to analog. The ride was uneventful up to the Pennsylvania border.

Coming up to the toll plaza, the signal would occasionally blend to analog. Pulling over to the side of the road immediately after the toll plaza showed a signal level of 1.4 mV/m at 40 degrees, 35 minutes, 46.6 seconds North latitude and 75 degrees, 20 minutes, 5 seconds West longitude as measured on a GPS, a distance of 66.4 miles from the center of the WOR antenna system at an angle of 258.7 degrees true. This location is just outside Bethlehem, Pennsylvania.

Driving further West showed that, after the toll plaza, the radio had dropped out of enhanced mode, but stayed locked on the digital carrier in core mode. This meant the stereo light was off, but the radio was still receiving a mono digital signal.

At the point where the stereo light no longer flickered on and the radio made its first blend to analog, the vehicle was pulled over and the signal level measured. The signal level was 0.96 mV/m at 40 degrees, 34 minutes, 3.5 seconds North latitude and 75 degrees, 24 minutes, 35 seconds West longitude, a distance of 70.7 miles at a bearing of 257.8 degrees true from the WOR transmitter facility. This location was right outside Allentown, Pennsylvania.

Continuing further West showed that the radio blended to analog permanently at a signal level of 0.75 mV/m at 40 degrees, 33 minutes, 59.9 seconds North latitude and 75 degrees, 29 minutes, 13.7 seconds West longitude, a distance of 74.6 miles at a bearing of 258.4 degrees true from the WOR transmitter.

It appears that, along this radial, WOR’s digital coverage takes us to Allentown, Pennsylvania. This is far outside the New York metropolitan area and is perfectly adequate for WOR’s needs.

It should be noted that WOR’s directional antenna forms a lobe aimed directly toward Philadelphia from Lyndhurst. The radial out to Allentown skirts this lobe to the north, and lies between the center of the lobe and WOR’s deepest null.

Comparing this coverage to a map provided by iBiquity, which unfortunately will not reproduce properly for this paper, shows that iBiquity’s tests list digital coverage along this radial out to Allentown. It appears that the Kenwood radio works to the iBiquity system design and is providing digital coverage to areas where iBiquity says their system should provide the coverage.

Why does the radio blend to analog at 0.75 mV/m rather than 0.5 mV/m? There could be numerous reasons. Among them: it could be the fact that the Explorer’s antenna is 32 inches long instead of a meter in length, it could be the excess antenna cable connected to the HD module, and it could be component tolerances in the front end of the HD module. It could be electrical noise generated by the Explorer. The bottom line, though, is that the coverage experienced with the Kenwood radio is exactly what iBiquity found with their test vans.

FURTHER OBSERVATIONS: The route the author drives into the City each day runs directly near the center of WOR’s null. iBiquity predicted that digital coverage would end at the New York/New Jersey border at the intersection of Route 17, I-287, and I-87. This is where our digital coverage ends with the Kenwood radio.

After this location driving South, there are a couple of quick blends to analog, as the WOR signal is shaky in this area. Once through Mahwah, New Jersey, however, the digital coverage is solid down Route 17, over to Route 3, and to the Lincoln Tunnel. It should also be noted that the pattern bandwidth of the WOR signal in the null is not very good. North of Paramus, New Jersey, the WOR upper sideband is greatly diminished. Yet WOR has digital coverage to the predicted location.

Further observations show that, driving cross-town from the outlet of the Lincoln Tunnel down 40th Street to Broadway, the digital coverage in the City is excellent. There are places along this route where the analog signal is very noisy. If a taxi pulls up along side the vehicle, the Radio Frequency Interference (RFI) generated by its taximeter generates a whine in the analog signal.

The Kenwood radio, however, stays locked on the digital carrier. In trouble spots for the analog signal, the digital signal goes to mono core, but it stays digital.

FM OBSERVATIONS: WNEW-FM is transmitting an IBOC carrier from the top of the Empire State Building. While Empire State Building may be the tallest location in the city of New York, reflections caused by the buildings blocking the signal along the West Side cause WNEW-FM’s signal to picketfence with multipath distortion all the way up the West Side Highway from 42nd Street to the George Washington Bridge.

WNEW-FM’s IBOC signal is free of this multipath distortion. One can drive up the West Side Highway, hit the button on the radio for WNEW-FM, and hear the analog signal spitting. The radio then blends to digital, and the multipath magically disappears.

While I have not measured signal levels on WNEW-FM, the Kenwood radio appears to hold the digital signal almost to the point where the WNEW-FM analog signal starts having problems.

Overall, from AM measurements to simple FM listening, I would say that this first commercial IBOC radio on the market performs as intended. I don’t think the public will be disappointed with the performance of this radio. Kenwood says that they have learned a great deal about the IBOC signal while designing this radio. The second generation of IBOC radios can only be better.

A.7  Installation of IBOC on AM

The following article was written for Radio World Magazine in late 2002.

While we’ve all heard and seen things regarding IBOC digital transmission for AM and FM stations, there hasn’t been much written in regards to what it will take to install IBOC on an existing installation. WOR, a 50-kW flame thrower in NYC, and one of America’s pioneer broadcasters, became New York’s first digital AM radio station at 9 AM, October 11, 2002.

First and foremost, before you delve into the world of IBOC, you must first evaluate your transmitter facility. You may have been seeing articles regarding what you need to do in order to make your facility IBOC ready. Would it be nice to have all-digital studios and a completely digital STL to make this happen? Sure. Is this practical for most stations? No. Most of us need to live within a budget, and replacing the entire plant, microphone to antenna, makes no sense initially. Since this article focuses on the transmitter plant, I’ll briefly say that WOR has older studio equipment: our consoles are literally serial numbers 1–5 of PR&E’s System 1 console, the predecessor to the BMX, and are mono. We will be concentrating on our studios within the next two to three years, and made no changes at the studio with the exception of our monitoring situation, as IBOC introduced an 8.5-second delay into our analog audio.

IS YOUR TRANSMITTER PLANT READY?: WOR’s transmitter facility has seen much work over the past five years. It started with a complete rebuild of the phasor and coupling circuits for our three-tower dogleg array. But this is not just any three-tower dogleg.

WOR’s transmitter plant is located in Lyndhurst, New Jersey. It is an RF oasis. On our street, all within 1½ miles of each other are: WOR, WLIB, WJWR, and WINS. Counting WOR, there is approximately 300 kW of RF over our site. You shut WOR’s transmitter off, ground a tower, and draw a considerable arc. We have almost 3V/m at 1010 (WINS) alone. Needless to say, our antenna system consists of detuning skirts on the towers to electrically shorten them at 1010 and 1190; detuning networks at the bases of the towers for 620, 1010, and 1190; traps for 620, 1010, and 1190; and then we actually get to the components that WOR can use.

The redesign, performed by Cart T. Jones Corporation, took WOR’s common point which looked like a roller coaster on either side of 710, and made it fairly flat ±15 kHz from carrier. After that, it goes to hell in a hurry, but for IBOC operation, your antenna and/or common point should look fairly flat and the reactance flat or at least symmetrical over the passband of ±15 kHz from carrier. WOR passed the first major test.

You next need to look at your transmitter. According to information I received from Harris at their IBOC road show, a tube-type AM transmitter pretty much will not have the stability required for the phase modulation components of the IBOC signal. The RF chain needs to be fairly linear, and, in the case of a PDM transmitter, the sampling frequency needs to be high enough and the filtering broad enough to allow the IBOC components to properly pass through to the final.

WOR utilizes a Harris DX-50 transmitter for a main. It is basically plug and play for IBOC. Our older auxiliary transmitter, a Continental 317C-1, pretty much doesn’t have a prayer of passing IBOC, though both Chief Engineer Kerry Richards and myself think it might be interesting to try IBOC on the beast to see what would happen. Since WOR had a main transmitter which is more than capable of passing the IBOC signal, and the antenna system looked good, it was time to play.

INSTALLATION: Kerry and I did some preparation at the site before Pat Malley, a field engineer for iBiquity, arrived one fine Sunday afternoon for installation. To get the IBOC signal into the transmitter, there is an interface box that needs to be installed between the exciter and transmitter. The connections between the interface and the transmitter are a BNC for a coax connection to the external RF input and a Phoenix-type connector for the audio connection. As an option, there is a control voltage input that can be paralleled to a relay in the transmitter to change between internal and external excitation.

Connection between the exciter and interface is done with Cat-5. I can now say that I have pretty much seen it all since I have connected my first 50-kW AM rig using shielded Cat-5 cable. Before Pat arrived, Kerry and I had run the Cat-5 from the racks to the transmitter, and made sure we had jacks available on the patch bay for input to the A/D converter. Time to do this was about one hour.

Pat arrived bearing gifts, and we put the auxiliary on the air. We proceeded to connect the Cat-5. Kerry worked on getting the connections between the patchbay, A/D converter, exciter, and processing correct. I installed the bypass relay in the DX-50, changed the audio input, and installed the BNC cable for external excitation.

We brought the DX up into the dummy load at 5 kW. We made sure that when the exciter was in the bypass mode, we were, in fact, operating on the internal oscillator and when not in bypass, we were operating on the external excitation from the exciter.

We then needed to set up the internal AM reference in the exciter. This is done by putting our Optimod 9200 into test-tone mode and watching the reference signal from the exciter on a scope. We set the exciter for 100 percent negative modulation, then fired up the DX, IBOC carriers off, and adjusted the gain on the interface box to produce a 100 percent modulated signal into the dummy. So far, so good.

Next, we turned on the IBOC carriers with the transmitter set for 5 kW. The display on the spectrum analyzer looked good, but on the receiver, the audio sounded awful in digital and analog. It was time to troubleshoot. Turns out the analog to digital (A/D) converter between the STL switcher and processing was bad. A quick replacement of the A/D resulted in clean audio.

There are several adjustments to make on the IBOC exciter. One major adjustment sets the phasing delay (from AM stereo, remember setting the group delay through the transmitter?) of the IBOC waveform as it passes through the transmitter. This is done by making an adjustment on the touch screen of the IBOC exciter while monitoring data errors on the receiver, or by watching the RF signal on a spectrum analyzer to minimize “spectral regrowth” around ±20 kHz from carrier. I’m happy to say that errors on the WOR signal are zero and it is extremely stable.

The next adjustment was to adjust the delay of the analog signal to precisely match the delay of the digital signal. The reason for this adjustment and the delay added to the analog is that the radios are designed to lock on a station in analog mode, then, when the digital signal is acquired, blend or cross-fade into the digital signal. Obviously, if the two signals are not time aligned, the listener will hear an abrupt switch between them. Additionally, the radios are designed so that, if the digital signal is lost, the radio will immediately revert back to the analog signal. Once again, if the signals are not time aligned, the change will be abrupt.

The easiest way to do this? We set the receiver to produce the digital signal on the left, the analog on the right, and proceeded to match them by ear! The result is that there is no abrupt change when the receiver blends.

We then brought the DX up to 50 kW (all the above tests were done at a lower power level such that the dummy would not get grumpy and take up smoking). Analog sounded really good. Digital, being that the processing is very light (as opposed to our analog which is very “in-your-face”), sounded close to FM quality. Then … GLITCH. The DX-50 reduced its power to 25 kW, and ramped back up to 50 kW We thought the problem might be that the dummy, which admittedly needs some work, was shifting impedance, so it was time to put the DX-50 back on the air.

ON-AIR TESTING: We switched back to the DX-50. It sounded really good in analog. We turned on the IBOC carriers. It sounded wonderful in digital. Then … GLITCH. The DX abruptly dropped to 25 kW, then ramped back up to 50 kW. It did this several times. Pat said he had seen this before.

A call to Harris showed that on the bandpass filter VSWR monitoring, the voltage and current sample points are taken in different locations along the circuit. At some point during operation, the zero crossing points of the voltage and current due to the phase modulation of the transmitter must occur at the same moment in time. The VSWR circuit takes this as a problem in the bandpass filter, and reduces the transmitter’s power for protection.

As a band-aid, per Harris, we have bypassed the bandpass filter VSWR circuit. Harris is working on a field modification that we will need to perform on our DX-50 in the near future.

Total time spent? About three hours. And IBOC is on the air.

RESULTS: Of course, Kerry and I rushed to our cars to see what the outcome was on the air. My car has a stock Ford AM stereo radio. Kerry has a high-end radio with a $2 AM section. Before iBiquity arrived for installation, we had reset our Optimod 9200 from its National Radio Systems Committe (NRSC) settings to a 5-kHz brick wall roll-off. We based the 5 k setting on our NRSC processing settings, and added our own equalization (EQ) curve. Listening on our car radios, we could barely tell a difference in the processing changes on the AM stereo radio, and could hear no difference at all on the high-end radio. And, since we’re not modulating above 5 kHz, we are now louder on these radios than we were.

Once IBOC was on, we listened for artifacts. On the supposed “wideband” AM stereo radio, I need to turn the volume up to ear-splitting levels before I can hear any noise under the audio. On Kerry’s radio, you cannot hear any artifacts of IBOC whatsoever.

WOR’s listeners are extremely loyal and are a vocal bunch. If they hear something wrong, they do not hesitate to speak up and make it known that “their” radio station has a problem and they want it fixed NOW. Our calls from listeners have consisted mostly of people wanting to know how they can hear our digital signal and where they can buy the radios.

We have had a few negative comments. One was from a gentleman who was restoring a 1930s vintage Atwater Kent radio, and wanted to let us know he heard hiss when he tuned across WOR on either side of us. The other negative comments have come from a group of AM stereo fanatics in New Jersey. These people live for the day AM stereo makes a come back. They are not listening on typical AM radios. They have verbally personally attacked both Kerry and myself, the radio station, and iBiquity. Keep in mind that this group thinks that AM radio is a high fidelity medium. They also started a rumor that WOR was operating illegally. The NRSC mask allows emissions to –25 dBc from 10 to 20 kHz. IBOC operation puts the IBOC carriers from 5 to 15 kHz at –30 dBc, perfectly legal.

The other complaint was from a person who was trying to get WLW. They live not all that far from our transmitter. Unfortunately, the IBOC carriers occupy space in the NRSC mask around 700, and they were not able to DX in the near field of the WOR antenna. But since WOR is legal, there is not much that can be done for this person.

The spectrum on the WOR signal was basically textbook perfect. The entire signal fits very nicely under the NRSC mask and is completely legal. And, even with all the detuning aspects of WOR’s antenna, the IBOC carriers are symmetrical.

And IBOC was installed without rebuilding our studios or replacing our STLs. The time spent for installation was about three hours, and we did it in the afternoon with the auxiliary transmitter on the air. If your antenna system is in reasonable shape, and your transmitter is of fairly recent vintage (i.e., not a 1955 BC-5P), you should lose very little time and spend the minimum amount of money putting IBOC on your AM station.

For more information, I would recommend iBiquity’s Web site at www.ibiquity.com. Or call your Harris representative. They can provide you with details from their IBOC road show, and can even guide you as to whether your IBOC installation will be as uneventful and easy as was ours.

A.8  Radio world guest commentary

I have written numerous guest commentaries in Radio World Magazine over the years. Many of them have to do with incorrect and false information being spread about HD Radio operation. Here is one where I have responded to a gentleman who has done just that.

I just finished reading the letters in the “Readers Forum” in the March 10th edition of Radio World, and the letter entitled “Slingin’ Hash” from Jim Jenkins stands out as having erroneous information regarding IBOC. While the industry has been ramping up for IBOC deployment, there has not been very much information available regarding IBOC deployment, and I believe many of the statements made by Jenkins and others have been brought about because of ignorance on the subject, and not necessarily through any fault of their own. The information is simply not readily available. I wish to address the four primary concerns brought up by Jenkins.

1.  IBOC will make all radios in the United States obsolete – When is this going to happen? There is no mandate to shut off the analog carriers, and your analog radio, AM or FM, works just fine with IBOC. This isn’t HDTV we’re talking about, where the analog carriers have a definite cut-off date set by the FCC. Your radio won’t be obsolete the day your favorite station turns on an IBOC carrier. I listened to WOR on the analog radio in my car for 14 months before I had an IBOC-capable radio to listen to, and this analog radio was far from obsolete. Anyone who tells you that your analog radio will be obsolete has no clue what they are talking about and does not have the facts. Eventually, though, the present radios will become obsolete when and if a mandate to turn off the analog modulation comes about. I do not see this day coming any time in the near future. We are most likely 20 years out for this to happen.

2.  The cost to small town radio – Yes, there are costs involved to convert to IBOC. Once again, there is, at this time, no mandate or timetable to do so. We’re constantly reading in the trade publications, Radio World included, about the loss of listenership in Radio. Much of this has to do with the programming, which is beyond the scope of this commentary. But when you have listeners leaving for a service such as XM or Sirius, and they then discover that your station offers programming locally similar to what they like on these services, very frankly, your 1955 vintage transmitter won’t cut it with them. Small town radio, much like small town television has had to do with HDTV, will need to eventually convert to IBOC, and the time to start planning and budgeting is now, not when and if the FCC mandates digital service. I would start planning to do this at some time down the road relatively soon and not ignore the fact that digital service is coming to AM and FM radio. If you put your head in the sand and pull it out to be confronted with a mandate to do it, you will be in trouble. Planning now will help soften the blow.

3.  The listeners, particularly in poorer areas, won’t be buying the new radios for years – And we should stop progress for the stations that are not in poor areas? How long did it take for FM and FM stereo to catch on? How long did it take for color television to catch on? It is going to be the same for IBOC implementation. It will take a while. No one has said that it will not. If we don’t move forward, you may find that all stations, particularly those in poorer areas, will be in big trouble in 10–15 years. Imagine radio listenership continuing on a downward spiral so that the poorer stations sign off first, as they will be the first to struggle and go belly up. Will IBOC be the savior of radio? No. But with digital services being in demand by consumers, it will help. We can’t ignore this.

4.  No one in the “trades” has addressed the adjacent-channel, sideband, frequency response curve and modulation limitation questions – I’m frequently a contributor in this particular trade. Ready for these issues to be addressed?

There are no modulation limitations. WOR is still banging away at –97 percent, +122 percent with the IBOC carriers on, with very aggressive processing, the same as we were pre-IBOC. This was a specification we put onto iBiquity when WOR became a test station in 2002. We do not lose analog loudness when the IBOC carriers come on at sunrise, and we do not gain analog loudness when the IBOC carriers go off at sunset.

We are still louder and prouder than our immediate competitors and make no apologies for that. IBOC did not hurt WOR’s analog modulation. We have checked with oscilloscopes and spectrum analyzers.

We have discovered, however, a few radios that seem to have a broader bandwidth in the sampling for the automatic gain control (AGC) circuit than the frequency response the radio can reproduce. On these few radios, WOR’s analog audio appears to reduce slightly when the AGC circuit senses the additional power in the sidebands when the IBOC carriers are turned on. These radios are few and far between and, very frankly, I see this as a design flaw with the radio. Here in the New York metropolitan area, there are a few AM stations that are awfully close together frequency-wise. I suspect these radios have the same issue with these close-spaced stations. Should we abandon IBOC because there are a few radios in the market with an inherently flawed design? Should we abandon the space program because there have been a few space shuttle accidents?

WOR has worked extremely closely with iBiquity’s engineers to make sure that not only is the IBOC portion of the signal the best it can be, but that the analog portion of the signal is not impacted. I recall one software change where a mistake was made in the analog side of the software, the result being no real positive peak capability above + 100 percent, and a distinct audio distortion. We identified the problems for iBiquity, and gave them one week to solve the issue before we would pull the IBOC carrier. They listened, resolved the issue, and then had a problem with the analog audio filtering. This resulted, two days later, in the “WOR Patch,” which has since been incorporated into the final version of the software. We would not let the analog suffer, and iBiquity came through. The result is that there is no impact on analog performance, while the new HDC codec has solved the artifact issues with the PAC codec that we started with.

What needs to be addressed with the frequency response curve? On the analog carrier, you are limited to 5 kHz. In actuality, WOR is limiting to 6 kHz. The majority of AM radios I have encountered generally do not reproduce anything above 3–4 kHz. The stock radio that was in my Ford Explorer reproduced 6 kHz, but only when I was sitting in front of the transmitter building. So, what is the issue? Extending your analog frequency response beyond 6 kHz (again, discovered in WOR testing) will severely degrade your digital coverage area and digital performance. And we are running a modified NRSC equalization curve in the analog processing. Operating with a 6-kHz audio bandwidth proved to have a better sound even on typical narrow-band AM radios and did not affect the IBOC coverage of WOR, so we see no reason to limit our frequency response to 5 kHz. Since the greater portion of AM radios cannot “hear” anything above 4 kHz, I ask again, what is the issue?

Another little known fact is that there are two general operating modes for hybrid AM IBOC. The normal mode, which limits the analog frequency response to 5 kHz, allows the HD Radio to constantly look at the sidebands and decide which one to decode the data from, upper or lower. In testing with iBiquity, we discovered that there are distinct differences in how the sidebands will be recovered at any given receiving location dependent on many factors.

The second mode of IBOC operation allows for an analog bandwidth of 8 kHz. In this mode, however, the radio must decode both sidebands at all times to recover the data and generate HD audio. Obviously, if there is a problem with one of the sidebands, or if one of the sidebands is severely diminished (in the WOR null, far field, our upper sideband all but disappears), the radio cannot reproduce an HD signal. This will have the effect of reducing your HD coverage, and the recovered data will not be as robust. The choice of operation is up to the individual station, but if the majority of radios do not reproduce analog audio above about 4 kHz, what is the point of limiting your HD coverage?

ADJACENT CHANNEL AND SIDEBAND QUESTIONS – Yes, there may be some issues due to the IBOC sidebands. Once again, no one has ever said that there might not be. However, you first need to consider that the IBOC system is designed to fit under the NRSC mask as mandated by the FCC. Energy that fits under the mask is completely permissible under FCC regulations, regardless of what anyone wishes to read into this rule.

One should refer to the FCC regulations, Part 73.37, to get the definition of the coverage area for a particular class of AM station. Yes, contrary to what some people believe, there are coverage definitions in the FCC rules for AM stations. There may be some background hiss heard out around and beyond the 0.5-mV/m contour of a station that has an IBOC neighbor. In listening tests, this was deemed more acceptable than the “Donald Duck Talk” from present adjacent sideband splatter. And let’s face it. In most areas of the country, if you can even hear an AM station out to the 0.5-mV/m contour because of the general electrically noisy environments we live in, you’re a lucky dog. If you can’t hear the station reliably in these areas anyway, there should be no reason to prevent implementation of IBOC.

If you think your typical listeners are listening to your station when the signal degrades to noise out to and beyond the 0.5-mV/m contour, perhaps you should do a study or survey of the average American’s radio-listening habits. I observe my 16-year-old son, his friends, and my neighbors. My son knows what AM radio is (he’d better, considering what his old man does for a living!), yet he and his friends refuse to listen to AM radio. They simply do not care for the sound of it. And this has nothing to do with programming. One of his friends is a die-hard New York Yankees fan. The only time I heard him listening to a ball game on the radio was when the playoffs were carried by an FM station out of Newburgh, New York, near where we live. I have observed neighbors punching the button at the first sign of noise on an AM or an FM station. They don’t like it, won’t put up with it, and won’t listen to a station regardless of the programming, once the signal gets noisy.

At this time, I won’t get into the issue of the fact that we have essentially lost an entire generation of listeners to AM radio. But for those who believe that every listener is an avid DXer, I firmly believe you are kidding yourself. Stay the course and not change to accommodate the younger listeners who have grown up on Internet audio and MP3 players, and you will drive even more nails into the coffin of AM Radio and radio in general.

Jenkins states that we have a mindless push to a flawed technology. Show me any technology that is inherently perfect. There are none. Jenkins further states that we are going digital because it’s “different.” I have made countless recordings (on linear Digital Audio Tape (DAT)) off my IBOC car radio, and play them to WOR staff members (not just engineers), clients, and listeners who happen to be in the station … on studio monitor speakers. I have given sales staff and neighbors CDs and asked them to listen to them in the car and tell me what they find objectional. Universally, all the people who have participated in these tests find it hard to believe that these recordings were made off of an AM radio (granted, IBOC capable), and ask when it will be available on more stations. They also ask where they can get one of these HD Radios. The AM HD sounds very good. I’ve been listening to it critically in the car since December. I can’t imagine going back to the standard AM on my drive in the morning. The WOR Morning Show sounds so much better in the HD environment.

And while Jenkins makes the point that independent record companies like to press vinyl, walk into any music store or the music department in any K-Mart or Wal-Mart. Do a ratio of CDs to vinyl. What do you find? The CD has replaced vinyl as the universal recording and distribution medium. People download MP3 files all the time and find them more than acceptable to listen to. I don’t buy the argument that because “old-fashioned records” give a warmer feel that they must be superior. Each medium has its pros and cons. Vinyl records are susceptible to surface noise, turntable rumble, typically have only 30-dB stereo separation, and generally have a maximum 50-dB signal-to-noise ratio. In this argument, the CDs have won.

Regarding the cascading of algorithms, yes, with IBOC, we’re all going to have to be more careful regarding data reduction use. At WOR, we record the Bill O’Reilly program for later playback, and do so in what can be considered a worst case scenario. We pull the program down at our transmitter on Starguide MP2, send it to the studios on an MP3 channel on a T1, record it into our ENCO system MP2, play it back and send it to the transmitter via a linear digital channel on a T1, and then put it through the HDC codec. Listening in the car, you don’t hear any “swimming” or raunchy data cascade effects. The average listener won’t notice. If, however, you put the air product up against a studio recording of the O’Reilly program, you will definitely hear the result of the algorithm cascading.

Yes, we did this test with Westwood One, recording identical segments at every point in the path where the audio coding would change. Studio receives off the phone line from the Fox studios at the Westwood One head end here in Manhattan at the output of a Starguide receiver, at the output of our MP3 card on the T1, at the output of the ENCO playback, and off of an HD-capable radio. It’s amazing to hear the changes along each leg of the path.

The overall result, however, is that no one will notice, unless he or she either knows what the studio product sounds like or is an audiophile listener. Don’t forget – audio memory is extremely short. Unless you know what you are listening for, chances are you wont notice, unless the result of the cascading is so raunchy that you can’t help but notice.

Now, before I am once again accused (and it was here in letters to Radio World) of being anti–small market radio and anti-AM in general, consider this. Buckley Broadcasting is one of the true small operators left in the United States. WOR is an anomaly in our company, as most of our stations are small stations in much smaller markets. We don’t have the deep pockets of investors to pick as do the Clear Channels and Infinity’s of the world. When we make a technology decision, it is something the company must believe in, and it is my job to bring the pros and cons of any new technology facing the company to the president. We are presently in the process of finding out what it is going to cost to outfit our 19 radio stations for IBOC operation. While this discovery, budgeting, and planning are taking place, I have discovered that iBiquity has reduced licensing rates at the date of this writing. While Buckley Broadcasting is not yet ready to implement IBOC operation on our other stations at this particular moment in time, I have recommended that the Corporation license all of our stations with iBiquity now at the reduced rates. It would have been irresponsible of me not to make this recommendation. I grew up on AM radio but I am a realist. If we continue on the “keep it the same because, after all, we’re Radio!” path, we will all be in trouble.

IBOC isn’t the three-headed monster, the fear mongers among us think it is. It’s time for a change in our industry. Education is the key to understanding and using this new technology to our greatest advantage. Remaining the same while the world marches past us will definitely place terrestrial AM and FM broadcasting among the dinosaurs and make us irrelevant.

A.9  Conclusions

I thank you for the purchase of this book. I hope that it has helped you gain the knowledge you need to add HD Radio operation to your station. I also hope that my observations in this appendix have helped you through my experiences with HD Radio.

As I stated in the opening section, broadcasting as we know it is in a constant change of flux. This is certainly an exciting time to be in the business. The digital future is at hand – and it is in your and my hands. While you may not fully agree with HD Radio, I hope you agree that it is our job to make it work the best it can. I hope you have a very successful HD Radio install!