Or rather occupied bandwidth. During a recent Alternative Inspection of an FM station, there was some question as to the accuracy of the modulation monitor. Truth be told, a modulation monitor is no longer required at a radio station, so long as the station ensures that they comply with relevant FCC regulations for their service. Many modulation monitors continue on, however, as air monitor receivers.
That is all well and good, however, many modulation monitors are notoriously inaccurate and tend to the overly sensitive side of the equation. If used when setting the modulation levels, this can lead to under modulation, which, as we all know leads to disaster, destruction and bad ratings…. Because the volume knob on every radio in the entire metro, Total Survey Area (TSA), or even the whole country has been broken off and listeners are unable to compensate for the low audio levels from an under modulated FM transmitter.
But anyway.
FCC 73.1560 gives the maximum FM deviation as +/- 75 KHz from the carrier, with some allowance for SCA injection levels, up to +/- 78 KHz. This is the definition of 100 percent modulation of an FM carrier. Thus the entire occupied bandwidth is 150 KHz, leaving a guard band of 50 KHz between signals. That is, unless IBOC is employed, then the guard band is -100 KHz which is good science no matter how one looks at it. On a spectrum analyzer, it looks something like this:
Occupied bandwidth of analog FM broadcast transmitter
This shows that the 5 second average occupied bandwidth of 90 percent of the transmitted energy is within 153 KHz, which is slightly high but within the margin of error of the measurement device. The vertical lines represent the -10 dB signal level as referenced to the carrier. Thus this station is in compliance with FCC rules regarding modulation in spite of the face that the analog modulation monitor shows it being 10-20 percent over. Had it actually been 110 percent, the occupied bandwidth would have been 165 KHz and 120 percent would have read 180 KHz.
Thus, before buying the latest squash-o-matic FM processor and setting it for full tilt boogy, a good engineer may want to check the occupied bandwidth with something other than the old analog FM modulation monitor in the rack.
There are a few FM stations around here that intentionally broadcast in mono. One is an FM talker, which from a technical standpoint makes a certain amount of sense since any particular human voice is a single point sound generator. There are some exceptions to that rule, such as those radio talk show hosts that talk out of both sides of their face, but that is an entirely different conversation.
The other FM station broadcasting in mono, WKZE, has a music format with a very eclectic play list. It is a full Class A located in north western Connecticut. The idea with this station is to garner a larger and more reliable coverage area.
It comes down to a simple physics discussion about free space loss. The basic equation for free space power loss is:
That formula works for a single frequency, say the carrier frequency, for example. As the signal gets spread out by modulation, the power density on any given frequency is reduced as the energy is divided between many other frequencies.
First, free space loss takes into account the spreading out of electromagnetic energy in free space is determined by the inverse square law, i.e.
where:
is the power per unit area or power spatial density (in watts per metre-squared) at distance ,
is the total power transmitted (in watts).
Second, with Frequency Modulation (FM), the power spectral density is a function of the differences in the highest and lowest frequency:
Therefore, the narrower the bandwidth of a signal, the higher the density of the received signal will be in relation to the transmitted power. An unmodulated FM signal will have a better, more reliable coverage area than a modulated one. Of course, we need to modulate the signal, otherwise there is no point in having the transmitter on.
A baseband or composite FM signal has several components:
FM baseband signal
An FM station transmitting a mono signal will have a much lower bandwidth. With wideband FM, the modulation index is generally 2fΔ or two times the maximum audio input frequency. Thus, a mono FM station will have an approximate deviation of 20 kHz (plus any ancillary services like RDS) vs a stereo FM station, which has a 75 kHz deviation using the same carrier power.
For higher power FM stations, FCC Class C and B, this is not much of an issue. Those stations tend to have a great deal headroom when it comes to power density, building penetration, multipath (picket fencing and capture effect). For Class A and LPFM stations, it is a different situation. For those stations, unless FM stereo broadcasting is truly needed, it should be turned off. On low power stations, stereo can be a great detriment to reliable coverage.
I love the sound of these units when coupled with an Optomod 8100A. Many people have (or rather, had) difficulty setting these things up. I found them to be very easy to deal with, just follow the instruction manual. If that doesn’t sound good, then there is something wrong with the unit. Over the years, there are only a few consistent problems. The first thing is with the voltage regulators. They have heat sinks attached with nylon screws. The screws get brittle and fall apart, making the regulator overheat and go bad. I have taken to replacing the nylon screws, and if the heat sink has fallen off, the entire regulator. There are also a few electrolytic capacitors in the power supply and on the audio board, it is always a good practice to replace those. Otherwise, unless the unit has been blown up by lightning, it should work.
As for set up, follow the directions in the manual:
Bypass the units using bypass switch
Turn on onboard pink noise generator
Using the test ports on the front of the unit, plug a Simpson 260 VOM set on 2.5 VAC important: use the ground port on the front of unit, not the case
For use with an Optomod 8100A, using the dB scale on the Simpson 260, set all the bands for a 4.0 reading. Set the density to 3/4.
Turn off pink noise generator and switch out of bypass mode.
Make sure the levels in the studio are where they should be.
Adjust the input gain so the “Buffer Active” light does not come on during normal level programming.
Adjust the output levels so that the input buffer on the Optomod reads between -7 and -3 vu.
The rest of the settings are on the Optomod:
Clipping = 0
HF limiting = 5
Release time = 2
Bass coupling = 2
Gate = 0
Set the input attenuators for about 10 dB total gain reduction, with peaks around 15 dB or so.
Then set the L-R null. To do this, make sure the program material is in mono, then adjust the L or R input attenuator for minimum reading. Also, if the Audio Prism has PR-1 (phase rotators) installed, bypass the phase rotator in the Optomod. There is also a replacement card 5 made by Gentner called the RFC-1 for the Optomod 8100A. I notice little difference between a stock Optomod and on RFC-1 Optomod.
That is a good starting point. Most people are quite happy with this, but if needed, the high and low settings on the Prism can be adjusted slightly to suit the station equipment. When properly adjusted, this equipment rides gain, adds a certain amount of loudness, while keeping the programming material natural sounding. Further, unlike some “modern” air chain processors, it does not boot up and it does not occasionally loose its mind, requiring a reboot.
The best paragraph in the manual, or any broadcast equipment manual is this:
There is a wealth of information available in the LED display. A few minutes of watching them in reduced light (emphasis added) while listening to a familiar program input will greatly help in understanding their action.
It will also greatly enhance your buzz, dude. It was the 70′s.
I am a strong proponent of non-computer based air chain processors. Something about listening to dead air while the computer reboots is annoying and every computer needs to be rebooted every now and again.
All of that being said, I recently had a chance to play around with Breakaway Broadcast audio processing software. I have to say, as a low cost, very versatile platform, it can not be beat. I would put it up against any of the high end FM audio processing, provided one uses a high quality sound card with an adequate sample rate.
Claesson Edwards Audio has developed several software based audio processors for a variety of end uses. They make several recommendations for hardware and operating systems, Pentium 4 3.2 GHz or better, dual core preferred. If one is interested in used the sound card to generate composite audio, then any sound card capable of true 192 KHz sample rate will work. They list several that have been successfully tested on their web site.
For approximately $1,200 dollars or so, one could buy a decent computer, the Breakaway Broadcast software and the Airomate RDS generator software. For a Mom and Pop, LP or community radio station that is looking to do some high end audio processing and or RDS, that is a good deal. I would add a UPS to the computer and keep back up copies of the software installed on an emergency computer just in case. One can never be too safe when it comes to computers, viruses, hackers and other malicious persons.
Things that I like
Inexpensive, the fully licensed version is $200.00. The demo version is free but there is a 30 second promo every thirty minutes.
There are several factory presets, but everything is fully configurable, changes can be named and saved allowing some experimentation.
Audio cards with 192 KHz sample rate or greater can be used to generate composite audio, eliminating the need for a separate stereo generator
RDS capable with additional software (Airomate2, approximate cost $35.00)
The same processing computer can be used for streaming audio and or AM audio processing simultaneously.
Full set of audio calibration tools for AM and FM transmitters, allows correction for tilt, overshoot and linerity. Can add pre-emphasis at any user selectable rate.
Fully adjustable phase rotators.
Things that I don’t generally like:
Computer based system using Windoze operating system
WXPK in White Plains, NY has been using this software to process their streaming audio for about 2 years now. The software itself is extremely stable running on a stand alone Windows box with XP service pack 2.
Okay, that’s a little better. I was just reading up on the newest, greatest, holy cow, gee whiz, gotta have that expensive box processor, also known as the Omnia 11. I have to hand it to Mr. Frank Foti and his marketing team. They have created one heck of a buzz about this thing, and it seems like folks are jumping on board to shell out $10 – $12 K for the box. But let us review a few things.
I will admit most freely that I tend to be an audio purest. I do believe that a limited amount of processing has its merits, especially for those listeners in high noise environments like automobiles, work sites, etc. With sloppy DJ’s working the consoles, there is a minor need for some limiting, gain reduction and so on, just to the air product levels aren’t all over the place. Those are the real world considerations.
Does and Ipod have an air chain processor? No, if the Ipod user want more loudness, they turn up the volume. Since most Ipod users are normal people and not some burned out DJ with bad hearing, the volume control on an Ipod has plenty of head room to satisfiy. Does a Droid or a Blackberry or whatever else people are listening to these days have an air chain processor? No. And most users/listeners of those devices are perfectly happy with the quality and quantity of audio.
Back in the day when loudness meant a bigger transmitter, more carrier power, bigger signal, was easier to tune manually with the non-digital dial readout, etc., perhaps a loudness war with the cross town rival was part of the game. Nowadays, nobody cares except the program directors. I repeat, NOBODY CARES. Ask anybody on the street what the loudest radio station is. They very likely won’t even understand what you are trying to ask and you likely could not explain it in terms that would make them understand, much less care about.
The average person doesn’t give a rat’s ass about loudness. Nor do they really care about how deep and full the DJ’s voice is, or how well the noise gate works, or the six band EQ or any of that crap. In fact, if the music sounded just like it does on the Ipod, e.g. completely unprocessed, they probably wouldn’t even notice. The competition has changed and radio is being left behind because many people are stuck with old ideas about how things used to be. Times have changed, what should be the driving force in radio, the listeners, want to hear the music that they like. That is what the program director should be worried about, finding and playing good music that the listeners want to hear. Or having the best talk show, the most interesting news, or whatever other programming the station carries.
If the programming content is good, compelling radio, they will listen. Never mind the air chain processor, the mic processor, the limiter, how loud the station is, what power the transmitter is running at, etc. That is for the Engineers to take care of.
One thing that I find a little annoying is the continuing need to reboot everything at some interval. Computers in the studio, audio vault servers and work stations, e-mail servers, files servers, network routers, and so on. Got a problem, first thing to do is cycle the power off and on…
One of the most irritating pieces of equipment is the audio processors on one of our FM stations. A few years ago, we purchased the whiz bang Omnia 6 processor. Every 6 or 8 months the thing losses its mind and sounds terrible. The station gets all bassy and the high end sounds distorted. I have tried everything I can think of to prevent this, including installing an UPS, extra grounding, extra shielding, software updates, etc. In the end, it just has to be rebooted, which of course, means several seconds of dead air. Naturally, this processor is at the FM transmitter site, where it is difficult to get to.
Truth be told, when it is working, it does sound pretty good on the air, but is it $10,000 dollars better than the older Optimod 8100A? No, it is not.
The old Orban Optimods sound pretty good as long as they are re-capped and aligned every so often. If fact, our number one billing station has an AC format and uses an Optimod 8100A and nothing else. Our other station in the same market uses an Optimod 8100A and a pair of Texar Audio Prisms. In the ten years I have been working for this group of radio stations, I have never had to reboot the Optimod or the Audio Prisms, they just seem to work continuously without problems. Imagine that.
I have seen this called a “retro audio chain” by some. Nothing retro about it, a little care and feeding and I’d stack this equipment up against an Omnia 6 any day of the week and twice on Sunday.
This is a grainy video of an 8100A in action:
That was taken in our rack room using off air audio on the rack room speakers and a cheap video camera. You get the idea.
So here is to Frank Foti and his marketing gurus that have sold all of the program directors in America on the need to “update” there air chain processors, because, you know, the Optimod, that is old skool.
Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or the right of the people peaceably to assemble, and to petition the Government for a redress of grievances.
~1st amendment to the United States Constitution
Any society that would give up a little liberty to gain a little security will deserve neither and lose both.
~Benjamin Franklin
...radio was discovered, and not invented, and that these frequencies and principles were always in existence long before man was aware of them. Therefore, no one owns them. They are there as free as sunlight, which is a higher frequency form of the same energy.
~Alan Weiner
Everyone has the right to freedom of opinion and expression; this right includes the freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers
~Universal Declaration Of Human Rights, Article 19
is the signal wavelength (in metres),
is the signal frequency (in 
is the distance from the transmitter (in metres),
is the 
is the power per unit area or power spatial density (in 
is the total power transmitted (in watts).
I am a strong proponent of non-computer based air chain processors. Something about listening to dead air while the computer reboots is annoying and every computer needs to be rebooted every now and again.
Open Mic