Broadcast Notes by Ray Voss

Size: px

Start display at page:

Download "Broadcast Notes by Ray Voss"

Rosamund Gordon
6 years ago
Views:

1 Broadcast Notes by Ray Voss The following is an incomplete treatment and in many ways a gross oversimplification of the subject! Nonetheless, it gives a glimpse of the issues and compromises involved in the analog audio broadcast system. (2) As to the "musical scale" and required frequency response for good "audio quality", there's been a lot written over the years and as we discussed, the 15 khz - 20 khz high end figure for "hi-fi" was established based on two things: The limits of human hearing and required "overtones" to allow for distinguishing between musical instruments, e.g. if a trumpet, oboe, piano, violin and pipe organ all play the same fundamental, it's the "overtones" that allow our brains to distinguish one from the other. I made a general statement about the maximum fundamental frequency which was not 100% accurate. The vast majority of acoustic musical instruments have fundamentals which fall in the range of an 88 key piano. (Musical octaves aren't precise doubling or halving of the fundamental frequency, but for approximations, I'll assume they are! Concert A above middle C is 440 hz. So, going up around 3 octaves brings the fundamental to or - hz (nominal). Going down 3 octaves puts the fundamental at around 55 + or - hz. The most used "notes" in a piano (and most other acoustic instruments) fall in the middle four and a half octaves, i.e. nominally hz (plus or minus) with the notes above and below being used only occasionally. Including overtones that fall in 1/3's, 1/5's, etc. most "harmonic" energy falls in the first harmonic and considerably less in the second and and much, much less in the third harmonic. Therefore, a hz (nom) bandwidth will cover the fundamental through the second harmonic with a hz.(and even narrower) bandwidth still providing reasonable recognition of different "notes" from different instruments! This is why we can do a surprisingly good job of distinguishing musical (and other) sounds from a highly bandwidth limited AM radio! Increasing the high end one more "harmonic" to 15,000 hz. gives "third harmonic" coverage for the most used notes and second harmonic coverage of the highest notes in the 88 key piano range! (It also pushes the envelope of human hearing! Hi!) (Medical human hearing testing for "intelligibility", BTW stops at 3,000 hz!) One exception to my statement about the frequency range of musical

2 fundamentals is the pipe organ and some of today's electronic instruments. Large pipe organs and some electronic instruments have low notes whose fundamentals drop down to 30 hz and below...frequencies which are often felt more than heard! Hi! At the high end, very large pipe organs and some electronic instruments have high notes that stretch to 7 khz and above!!! (We had an organist at one church who could no longer hear some of the high pipes!) Because these are rarely used by themselves, not hearing them reduces the aural experience, but does not take away basic recognition of the sound of the music! As you know, for safety and intelligibility reasons, the human ear is significantly more sensitive to sounds in the hz range than to those below (the Fletcher-Munson curve)! Being able to hear the "cracking twig" as the bear approaches is important for our safety! These are also the "consonant" sounds in many languages around the world! (Some eastern languages emphasize vowels and so, for them, the higher frequencies are less important.) It also happens that the primary result of audio "distortions", e.g. harmonic distortion, "intermod" distortion, etc. tend to fall, again, in that hz range. We are also more sensitive to "variations" or "differences" in levels than we are to "absolute" levels! I can differentiate relatively modest "differences", but cannot guage actual level well at all. Background NOISE, also plays a part, both in the original material and in the listening environment. The quieter the general "noise" level, the higher the perceived "quality" of the audio. However, the sound must cross a "threshold" of reasonable audibility above the background and still be within the maximum "comfortable" level for listening. So, a quiet original in a quiet environment can have a much wider dynamic range! Conversely, if the listener is in a high background noise environment, the dynamic range must be compressed if the loud passages are to remain "comfortable" and the soft passages are to be heard! So, "psycho-acoustically", people can be convinced that they are listening to a "hi-fi" sound, IF...the frequency range is adequate to distinguish among the sounds...if... the sound seems to be free of distortion...if...the sound is perceived to provide levels at different frequencies which seem to match with "real life" experience and IF the dynamic range fits the listening environment, e.g. emphasize mid-highs for better intellgibility and better ability to distinguish among musical notes. Finally, because there is much less "energy" in the higher frequencies,

3 e.g. above 3000 hz, it is not necessary to have an audio system that can reproduce these frequencies at particularly high levels! (Except for some newer electronic instruments, which can BLOW OUT a tweeter!) So, now you have the components of a "high fidelity" audio signal: (1) Reasonable frequency range hz for music, for voice (nominal) (2) Low distortion (3) Low noise level (4) Appropriate dynamic range. (5) Appropriate "equalization", e.g. an emphasis of the mid-highs. (3) Broadcast OCCUPIED BANDWIDTH is an aesthetic as well as practical issue, i.e. the broader the bandwidth, the more information can be conveyed and the higher the quality of the signal. At the same time, the fewer stations that can occupy the same amount of frequency space without causing interference that is perceived by the listener/viewer as detrimental! Using VERY simplistic terms, in AM modulation schemes, the occupied BANDWIDTH is determined by the FREQUENCY of the modulating signal. In double sideband AM, the bandwidth is twice the modulating frequency. So controlling "harmonics" becomes quite important in limiting occupied bandwidth and interference to other stations. 100% modulation LEVEL or loudness is established at the zero crossing point of the carrier (assuming balanced + and - modulation) because serious distortion artifacts occur when one tries to drive negative modulation beyond the zero crossing point! (The FCC in the past, allowed unbalanced modulation which I think was up to +110% or +120% (I don't remember which) while maintaining a limit at -100%, but I'm not sure if that is allowed any more. This helped stations increase their "loudness" and perceived signal to noise ratio.) As you know, FM modulation is, in fact, a quite complex mathematical model and there is debate, in fact, over the differences and similarities of "FM" versus "Phase" modulation! Hi! Again, in VERY, VERY overly simplistic terms, the deviation FREQUENCY (from the center carrier) is determined by the LEVEL of the modulating signal and the RATE of Frequency Deviation is determined by the FREQUENCY of the modulating signal. (Any AM modulation of an FM carrier is considered distortion!) Unlike AM, in FM there is no magical "zero carrier crossing point" to establish a 100% modulation level. So, what constitutes 100% modulation level is, in fact, an arbitrarily established amount of frequency deviation from the carrier. In FM Broadcast in the U.S. it's + or - 75 khz. In TV Broadcast, it's + or - 25 khz! (And you are familiar with two-way radio FM maximum deviations.) There is a correlation, however,

4 between dynamic range and signal to noise ratios as the maximum deviation utilized changes. Receiver design is part of the equation because as deviation exceeds the design maximum for the receiver, distortion increases. At the same time, as the maximum deviation level utilized as 100% "modulation" increases, the amount of POWER required for comparable coverage area (acceptable signal to noise ratio) goes up. (Modulation power density!) When an FM station broadcasts an FM subcarrier, it must REDUCE it's "main" signal modulation as it injects the subcarrier, IF it is going to keep its overall occupied bandwidth (deviation) within the prescribed + or - 75kHz. And, since the FCC's table of allocations ASSUMES that stations will stay within their prescribed occupied bandwidth, this is important! Hi! The FM STEREO modulation scheme is a somewhat ingenious approach to minimizing the amount of "modulation budget" used to provide the stereo information. First, the "main modulation" is L+R which means that anyone with a monaural receiver will receive a complete signal (backwards compatibility). It is also bandwidth limited to 15 khz and filtered in the receiver so that the 19 khz pilot tone and subcarrier components won't mix or the 19 khz tone be heard by those with SUPER EARS! Hi! Second, the 38 khz suppressed carrier AM L-R subchannel signal, is carrying the least amount of information necessary AND, by using a suppressed carrier approach, with a low level 19khz pilot tone to lock the local receiver's BFO for demodulation, the injection level of the "Stereo subchannel" is kept to a minimum! (Why waste "modulation budget" on an AM "carrier"!) This also left enough "modulation budget" so that stations could continue to offer subchannel services, e.g. talking books for the blind, commercial background music services, etc. without UNDULY reducing their "L+R" main channel modulation level which would decrease their effective coverage area as defined by the point at which the S/N ratio became unacceptably high. (As I recall, the maximum FM broadcast modulating frequency is 75 khz which puts the subchannel carrier in the khz range.) (4) Now, you can begin to see why the FCC sets certain "technical standards" or "proof of performance" requirements on broadcast operations to both provide both a perceived quality of service as well as to limit interference. (5) Occupied bandwidth is only ONE factor in the FCC's table of assignments, of course. Local intercarrier effects in the receiver are also part of the issue. For example, the 10.5 mhz IF strips used in FM receivers establishes "prohibited pairs" in each area. The same holds

5 true for the common 455 khz IF strips in most AM radios. The receiver's creation of "Phantom" stations, again due to mixing in the receiver front ends, is also a factor. And, then of course, comes the whole issue of "acceptable" interference to the coverage areas of other stations, co-channel, adjacent channel and second adjacent channel!

SUBCARRIERS IN MICROWAVE AND SATELLITE SYSTEMS

SUBCARRIERS IN MICROWAVE AND SATELLITE SYSTEMS By: Frank McClatchie FM SYSTEMS, INC 1-800-235-6960 SUBCARRIERS DEFINED: In the early days they were called Diplexers, alluding to their main function at