Chapter 2 SCA TRANSMISSION PARAMETERS (And You Thought Only Doctors Gave Injections?) The SCA signal, hereafter called the "carrier" (for the same bad reasons expressed in Chapter 1), also uses frequency modulated (FM) technique to convey program information. Briefly, what is FM? Well "F" stands for "frequency" and refers to the number of complete cycles of electrical action which occur each second; "M" stands for "modulation" which means to change or vary. (So much for the English lesson.) Thus FM means that the normal frequency that the SCA carrier resides at, 67KHz, is varied! What varies it? Why, the program information you hope to broadcast, of course! I assume we all understand that through the "magic" of microphones, etc., and electronic "black boxes" the sounds we hear such as voices, music, and noise (sometimes these are the same thing to some of us) are converted from being sounds to being electrical signals. In general, the louder the sound then the bigger in size, or "volts" will be the electrical signal ("amplitude"). Also, the higher the pitch of the sound then the more often the electrical signal it produces will go through complete cycles (frequency). When these electrical signals, which represent the sounds, are "plugged into" the device which creates our SCA carrier, they will cause the carriers 67KHz frequency to vary. That is, our "program material" will "modulate" the SCA. The manner which FM operates results in the carrier modulation to be directly proportional to the loudness (amplitude) of the program material. In other words, loud sounds will produce large electrical signals which will move the 67KHz frequency of the SCA carrier a large distance; either up or down, depending on the nature (direction) of the electrical signal. How far the SCA carrier moves is called it's "deviation". It is possible to create such loud sounds that our SCA carrier could "deviate" right up beyond the legal limit set by the FCC at 75KHz. Likewise, going the opposite direction, it could deviate so far downward as to "crash" into the stations L - R stereo information. Neither of these actions would be viewed with pleasure by the FCC, the station management, and perhaps the stations main channel listeners. The Radio Talking Book listeners may or may not be displeased -- their radios would, of course, sound louder but they also might sound "distorted" depending on the radio's design characteristics. In any event, there is a maximum "loudness" that can be used to "deviate" the SCA. The standard value which we arrived at for providing acceptable SCA (and stereo) performance was a maximum deviation of +or- 6KHz on the loudest program peaks. Stating this in engineering terms we would say that...."lOO% modulation equals peak deviation of 6KHz". (An aside comment - What I'm defining here is for Radio Talking Book, speech type, programs. Music services, such as MUZAK, generally use lower deviations in the order of 2KHz to 4KHz. This has to do with "complex" considerations of fidelity, signal to noise acceptance, dynamic headroom required, and the peak-to-average ratio of the program material and the statistical probability of this and the high frequencies being present in the program materials.) Let me simply state that if you don't keep you SCA modulation hitting 6KHz peak deviation ALL THE TIME it won't work very well. This can be interpreted to read, you won't "reach out" very far with your Radio Talking Book service. So, how do we accomplish this very important task? Simple. We use expanders/compressors/peak limiters in our program audio processing system! This is absolutely mandatory to the systems successful operation and will be discussed again in Chapter 4, "SCA Program Studio Facilities". Now that we all understand what "deviation" and "100% modulation" of the SCA means, what say we consider "fidelity" (ability to broadcast all audible sounds over the SCA)? The SCA channel, being "crunched" between the main channels L - R stereo information and the FCC's legally defined barrier of 75KHz, doesn't have much "elbow room" to show off it's stuff. It turns out in the practice of FM (Don't forget - the SCA is an FM system too!) that the width of the channel occupied by the signals is determined not just by how "loudly" it is modulated (the amount of deviation). It also is a function of the "pitch" (or frequency) of the information being used to cause the deviation. In engineering terms, the bandwidth of the channel is a function of the quotient of the deviation and the modulating frequency. This is such an important concept in FM that we even created a term for it...."modulation index". Mathematically, (for anyone who cares), MI = DEV./Fm. Well, to go much further with this concept we would get into calculus mathematics and involve "Bessell functions". (I, for one, don't care to do that here, and as was stated previously, this is a subject for another paper.) The important thing to understand from this part of the discussion is that both loudness and high pitched sounds affect the amount of space occupied by the SCA. And, with very little space to "splash around" in something has got to give in! In short, "you can't have your cake and eat it too!" Hey, but there really is good news! The practical fidelity limit is not really very severe! We have proven there will be no problems using audio program materials with information who's frequencies extend to 5KHZ. (Actually you can "get away" with quite a bit more than this, but we then start to encroach into an area which might be classified as "bad engineering" practice.) What does this fidelity limit of 5KHz mean to us in practice? Not much! Oh, don't limit it to less than this else all kinds of "nasty" effects come under consideration, but 5KHz is quite acceptable. Most speech material certainly contains sounds that exceed 5KHz, however, there is very little information contained in those sounds. All vowel sounds are lower frequencies and constitute the loudness of speech. All consonant sounds are higher frequencies and constitute the intelligibility of speech, however, they are concentrated near 3KHz. Thus if 5KHz of audio fidelity bandwidth is allowed for, then virtually all of the important information is conveyed and with very good usable quality. (We would all like to have the full l5KHz audio bandwidth of main channel FM broadcasts available, but fortunately it isn't necessary.) Again, do not use less than 5KHz audio bandwidth! Loss of intelligibility is a subtle factor which creates listener fatigue and eventual psychological disinterest in your broadcast services by the listeners. (The same can be said for poor microphone techniques; more on this in Chapter 5, "Microphone & Technique"). So - how to determine that you provide a 5KHz audio bandwidth on the SCA? Again, 'tis very simple; use the best "Hi-Fi" techniques in your studio audio systems and tape recording methods. Next, deliver the audio to the SCA transmitter location with as little degradation as possible. (More on this in Chapter 6, "Studio to Transmitter Link") In the SCA generator (That's the "black box" at the transmitter location which creates your SCA carrier) install the 5KHz filter designed by the manufacturer of the SCA generator. Else, and preferably, use a 5KHz (Low Pass) filter in the "limiter" which is installed just ahead of your SCA generator and don't use the filter (or pre-emphasis either for that matter) in the SCA generator. Yes, you MUST have the "limiter" just refered to! This device, "limiter", will prevent your program audio from ever getting loud enough to exceed the 6KHz peak deviation (100% modulation) of your SCA. As you can see, with very little effort we have controlled the loudness (deviation) and the fidelity (bandwidth) of our SCA system. All it cost us , besides the SCA generator you must have to create the service on the FM stations channel, was a 5KHz audio "low pass" filter and a decent quality audio peak limiter. Seems like we're all done. But, nope, there is one more very, very, important consideration! (Move over Doc', here's where we get into injections!) As we've just described, the amount of deviation of the SCA affects the loudness of the programs heard on your listeners radios. Another way this could be stated is that the power output from the radio's speaker gets greater when the SCA is deviated more. We can draw a parallel to this. Remember that the SCA carrier itself exists because the FM station's main carrier was modulated (deviated) by the SCA signal at 67KHz. (Just like all the rest of the main channel program information did.) And, of course, that also is an FM system technique. Therefore, the loudness (amplitude) of the 67KHz signal itself when put into the FM station's carrier will determine how big, or strong, or powerful (you pick the word that feels best to you.) will be the SCA carrier that is transmitted by the FM broadcast station. Thus the more you deviate the main FM station carrier with theSCA signal, the stronger will be the SCA carrier which is broadcast and, therefore, the farther the SCA program service will reach out to your listeners. We say that the SCA is "injected" into the main carrier. Now, there are some FCC inspired legal limits imposed here. The 'Feds' tell us we are not allowed to inject (deviate) the main carrier with our SCA signals greater than that which would result in 30% of maximum allowable 100% modulation of the station, if we are broadcasting in monophonic. However, if we are broadcasting in stereo, that limit is reduced to 10% of the maximum allowable 100%. Well, since 100% modulation of the main carrier is defined (by the FCC) to be a deviation equal to 75KHz, then 10% SCA signal injection means that we are allowed to deviate the main carrier 7.5KHz with our SCA. (Obviously 30% would equal 22.5KHz deviation, but virtually no one does this anymore so 'tis just an academic point.) The point of all this is just simply that the coverage range - (or "power", if you prefer to think of it that way) - of your Radio Talking Book SCA program service is directly dependent upon the level at which it is injected into the FM station's carrier. Do not exceed 10% 'cause it makes 'uncle' very irritated; likewise, do not use less than 9% or it makes your listeners several miles from the transmitter wonder why your signal is not there anymore, or at least why did it get so weak and noisy! Actually, you could "live with" injection levels as low as 8% but again this is bad engineering practice and there is no defendable reason to do it. At this point I must mention that many very misguided souls (most of which should know better, but don't!) would claim that better stereo operations will occur if the SCA injection is decreased. This is pure fallacy based upon incorrect conclusions drawn from improper observations. These same people might just as well remark that automobiles perform quieter when the engines are shut off - of course they really wouldn't perform at all, but they would be quieter! Seems like a huge price to pay to fix an unrelated problem. And so it is with SCA injection and stereo. But we will examine that subject in the next Chapter, "SCA and Stereo Compatibility". Back a few paragraphs, I dropped "pre-emphasis" into the discussion and just let it lie there. Surely someone is gonna' call my attention to this, so let's drag it out and describe what kind of beast this is and why we need one. "Pre-emphasis" is another one of those 50. terms we technical people invent so we won't have to explain what we mean. I'm really not sure if this is inspired by laziness or snobbery - perhaps both? All it really means is a deliberate increase of the volume (amplitude) of the high pitched (frequency) sounds. I could just as well have said that it means to turn up the 'treble' tone control. Let me start by disclosing that all FM stations use pre-emphasis. Why? Well, basically, the reason is because it improves the ratio of the noise to the program sounds at the listeners radio. It is a well known fact that in FM radio the presence of the broadcast station's signal when tuned in on the radio causes the radio to become quiet. Tune off of the station and the radio gets very noisy. Likewise, as the stations signal becomes weaker, the radio gets noisier. It is also a well accepted fact that this noise is comprised of mostly high frequency sounds. So, if the program information is transmitted with it's higher frequency sounds increased, and then at the radio the high frequencies are reduced back to normal levels, any high frequency noises in the radio are also reduced, yet the program is at normal and correct loudness. (Reducing the high frequency sounds at the radio is called "de-emphasis".) This can be done, so the theory goes, without exceeding the 100% modulation level of the FM station since high frequency sounds occur in nature at much lower amplitudes than do the lower frequency sounds. Therefore, after much study of this phenomena, the FCC determined that we should boost the high frequencies before transmitting them with a circuit network which has a characteristic time constant value of 75usec. (Pronounced "micro-seconds") Every FM radio produced also has a circuit in it which reduces the high frequencies by 75usec to be compatible. (Note, these FCC rules apply only to the FM main carrier, not to the SCA. However, preemphasis is used on the SCA but the values are often not 75usec, but are l50usec.) Today, it turns out, there is an interesting problem with 75usec pre-emphasis. When that value was derived, audio processes were not as refined as they presently are. With todays equipment, the correct value should be about 25usec. Mr. Dolby of Dolby Labs (noise reduction systems) deserves most of the credit for study in this field. What this means is that FM stations are overboosting their high frequency sounds, forcing the station to reduce it's average modulation to accomodate this. Because of the many thousands of radios out there in radio-land, the FCC is not anxious nor willing to change the system. What's this got to do with SCA, you ask? Only a little bit. It's kind of an interesting situation so I sometimes get carried away. But, there also really is some relevance. Pre-emphasis on the SCA is used to accomplish exactly the same noise improvement effects as for the main channel. Most SCA's are using l50usec, but I strongly prefer the use of 75usec. My gosh, I can already see the gnashing teeth and clenched fists out there! Just like some folks like Fords and others prefer Chev's for either real or emotional reasons, some of us do and some of us don't like l50usec. I believe it is too severe of a boost and limits the nominal average loudness excessively. Most of the mathematics will favor it. I've calculated the differences, measured the differences, and tried them both (plus a few other strange ideas) in practical operations. I still keep prefering what I get using 75usec, so I keep recommending this practice to my friends and anyone else who will listen. (Come to think of it, most of my friends don't want to listen to me anymore.) So, I'm quite willing to get into a not too rousing debate of this point with anyone, at the first offer of a glass of beer to do it over. OK, let's summarize: SCA 100% modulation = +or- 6KHz deviation (peak) SCA injection level = 9% to 10% SCA audio fre. max. = 5KHz SCA pre-emphasis = 75usec prefered, l50usec optional I can't close off this chapter without one additional practical comment. How do you know if your SCA is operating within these parameters? Well, there are several tests you could conduct if you had the test equipment. (If you don't, you should very seriously consider doing so, along with getting someone who's competent at using it.) Lacking that, every FM station is required by the FCC to have a station monitor unit which, by merely flipping a switch or two, will display the deviation and injection levels for you. Get acquainted with this device and use it. Your Radio Talking Book service depends upon it, so don't be bashful. It also is not a bad idea at all to purchase a monitor and install it at the Radio Talking Book studios so you can keep track of things. Expect a jolt though, these things are not "cheap" - maybe you can find a wealthy friend? (It could be the best favor he will ever do for you, and for the users of your service!)