Shavano Music Online

    General Sound Information

    9/2004 - Jens Moller http://www.colomar.com/Shavano/general_sound_info.html
    Note: This article is aimed specifically at Pro-Audio (PA Systems, Guitar Amplifiers, etc), not Home or Car Audio. While many of the concepts will be the same for any audio system, some things will only apply to Pro-Audio. The information discussed here is primarily for creating music (Live Music), not reproducing previously recorded music (ie. CDs, MP3s, etc).

    Overview

    Many people want to build multi-speaker sound systems. There are many generalities that come into play when doing this. This article attempts to give you some guidelines to help you make more informed decisions. Please recognize that your needs may be different.

    Voice Frequency Ranges

    Speaking fundamental frequency is the average pitch at which one speaks. Research has been done to determine average speaking pitch for different age groups.

    How this was done, and how to find your pitch:

    • Say 'mm-hmmm' several times
    • Say 'mm-hmmm' a last time and sustain the 'mmm'
    • Match the pitch of your 'mmm' to a pitch played on well-tuned piano or other musical instrument
    • When you find the note; this is your speaking fundamental frequency.

      The following speaking fundamental frequency norms have been provided as a result of Dr. Daniel R. Boone's work published in 1991:

      SexAgeNormal Speaking PitchFrequency
      Women 21 G below middle C  196 Hz
      Men 21 C below middle C  130 Hz
      Women 51 F below middle C  174 Hz
      Men 51 A2 (8 notes below middle C)  110 Hz

      Middle C = 262 Hz. For more info, see: Frequencies of Musical Notes

    The added issue is that voices are not simply the fundamental frequency, bit they also include harmonics (which are multiples of the fundamental) and something called Formants. A Formant is a resonance within a vibrating body - the singer's vocal tract. Because every singer's vocal tract (vocal cords, mouth, nose, tongue, lips, etc.) has a slightly different structure, the Formant of each person's voice is unique. Even when the pitch changes, the Formant stays the same, giving that person's voice its own characteristic sound; Instruments also have Formants.

    For men, the main Formant range is around 2.5 kHz to 3 kHz, while in women, the range is roughly 3 kHz to 3.5 kHz. There are also low Formant ranges: 500 Hz for men and 1 kHz for women. There are harmonics to deal with as well, but they will not be as powerful as the Formants - assume that the will be at least one octave above (2 times the frequency indicates a 1 octave increase).

    As a result, the Midrange frequencies from 1 kHz to 4 kHz are critical for Vocal intelligibility in a PA system, however in order to maintain richness in the sound, you need to also provide for the Fundamental (which should be at least as low as the persons normal speaking voice). You should also be able to provide up to 18 kHz to keep the voice crisp.

    PA systems were created primarily for voices, however, they often include other instruments run thru the PA system. In order to determine what ranges you need to cover, you need to know more about the frequency ranges they produce.

    Limitations of Woofers, Horns and Tweeters

    Reproducing a full spectrum of sounds is pretty hard to with a single speaker. It becomes impossible when the volume levels get cranked to what is needed for Live Music. To address this, purpose built speakers have evolved to fill that need.

    Realistic Usable Frequency Ranges

      Bass

      20 to 200 Hz - Sub Woofer

      100 to 5000 Hz - Woofer

      Midrange

      500 to 5000 Hz - Midrange Speaker

      800 to 7000 Hz - Midrange Horn/Compression Driver

      1800 to 20000 Hz - Midrange Piezo

      Highs

      2000 to 8000 Hz - Tweeter

      4000 to 20000 Hz - Piezo Tweeter

    While it looks like the Woofer should be able to handle most everything by itself, what you see are the ranges that it makes sound at, what these specifications don't tell you is the level (in dB) that is output for each of these frequency ranges. A goal of high end Home Audio speakers is to attempt a flat a response as possible across the whole range - this is not often the case with Musical Instrument speakers - they are optimized to give you more output volume (SPL) at the expense of flatness, and its assumed that you will have to adjust the EQ to compensate.

    To get a good sound thru the PA system, you really need to have at least 2 speakers (a Woofer for the lows and something to cover the mids and highs), or a 3 Speaker system that has optimal coverage of the Bass, Mids and Highs. For a vocals only PA system, there is not much need for a Sub Woofer. If you need support of the extreme low end, then you add Sub Woofers as separate boxes.

    Please see Speaker SPL Information before running out and buying some thing that you see on sale. You get what you pay for and for PA and Musical Instrument systems, you do not want to use Low SPL Home or Car Audio speakers.

    Purpose of a Cross-Over Network

    As it turns out, Individual Audio speakers usually have a lot of trouble trying to reproduce the full spectrum of sound that is necessary to produce the richness of sound that you need. Depending on the need, its possible to simply pick a speaker that covers the most important range for Vocals and leave it at that.

    An example of that is the outdoor public address systems that are basically a bunch of waterproof horn enclosures hung high above a crowd. These Horns are very specific for Midrange and above. They have very small diaphragms and cannot move enough air to reproduce much signal below 1000 Hz. When you hear people talking thru these, you hear the Formants in action, but you never hear the fundamental frequencies. To understand a spoken message, you don't need to hear the fundamental. When music is played thru these, all of the bass portion is lost.

    To regain the missing Bass, you could add a Woofer, however you really don't want to load of 2 speakers on the power amplifier when all you want is the Woofer to cover for the Horn when it can't get down to specific notes.

    The solution to this is a Cross-Over network. A Cross-Over network splits out frequencies, separating highs and lows from each other. Complex waveforms will have high and low frequency components; all the Cross-Over network cares about is separating them into 2 or more parts based on frequency range. There are 2 types of Cross-Over networks, Active and Passive.

    Active Cross-Over networks are used between the Audio Mixer and the Power Amplifiers. I say 'Amplifiers' because you will need a separate power amplifier for each frequency range that you have broken the signal out to. This tends to be expensive, but is by far the most efficient and controllable use of Power Amplifiers and speaker systems. All high end stadium system split at least the Sub Woofers from the PA speakers using this method (and usually more parts).

    Passive Cross-Overs attempt to split the raw power amp signal into different frequency ranges that you assign a specific speaker to - normally, you choose a speaker/horn/tweeter (all with the same impedance) that best matches the range desired. Since the individual speaker/horn/tweeter is not conducting the same frequencies, your Power Amplifier thinks that it is seeing a single speaker out there instead of multiple speakers. Of course, you have to make sure that you choose the Cross-Over points so that there is no overlap. Most Multi Speaker Pro-Audio cabinets use Passive Cross-Over networks.

    For more information on Passive Cross-Overs see:

    • First Order Cross-Overs
    • Second Order Cross-Overs
    • 2 Way/3 Way Cross Overs

    Power Usage and Frequency Ranges

    When selecting speakers, you will find that there are many different power ratings for the various types of devices. Logically, it would make sense that all the speakers in 2 or 3 Way cabinet should have the same power rating, but that is not the case. If you take an average music mix (Pop music for example), where no Sub Woofer is in use, you will find that the power requirements for different frequency components vary quite a bit. As a general 'rule of thumb', in the average music mix, the frequencies below 500 Hz use approximately 70% of the power. The frequencies above 500 Hz use the remaining 30 % of the power. The signals above 4000 Hz use approximately 15% of the power. This may help to explain why Woofers tend to have large magnets and high power ratings and Tweeters tend to be very lightweight and have fairly low power ratings - separating the signals into frequency ranges allow you to size the speaker/horn/tweeter for its use.

    Example 1

    You have a 200 Watt power Amplifier. You want to build a 2 way speaker system for it that has a 12 inch woofer and a Midrange Piezo horn (No Cross-Over is required for this configuration). The Piezo naturally starts responding at 1800 Hz. What should the speaker power ratings minimally be for safe operation?

    • Woofer = 70% of 200 = 140 Watts RMS
    • Midrange Piezo = 30% of 200 = 60 Watts RMS

    Example 2

    You have a 300 watt power Amplifier. You want to build a 3 way speaker system for it that has a 15 inch woofer, a midrange compression horn and a tweeter. The Cross-Over points are 1000 Hz and 4000 Hz. What should the speaker power ratings minimally be for safe operation?

    • Woofer = 70% of 300 = 210 Watts RMS
    • Compression Driver = 30% of 300 = 90 Watts RMS
    • Tweeter = 15 % of 300 = 45 Watts RMS

    Your music may be quite different than the 'average' cited; many people push more bass than average recorded music, so you might want to make the Woofer closer to the power amplifiers rating. Keep in mind that the peak power rating of speakers for Musical Instrument use is normally 2 times the RMS rating. So, in the above 'Example 2', the 210 watt Woofer can handle peaks of 420 watts.

    Adding a Sub Woofer

    Normally, the lower the frequency, the more power is needed to get it heard. Sub Woofers need a lot of power when used in a Pro-Audio situation. Often these will have their own power amplifier (many have built in Active Cross-Overs). If using a Passive Cross-Over, you typically split into a Low Pass out for the Sub Woofer and a High Pass out for the rest of your speaker system. In many cases, effectively driving a Sub Woofer will require that you increase the power Amplifiers output by 50% or more. Plan for this if you expect to add Sub Woofers later.

    Feedback is a bad thing for your Tweeters

    As you will note. The Cross-Over network can minimize the power going to different speakers. One exception to this rule is Feedback. A blast of sustained feedback can provide close to the full power of your power amplifier into your tweeters - their voice coils will not last long if this happens. Be aware of things that cause feedback and do what you can to avoid it.

    Guitar/Bass Guitar use

    In most cases, these instruments are best served by speakers alone, and you can assume 100% of the sound should go to speakers. A lot depends on your choice of speakers for the sound that you want. Guitar and Bass Guitar have interesting Formants and your style of playing can really affect the need for additional high frequency support. When I add high frequency support to a guitar or bass guitar speaker cabinet, I'm trying to boost the overtones and harmonics. Distortion boxes create odd harmonics, which results in high frequency that cannot be represented in a Woofer alone (sometimes you want the speaker to roll this off naturally for you). In the case where you do decide to add a high frequency boost in this environment, assume that you will need to be able to handle approximately 25% to 50% of the normal amplifiers power.

    Return to Shavano Music Online Home page

    Questions? Comments? .

    © 2004, Shavano Music Online