Loudness
Loudness is the subjective perception of the intensity of sound pressure waves
in the air. There are many different standards for measuring Loudness. Measurements
of loudness may be helpful, but the human ear is very complicated, making the
correspondence between any measurement and a person's perception of loudness imprecise.
The most reliable unit of measurement of
loudness is the phon. Yet another unit
is called Sound Pressure Level or SPL. Only under certain
prescribed conditions will a phon correspond to SPL. (specifically, a db-SPL at 1 kHz).
Two other units that relate to loudness are LUFS and sones. Yet another unit that has
something to do with loudness is the VU or Volume Unit, but remarkably,
of all the units the VU is the least reliable indicator of loudness. The db-VU is a very
easy measurement to make and is very useful for maintaining signal integrity, therefore
it is very commonly used, even though it does not predict loudness well.
Loudness depends on the frequency, the spectral density, and duration of the sound.
The human ear is more sensitive to mid-range frequencies than it is to low or high
frequencies. This means that for a given perception of loudness in phons one needs
more power in low or high-frequency sound-pressure waves. Loudness also depends on
the duration of the sound. Short brief sounds that happen in a generally quiet
environment seem louder than droning long-duration sounds that create a background
of sound. Our ears seem to naturally filter out the background sounds and direct
our focus toward changing sounds. If your brain needs to do a lot of this filtering
work to pick a voice out of background noise for example, it tends to make the voice
seem more powerful than it really is, rather than make the background noise seem weaker.
These are reasons why loudness is difficult to measure analytically. The absolute
measure of sound that best accounts for all these effects is the phon.
Overview of some units for measuring sound levels
decibels
You might be wondering about the loudness unit called the "just plain"
decibel. Actually, there is no such unit of loudness. The
decibel is a mathematical technique
to express a measurement on a logarithmic scale. Any kind of measurement, say the
brightness of light in a fiberoptic cable or the gain of an antenna, can be expressed
using decibels. Just as there are many contexts in which the word micro comes
up, so with the decibel. In order to express something like SPL on a logarithmic
scale we do the mathematical processing and then label the result using the word
"decibel," finally arriving at something like db-SPL or decibel Sound
Pressure Level.
When you see "dB" or hear "decibel," it is just a scaling
prefix. Pay attention to the modifier that follows. That modifier is the real
unit being discussed. Similarly, when you hear "nanometer" you understand that
the basic measurement is length in meters and "nano" is a prefix that scales the
number for convenient use in contexts of very short lengths. Occasionally
someone uses a prefix in a colloquial way. "How many kilos of sugar did do
you want?" We then understand that a unit, such as "gram," is implied. "How
many kilograms of sugar do you want?" is what we understand, even if that was
not exactly what was asked. Similarly if someone talks about "decibels" there
is a unit that is implied and hopefully can be filled in from context.
LUFS or LKFS
Another measure that relates to loudness is called LUFS (pronounced
as a word that rhymes with "cuffs") or Loudness Units Full Scale.
Like the phon and the db-SPL, the LUFS is a logarithmic unit. In spite of the word
"loudness" in its name, LUFS is not really a measure of absolute loudness like phons is.
(A side note on grammar: LUFS is a treated as a singular word. Counterintuitively,
we say, "LUFS is logarithmic." We do not say, "LUFS are logarithmic.")
LUFS is an unofficial synonym for LKFS. The LKFS measurement makes use of
a "K-weighted" frequency filter. The standards document that defines LKFS
(ITU-R BS.1770)
tries to make the K-weighted filter more obvious by naming the measurement with
a "K." Most people call it "LUFS" anyway, probably because how does one
pronounce "LKFS?" (The definition of LKFS is Equation 2 on page 6 of the standard.)
The key to understanding LUFS is in the last two words of its name: "full scale."
LUFS assumes there is some context that establishes a maximum loudness that can be
produced. For example, say you have an amplifier with a volume control knob labeled
with numbers from 0 to 10. (A certain rock band uses amplifiers labeled up to 11. How
the knobs are labeled makes no difference when it comes to LUFS!-)
Now, suppose also that you have set that volume control to the number "4." That
setting establishes a context for the maximum loudness that the program can achieve. That
maximum achievable loudness is defined as 0 LUFS. Everything of the programming that gets
reproduced at a softer volume than that maximum results in a negative number of LUFS.
In programming for streaming media (YouTube, Facebook, etc.) and TV and radio the
average LUFS within the entire mix is designed by the mixing engineer to typically
be somewhere between –6 and –30 LUFS. Most platforms work best with
the average LUFS near –12 LUFS.
The lower the average LUFS is the more contrast the louder sounds of the programming
may have against the average. But an interesting thing about LUFS is that if you
change the playback volume control you do not change the LUFS measurement of
the resulting loudness of the program! That's because if you turn up the volume,
say from "4" to "6," everything gets louder, so the same moments in the
programming that produced 0 LUFS with the volume control at "4" also are the moments
with the loudest part of the programming when the volume control is at "6"!
That new loudness is still 0 LUFS in the context of setting the volume control at "6."
So LUFS is not really a measure of absolute loudness. It is a measure of the present
loudness compared to the maximum possible loudness in the context in which you are
listening. LUFS are an important measurement in mixing, but we need to remember that
LUFS do not measure absolute loudness like phons do. A playback volume control or
any other typical level control changes phons, but not LUFS. (Changing the recorded
or streamed mix—making one or another part of the mix softer or louder—can change
the LUFS.)
Sone
A related but non-logarithmic measure of loudness that one might occasionally encounter
is the sone. One sone is 40 phons, 2 sones is 50 phons, 4 sones is 60 phons, etc.
Each doubling of the sones adds (about) 10 phons to the perceived loudness of a sound. Phons
and dB SPL are the absolute units usually encountered in audio engineering, so no further mention
of sones will be included here. (Sones are used in some industrial noise analysis
contexts.)
VU and dB-VU
The VU or Volume Unit is a measure of a weighted moving average of the
rectified voltage of the signal. It is an easy measurement to make and is very helpful
in preventing clipping and keeping the signal sufficiently above background noise and hum.
The VU is practically always expressed in decibels, in which case it is abbreviated as dB-VU.
Since it is an easy measurement to make and also a very useful measurement of signal
integrity, most mixing boards, recording equipment, and other devices have VU indicators,
either as traditional electro-mechanical meters with needles that swing over a scale,
or with bars of LEDs or some emulation on a screen. VU meters (or bars) are by far
the most common displays of signal amplitude in the audio industry.
But loudness is more related to power and spectrum density than to voltage. Since the VU
is at heart a voltage measurement, of all the measurements so far discussed here, the VU
is by far the least reliable indicator of perceived loudness. To illustrate, YouTuber
Tom Scott has put up a demonstration video titled, "Why are adverts so loud?."
In this demonstration a person standing in a forest whispers into the microphone,
which sounds appropriately calm and soft. Then the person, now standing next to a
highway, shouts into the microphone which sounds appropriately much louder.
Analyzing the VU levels reveals amazingly that the whispered voice is mastered to a
level of –4 dB-VU (quite high) and the person shouting is mastered to a level
of –12 dB-VU, eight decibels lower! However, when measured in LUFS, the person
whispering is 7 LUFS softer than the person shouting. Clearly a dB-VU measurement
is not a reliable way to assess perceived loudness. (I have analyzed this video myself.
Even after playback through YouTube, Tom Scott's claim of the eight dB-VU difference
is correct! It is a bit of a gaffe that in one part of the video he calls the unit
of dB-VU simply the "decibel" and then later discusses dB-SPL and again calls that
just the "decibel," leaving it to the audience to sort out the unit from the context.
But the demonstration is well designed, vivid, and correct.)
Tom Scott explains that the whispered voice has the highest dB-VU measurement suggesting
that it will be perceived as louder than the shouted voice when in fact it is perceived as softer.
(Screenshot from his video at about
6:37. [slightly edited, "dB" changed to "db-VU" here])
The units discussed in this section (LUFS and dB-VU) are popular, but they are not true
measurements of the absolute loudness of a sound. They are characteristics of a signal
that only relate indirectly to the loudness of the related sounds. The units discussed
next are less often used, but they have much more to do with perceived absolute loudness of
sounds.
Sound-Pressure Level (SPL) Defined
Sound-pressure level (SPL) is a more easily defined quantity than loudness is.
Sound-pressure level is the root-mean-square (RMS) deviation of a sound wave's pressure
from normal atmospheric pressure. (Yeah, loudness is still more complicated than
that!) The SI unit of sound-pressure is the pascal (Pa).
The threshold of human hearing is usually considered to be 20 μPa. (micro-pascals)
The threshold of pain (where discomfort is first noticeable, not outright pain) in
human hearing is usually considered to be 20 Pa. The maximum possible sound-pressure
level in the normal atmosphere near sea level is 101325 Pa. It is a maximum
because 101325 is 1 atmosphere of pressure. You cannot take more air molecules
out of a volume of space that is already lacking any air molecules! Air pressure
disturbances that go beyond 101325 Pa are possible, but these would best be
characterized as blast-waves, such as are associated with bombs or objects moving at
supersonic speeds. These can create very high positive pressures, but no pressures
lower than 0 Pa. Audio equipment cannot create the asymmetric pressure profile of
a blast-wave. Thus, it is fair to say that the highest possible sound-pressure level
is about 101325 Pa.
Because our ears respond logarithmically, we usually convert SPL in Pa to a logarithmic
scale. This is called the "decibel sound-pressure level" or "dB-SPL" ("dee, bee, ess,
pea, el"). It is twenty times the base-ten logarithm of the ratio of the sound-pressure
level in pascals to 20 μPa. The formula for conversion is:
dB-SPL = 20log10([SPL in Pa RMS]/[20 μPa])
The formula above requires the sound pressure level to be in "Pascals RMS."
To convert peak Pascals to Pascals RMS, as a rough estimate dividing the peak
value by 1.4 (√2) will get you close most of the time. (An exact conversion is
waveform dependent and may require some application of calculus—a topic beyond
the scope of these pages.) The maximum possible peak
SPL, as explained above, is 101325 Pa. To convert this to Pa RMS divide
by √2. This gives about 71648 Pa RMS. Using the formula above
one finds that 71648 Pa RMS is equivalent to 191 dB-SPL.
Here are some typical sound-pressure levels:
191 dB-SPL is the maximum possible sound pressure level in a normal atmosphere.
120 dB-SPL is the threshold of pain (discomfort—not outright pain).
110 dB-SPL The audience experience about 50 feet from the stage at a rock concert
85 dB-SPL A factory's production area. (OSHA regulations become an issue!)
80 dB-SPL Vacuum cleaner
60 dB-SPL Normal human speech about 3 feet from the person speaking
50 dB-SPL A lightly occupied department store with background music playing
40 dB-SPL Empty classroom with a fan or ventilator running
30 dB-SPL Empty movie theater
20 dB-SPL Outdoors in the countryside far from traffic, slight breeze, no birds, etc.
20 dB-SPL is also a desirable goal for an empty sanctuary or concert hall.
10 dB-SPL Two people 3 feet apart in an anechoic chamber.
They hear each other's breathing and their own heartbeats.
0 dB-SPL The sound of a pin dropping about 10 feet away in an anechoic chamber.
Quietness is important too
In the chart above 20 dB-SPL is mentioned as a goal for the quietness of an empty sanctuary. The rumble of heating ventilating and air-conditioning equipment, the buzzing of lamp ballasts, the bustle of traffic on the streets outdoors, and similar sounds can distract from a sense of reverence and peacefulness while also obscuring the details of softly performed music. Even achieving a quietness of 30 dB-SPL is non-trivial and will require some deliberate effort to quiet down various aspects of the venue. If care for quietness is not a deliberate effort, the level of quietness will likely be degraded to something above 35 dB-SPL in today's typical world of noisy equipment.
SPL Is Not Loudness!
BUT. . . dB-SPL is not a good measure of loudness. Consider that using a "vacuum
cleaner" might seem louder than working in a "factory production area." As another example,
if you could hear the (normal) breathing of a person 3 feet away from you (easy if you
are in an anechoic chamber) you would think the breathing is unusually loud, but the same
breathing in most situations is completely "silent," masked by other sounds. Or you
might perceive that an empty classroom is just as quiet as the outdoors (far from traffic),
even though the dB-SPL levels are far apart. The power that a sound wave carries is
accurately expressed by dB-SPL, but our perception of that can be very different than the
dB-SPL numbers would predict.
As another example of loudness vs. dB-SPL, consider that in the USA the FCC requires that
advertising may not have an average dB-SPL level above that of the normal programming
on radio and TV. Just listen to some programming. Some advertisements sound much louder
than the program they support. This is possible because loudness, being a complex issue,
is not what is measured and regulated. Specifying that comparable dB-SPL levels much be
achieved implies that voltages are actually what is regulated, limiting only sound-pressure
levels. Europeans have it better because their regulations are in terms of LUFS.
Loudness depends on frequency and dB SPL
In the 1930's engineers at Bell Labs experimented to try to discover a
relationship between sound pressure level (in dB-SPL) and the resulting loudness. They
discovered that the frequency of the sound used for the test changed the results. They
proposed a set of "curves" now called the Fletcher-Munson curves. These curves are the
best information on this topic that are in the public domain. (Some companies have
done their own research and use different curves when designing their
products.) The Fletcher-Munson curves (red lines) are shown below. One of the characteristics
of the Fletcher-Munson curves is that a phon is defined as equal to a dB-SPL at
1000 Hz for short-duration sounds against a background of silence. If a pure tone
is at a frequency other than 1000 Hz, find the point (Hz, dB-SPL) on the Fletcher-Munson
chart and interpolate the red line that would intersect that point. That is the
loudness level in phons for that sound. Example: Measuring a 30 Hz sound reveals it
has a sound-pressure level of 115 db-SPL. This sound has a loudness of only 90 phons.
The entire set of
Fletcher-Munson curves is only valid for short-duration sounds against a background
of silence. These curves also represent an average of all the samples from a large
number of young people who claim to have no problems with their hearing. The curves
do not account for the effects of age. Any individual person's hearing can deviate
significantly from these curves.
As explained above, the unit of loudness called the phon is derived from sound
pressure level in dB-SPL and from frequency. The Fletcher-Munson curves express this definition.
Not all manufacturers who make audio equipment are satisfied with the Fletcher-Munson
curves. After all, that data is about a century old now. Many manufacturers use curves of
their own devising which they hold as trade secret information. The general name for
these curves, be they Fletcher Monson curves or some other proprietary set of curves,
is equal-loudness contour curves. The curves shown above are the Fletcher-Munson data
for equal-loudness contours.
Loudness compensation
Because loudness depends on frequency, artificially changing the db-SPL level of a sound
will change the way we perceive the sound's tonal balance. Every time the "volume" control
of a radio is adjusted, our ears perceive the change as if the bass and treble controls of
the radio are also adjusted. Turning the volume up increases our relative perception of
bass and treble relative to the midrange frequencies. The sound seems more full and
pleasing to most people when the volume is turned up. Conversely, turning the volume down
decreases our ability to perceive the bass and treble sounds relative to the midrange
sounds and the sound seems thin and low-quality.
Some radios and consumer-grade audio equipment include special tone-control circuits
that automatically adjust themselves as the volume control is adjusted to compensate
for this nature of human hearing. Sometimes the "volume" control is then labeled as
a "loudness" control instead to inform the user that a loudness compensation circuit
is included. Loudness compensation is usually included only on the more expensive
consumer-grade audio equipment and in some powered speaker systems (professional or
consumer grade) that might be used in bar or restaurant type contexts to play recorded
music. Usually if loudness compensation is available it can be switched on or off.
The switch is usually labeled something like "loudness" or "contour."
This Stereo system has loudness compensation. If the "LOUDNESS" button is pressed
(turned on), then when the volume control is adjusted the bass and treble levels change
automatically to compensate for your ear's change of frequency sensitivity. The bass
and treble controls act in addition to this. The loudness compensation is all electronic.
The bass and treble controls remain stationary as the volume control is turned, but the
amounts of bass and treble are changed by the compensation circuitry. (Pixabay)
Front view and rear-panel view of a Mackie powered speaker that has a loudness compensation
circuit. If the "CONTOUR" button is pressed (turned on), then the bass and treble
levels are boosted to compensate for your ear's change of frequency sensitivity When
the volume is low. Mackie recommends this be used for playing recorded music at low
levels. (Fair use)
If loudness compensation is available, it
is not appropriate to use it in a a live-sound context since live-sound is the reference
against which loudness compensation is supposed to adjust itself. Use it only to
play back recorded music. (Usually it is more appropriate to use equalization in
the channel on the mixer than a loudness contour on a powered speaker anyway.)
Since loudness compensation is almost never used in professional equipment intended for
live-sound reinforcement (with an exception for some powered loudspeakers that
could be used for playing recorded music), the circuit is almost never included
in the first place.
Suppose you attend a live worship service in which the musical portions of the worship
service are at about 80 phons of loudness. Now you go home and listen to a recorded
version of the very same worship service but you set the volume control to provide about
60 phons of loudness. This is a very common scenario. Usually when we listen to
programming at home we listen at a lower loudness level than that of the venue at the
live event. The greater curvature of the 60 phon curve compared to the 80 phon curve
tells us that the lower-loudness playback of the sound is not going to sound as good
as the original sound, unless some steps are taken at the playback equipment to
compensate for the lower loudness.
For example it takes about 80 dB-SPL at a mid-range frequency such as 440 Hz
(A4 on a piano—the A above middle C) to make 80 phons of loudness.
But at the lower frequency of 55 Hz, (A1 on a piano—the lowest A on
a piano) it takes about 100 dB-SPL to achieve the same loudness. That is a 20 dB
difference. But if the music is now reproduced at 60 phons of loudness, then 440 Hz
requires 60 dB-SPL (20 dB-SPL less than the live sound) and 55 Hz requires 90 dB-SPL
(only 10 dB less than the live sound). Relatively speaking, the low notes in the
music will have lost 10 phons of loudness in the reproduced music if all frequencies
are reduced 20 dB-SPL from the live sound level. Thus hi-fidelity sound systems
have bass and treble tone controls so that you may boost the low and high frequencies
back to a perceived normal level. A few high-fidelity sound systems have a loudness
compensator circuit connected to the volume control to automatically boost the bass
and treble frequencies as the volume is turned down. Although perfect calibration
of such an automatic system is unlikely, this is usually a helpful feature, if it is
provided. Loudness compensation can usually be turned off, if desired, with a switch.
More about loudness compensation can be found at the Wikipedia: loudness compensation.
The types of tone-control adjustments found on home stereo equipment are not the types
of adjustments needed in a live-sound or live-stream mixing environment. They are
appropriate for playback so that the various listeners may adjust the
tone to their liking based on the loudness levels they are using. These adjustments
should not be baked into the live-stream or a recording mix because every listener's
situation will be different. Tone control adjustments such as used in home sound
systems most certainly will not sound pleasing in live-sound reinforcement because
the live sound is by definition at the loudness level of the live sound! Tone
control for live-sound is an entirely different topic, called equalization.
(More will be added to this web site on equalization in the
future, hopefully.)
Loudness Fatigue
Loudness Fatigue, also known as listener fatigue, auditory fatigue, ear fatigue,
and other similar names, has two aspects to it.
First, there are some muscles in the inner ear (the tensor tympani and the stapedius) that allow your ear to adapt to various
sound-pressure levels. This phenomenon is called the acoustic reflex. In a quiet environment, probably below about 60 dB-SPL, these
muscles completely relax and allow the maximum motion from the diaphragm in the ear to
be transmitted into the cochlea. As the average sound-pressure level of environment
increases these muscles tense up and damp out a portion of the sound to reduce the
pressure waves transmitted into the cochlea. For most people these muscles are
significantly tensed at sound levels above about 80 dB-SPL. This muscular action in your ear is
entirely subconscious and unnoticed. This is one reason why various sound-pressure
levels do not relate so directly to loudness. Our ears have a sort of internal
protection mechanism to muffle loud sounds.
After a long duration of loud sounds
these muscles fatigue. Their protective value is reduced and damage from excessive
pressure waves begins to happen to the cochlea's stereocilia. Not only is there
damage, but the details of the sound are not properly perceptible under these conditions either.
If a person operating a mixer board suffers this type of loudness fatigue, this person
will be handicapped in creating a good mix. Usually taking a break from the loud
environment for 15 minutes or more will restore some quality of hearing. Usually
most of the damage will heal itself in a matter of hours or a day or so, but some
of the damage can be permanent.
When muscles near your brain tense up the electrical
activity of the related nerves creates a whining sound in your ears. Try biting your
teeth together hard while listening. You will hear the whine. If you hear this
whining sound even when you know it is silent around you and all your facial
muscles are relaxed, you may be hearing the muscles in your ear! You should protect
your ears by minimizing your exposure to sounds that cause these muscles to tense
up a lot and for long durations. It is not good to let these muscles get fatigued.
Sometimes when people speak of loudness fatigue, these muscles in your ears are what
they are referring to.
By the way, those
under the age of three might not have such fully developed protective muscles in their
ears. If you see toddlers in the congregation with their hands over their ears, it
is too loud! (Sadly, I have seen this in certain loud praise-and-worship situations.)
Secondly, research has shown that loud sounds over a duration of time cause a general
bodily fatigue, in the sense that one loses focus on the task at hand, whatever it is,
and just wants to take a break and rest or sleep. If a sermon is blasted too loudly
at the congregation, this type of fatigue can make it harder for the congregation to
listen attentively. This is a second way in which loudness fatigue presents itself.
Loudness fatigue, in either form, makes it tempting to keep turning up the volume.
If one is having trouble hearing detail in the mix, the natural instinct is to turn up
the sound in the headphones. It will work briefly, but then the fatigue will set back
in. If the congregation does not seem to be paying attention, heads are nodding, turning
up the volume will wake 'em up—briefly. In any of these situations it is best to
recognize what is happening. If necessary, take your brief advantage of louder sound to get over
the moment, but then the best thing to do is to gently and slowly back the volume down
to a lower level that it was. Everyone will feel a sense of refreshment!
The matter of loudness fatigue also relates to dynamics in music. It is not a
good idea to ride a fader with the goal of keeping the loudness totally uniform. Usually,
performance levels are well above 50 dB-SPL and the event is more than 15 minutes long so
loudness fatigue is a factor to minimize. Let the musicians swell and fade their sound,
maybe even help exaggerate that ever so slightly. The softer moments will help avoid
loudness fatigue. Especially if you know that a loud crescendo is coming, plan for some
relaxation ahead of it. Over the course of a verse (e.g. of a hymn) slowly ramp down
the faders achieving a 3 or 6 dB net reduction over the entire span of the verse.
Then as the crescendo starts, slowly raise the faders back up to normal or maybe a few dB
above normal, saving about 2 dB for the final coda. This will usually be more
satisfactory than doing nothing or only pushing the fader up above normal. Because you
will have allowed the listener's ears to relax and rest a little in preparation for the
crescendo, the crescendo will sound appropriately louder, whereas doing nothing or
just making it louder will cause the ear's muscles to damp out much of the perception
of the crescendo. Similarly, if you see a quiet hymn of confession coming, pull back the
relevant faders maybe 3 dB just before the hymn starts to provide a sense of restful
attentiveness. Take advantage of that moment. (It is important to be careful that
the volume changes agree with the music and do not seem obvious.)
Within a musical mix it is not necessarily desirable to maintain the same mix throughout
a hymn or song. Especially if there is a solo part, there is an opportunity to reduce
listener fatigue. For a solo part it is usually good to push up the fader on the soloist
to help the vocal to ride over the rest of the mix. But you can also pull down the
group master (or if not using groups, use the main master faders) about the same amount
at the same time to maintain the overall dB-SPL level while leaving the perception
of the soloist being louder. Even if there is no solo part, you can make a choice to
slightly solo one different vocalist on each verse and then maybe slightly solo an
instrument on the refrain or chorus. If you do this in a coordinated and subtle way
(keeping changes to just 3 or 6 dB), so that you do not keep pushing up the dB-SPL
level, you will avoid increasing the listener fatigue factor and the whole mix will
sound louder and better.
Safety Matters Related to Loudness
There are many standards relating to loudness that apply in various contexts. Churches,
generally speaking, are exempt from these legal regulations. Nevertheless, churches have an
ethical responsibility to keep loudness levels down to safe levels. The best complete
overview of this issue that this author is aware of is from the World Health Organization.
It is the WHO global standard for safe listening venues and events. The Standard comprises
six features which, when implemented, "allow audience members around the world to enjoy
amplified music with protection of their hearing, while also preserving the integrity
of the artistic experience." Briefly, the six features are. . .
1.) Establish an upper limit for the sound level.
2.) Continuously monitor the sound level using appropriate measuring equipment.
3.) Understand (and if necessary, minimize) the variations of loudness in the venue.
4.) Provide personal protective equipment when and where needed (e.g. for musicians.)
5.) Provide quiet zones for persons with compromised hearing (e.g. colds, illnesses)
6.) Provide appropriate training for various personnel.
Interestingly, this document also recommends flying subwoofers to keep sound pressure
levels from the subwoofer more uniform across the venue (See Annex 9 in the document).
Subwoofers are heavy and flying them is usually an expensive proposition. Thus they
are usually placed on the floor at the front, exposing the front row seating area to
higher low-frequency sound levels than the rear seating areas. The author of this
House-of-Worship guide agrees with the WHO document both on the matter of safety and
also, it just sounds better when the subwoofer(s) is (are) flown. ("Flying" a speaker
means hanging it up high near or against the ceiling.)
Use your ears to judge loudness
The "VU" meters on a mixer board relate more to dB-SPL than to loudness. For technical reasons the VU meters need to be bouncing in the normal ranges, but this has little to do with the artistic value of the mix. The type of programming matters. As one example, an electric guitar that has been processed with a fuzz box can sound much louder than the vocalist, even if both are peaking at the same levels on a VU meter. Similarly, a vocalist can swamp out some instruments, such as a triangle, even if the meters say the levels are about equal. Use your ears and even your sense of fatigue as the judge of what sounds good and well-balanced.
Return to HoW Sound homepage