Engineering Calculations Using Mathcad

Engineering Calculations Using Mathcad
PDHonline Course E126 (2 PDH)
Sound System Design for
Cafeterias, Auditoriums and
Small Churches
Instructor: Thomas Mason, P.E.
2012
PDH Online | PDH Center
5272 Meadow Estates Drive
Fairfax, VA 22030-6658
Phone & Fax: 703-988-0088
www.PDHonline.org
www.PDHcenter.com
An Approved Continuing Education Provider
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Sound System Design for Cafeterias, Auditoriums
and Small Churches
Course Content
Sound systems have been in use since the 1940’s, when Frank Sinatra first
used a microphone to convey intimate music to the audience. Early sound
systems were limited by tube-based electronics - noise, frequency limits and
non-linearity. Modern solid-state electronics have almost unmeasurable noise,
response well beyond human hearing and linearity beyond human detection.
The remaining problems are selection of equipment, location and installation of
the equipment and room acoustics. These problems, some underlying
definitions and examples are illustrated and addressed after the system
graphic below:
Microphone
Microphone cable
Microphone
connectors
Preamplifier
Mixer
High-end mixer
(Remote control)
Signal processor
Assistive Listening
Power amplifier
120V
120V
120V
V
120
120V
V
120
Speaker wiring
Speaker
Sound System Graphic
•
Sound characteristics - waves, complex waves, frequency, low
frequency, high frequency, linearity, noise, sound intensity.
•
Room acoustics - size, reflectance, resonance, aesthetics.
•
Sound system components - sound lectern, microphone, microphone
cable and connectors, preamplifier, mixer, high-end mixer, signal
processor, power amplifier, speaker wiring and speaker.
•
Sound lectern - power source, components, limitations, cost.
1 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
•
Microphone - types, power source, pickup pattern, limitations, cost.
•
Microphone cable - conductors, shield, jacket, limitations, cost .
•
Microphone connectors - types, conductors, shield, limitations, cost.
•
Preamplifier - types, limitations, cost.
•
Mixer - types, limitations, cost.
•
High-end mixer - features, limitations, cost.
•
Signal processor - functions available, limitations, cost.
•
Power amplifier - features, limitations, cost.
•
Speaker wiring - conductors, jacket, limitations, cost.
•
Speaker - types, limitations, cost.
Each of these concepts and components will be discussed in some detail.
Sound Characteristics:
Noise is the phenomenon of vibrations reaching your eardrum and being transferred to your
brain where no meaning or pattern is detected. Sound is the same thing, except your brain
detects music or a voice message or a meaningful signal tone.
Vibrations which can be converted by your eardrum are in the range of 20-15,000 Hz with an
intensity of 0-100 dB(C) SPL.
“Hertz” is the new name for “cycles per second”, The way you find the frequency, or Hertz, of
a vibration is by connecting it to a spectrum analyzer. Until recently, a spectrum analyzer was
a sophisticated $10,000 instrument. Today, it is a free software program you can download
from the internet. (See CoolEdit in the references at the end of the course.)
The display on a spectrum analyzer is frequency on the horizontal axis and amplitude on the
vertical. You can adjust the controls for wide-spectrum, 20-15,000, or your choice of narrow
spectrum, perhaps 300-7000, the critical voice frequencies. The spectrum gives graphic
evidence of low-end roll-off (the lowest tones present) and high-end roll-off (the highest tones
present). Over time, it shows how uniform the amplitude is over the range and if any peak
frequencies are present. When no signal is present, it shows the amplitude and frequency of
system noise.
With very little trouble, the spectrum analyzer produces numeric measurements of the actual
spectrum present, linearity, peaks and noise. This is how to scientifically compare music
reproduction systems.
To test the air-borne music quality and room acoustics, you need a microphone which has
better response than the system being tested. A studio-quality large-diaphram condenser
microphone costs about $60 and plugs directly into the computer sound card.
It is interesting to watch the spectrum of an Elvis Presley song, but for sound system design,
pure tones and pink noise are used as sources. Both of these sources are available as free
software programs from the internet. (See SweepGen222 in the references at the end of the
course.) A single pure tone is useful for multiple measurements of the system. The speaker
distribution chart below was made at 1,000Hz, a common test frequency. Obviously, the
2 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
results relate only to system performance at 1,000Hz. The non-uniformities in the sound
pressure level gradient and the extreme sensitivity to sound meter location are results of
interferences produced by a pure tone in a real-world space. Fortunately pure tones rarely
exist outside the laboratory.
The goal of the measurements was to verify the predicted drop-off in sound level and
determine the off-axis performance of a commercial enclosed speaker.
Field Measurement dB(C) SPL
Test Location
13Ft
80
7M
84
85
78
6M
84
79
84
5M
82
89
83
4M
73
3M
77
87
80
2M
80
81
2X
1M
77
94
Ref
100
Base
Speaker
5M
81
88
74
4M 3M
13Ft
82
87
83
73
75
87
2M
1M
83
90
79
78
91
0
1M
Speaker
73
85
76
8X
8M
87
77
4X
71
80
77
74
2M
3M
4M 5M
13Ft
Distance Ratio
26Ft
Free-Field Sound Distribution, 1,000Hz
20W Bass Guitar Amplifier(System)
Field Note: Very location sensitive measurements.
1/2-in meter location shift produces +/- 3dB value shift.
On-Axis Data Analysis
Location
Distance
Ratio
1M (3Ft)
Base
2M (6Ft)
2X
4M (13Ft)
4X
8M (26Ft)
8X
Measured
dB(SPL)
100
94
90
83
Theory
dB(SPL)
100
94
88
82
Difference
Percent
0 dB
0 dB
2 dB
1 dB
-2.3%
-1.2%
A more valuable pure tone test uses the sweep generator function. You can adjust the start
and end frequencies and the rate of “sweeping” from the lowest to the highest. A sweep
generator is very good for manual tests. For instance, you can check the low-end roll-off and
high-end roll-off of a sub-woofer system by connecting the sweep generator and listening. No
measure of linearity, but clear measure of system range. Because of multiple frequencies,
interferences effects cancel out.
A sweep generator into a system and a recorder on the output produces a record of system
performance which can produce many interpretations in off-line analysis. (Remember the
requirement for a microphone which is better than the system being tested and expected
interferences.) The off-line analysis can be very tedious and technical, but produces heavyduty results.
3 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
A quick, graphic way to evaluate overall system performance is to put in “pink noise” and read
the output on the spectrum analyzer. “Pink noise” is a random collection of all frequencies in
the music range, 20-15,000Hz with uniform amplitude. A very consistent signal goes in; any
non-linearities are measures of the system under test and immediately visible. The signal can
be recorded or a screen-capture can archive the results. (Remember the requirement for a
microphone that is better than the system being tested. Interferences are valid measures, so
multiple microphone locations are essential - even 12-in apart.)
Field Measurement dB(C) SPL
Test Location
13Ft
76
7M
77
81
77
6M
77
82
77
5M
77
84
78
4M
77
3M
78
89
78
2M
79
79
2X
1M
78
91
Ref
100
Base
Speaker
5M
78
80
81
4M 3M
13Ft
80
83
81
88
84
87
2M
1M
81
87
81
86
88
0
1M
Speaker
80
79
77
8M
83
81
8X
77
4X
85
82
78
2M
3M
4M 5M
13Ft
Distance Ratio
26Ft
Free-Field Sound Distribution, Pink Noise
20W Bass Guitar Amplifier(System)
On-Axis Data Analysis
Location Distance
Ratio
1M (3Ft)
Base
2M (6Ft)
2X
4M (13Ft)
4X
8M (26Ft)
8X
Measured
dB(SPL)
100
91
87
81
Theory
dB(SPL)
100
94
88
82
Difference
Percent
0 dB
3 dB
1 dB
1 dB
-3.3%
-1.1%
-1.2%
“Free field” means that no walls or objects interfere with the sound around the source and test
instrument. For valid, scientific testing, an anechoic chamber is used. This is a room with
absorbent walls. For the data collected above, the author’s front yard was used, with the
source about 8-ft above the ground, sitting on the edge of an SUV and the author walking
around a grid on the grass, holding the sound meter above his head. As with all field data, the
discrepancies from theory are as meaningful as the demonstration of the theory. Note,
however, the uniformity on-axis and off-axis of the low-cost commercial sound reinforcement
system.
The measurement of sound, dB(C) SPL, confuses many. For our purposes, it is a valid
numeric measure of sound intensity that closely matches human ear response. The human
ear/brain is logarithmic in response. That is, increases of 2x, 4x, 8x and 16x sound like equal
increments. In dB notation, these are 3dB, 6dB, 9dB and 12dB. Historically, 1dB is the
4 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
smallest change perceivable by the human ear. SPL means Sound Pressure Level, indicating
that the measurement started with a microphone. (C) means linearly-weighted. The Aweighting is limited to voice frequencies. The 100-ref means that the system under test was
adjusted to 100 dB(C) SPL at a point 1M in front of the speaker, then the measurements were
made. 100 dB SPL at 1M is a very standard speaker output for 1 watt of power being applied
to the terminals.
To provide an intuitive understanding of sound pressure level, the residential neighborhood
background noise for the free-field test was less than 50 dB (the lower limit of the meter used);
80 db is the noise level of a new Toyota Corolla traveling at 70 mph, 85 dB is the preferred
radio sound level in the Corolla at 70 mph, and 85dB is the OSHA limit for 8-hr exposure in
the workplace. 100db is generally considered to be the pain threshold (which was confirmed
during the tests).
The use of dB is widespread in sound system specifications. A 10-watt amplifier is 10dB
stronger than a 1-watt amplifier. A 100-watt amplifier is 20dB stronger than a 1-watt amplifier.
The 100-watt amplifier (operating at full rated gain) amplifies the 80 dB Toyota noise to 100 dB
(pain). This is why it is important to record in a very quiet studio or edit out all no-music or novoice portions.
dB
0.0
0.5
1.0
1.5
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
Power
Ratio
1.00
1.12
1.26
1.41
1.58
2.0
2.51
3.16
3.98
5.01
6.31
7.94
Voltage or
Current Ratio
1.00
1.08
1.12
1.19
1.26
1.41
1.58
1.78
2.00
2.24
2.51
2.82
dB
10
15
20
25
30
40
50
60
70
80
90
100
Power Voltage or
Ratio Current Ratio
10.0
3.2
31.6
5.6
100.0
10.0
316.0
18.0
1,000.0
32.0
10,000.0
100
10exp5
316.0
10exp6
1,000.0
10exp7
2,162.0
10exp8
10,000.0
10exp9
31,620.0
10exp10
10exp6
dB(power) = 10log(P2 / P1)
dB(voltage) = 20log(V2 / V1)
dB(current) = 20log(I2 / I1)
dB(SPL)
= 10log(SPL2 / SPL1)
Room Acoustics: .
There are two problems with room acoustics - reflections and resonances. Reflections are
largely undistorted echoes. In small rooms, the echoes come back while the word is being
spoken and cause no difficulty beyond limiting the gain available before feedback occurs. In a
large room, legitimate theater, music hall or cathedral, the reflections seriously interfere with
intelligibility.
5 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Resonances are non-uniform reflectances. There is always one frequency that reflects more
than the rest, sometimes very much more than the rest. This is the tone that first causes
feedback squeal.
Reflections can be measured with a pulse test that is not pursued here. Resonances are
easily recognized by spectrum analysis of system response with a pink noise source. It is
valuable to know the resonant peaks, because they can be diminished by signal processing in
the sound system, permitting larger overall gain and more pleasant listening. Some purists
object to response adjustment to eliminate feedback, but remember that the room acoustics
are providing very non-linear response before the adjustment.
Qualitatively, a space with many reflections and resonances is referred to as “live” and is
easily recognized by hard surfaces. A space with few reflections and resonances is referred
to as “dead” and is recognized by soft surfaces, carpeting, upholstered furniture, possibly
tapestries on the walls.
The same tests were performed in a conference room used for employee informational
presentations and web casts to remote offices. The goal was to see what effect the reflecting
walls have on sound levels.
6 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Field Measurement dB(C) SPL
26Ft
Test Location
13Ft
85
8M
85
7M
87
6M
87
5M
87
4M
91
3M
94
2M
92
Ref
100
84
1M
Speaker
5M
85
4M 3M
13Ft
87
92
2M
1M
85
89
0
1M
Speaker
87
84
84
2M
3M
4M 5M
13Ft
"Live" Conference Room, Pink Noise
20W Bass Guitar Amplifier(System)
On-Axis Data Analysis
Location Distance
Ratio
1M (3Ft)
Base
2M (6Ft)
2X
4M (13Ft)
4X
8M (26Ft)
8X
Measured
dB(SPL)
100
95
93
92
Theory
dB(SPL)
100
94
88
82
7 of 26
Difference
Percent
0 dB
1 dB
5 dB
10 dB
+1.1%
+5.7%
+12.2%
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Field Measurement dB(C) SPL
26Ft
Test Location
13Ft
8M
85
85
7M
87
6M
87
5M
87
4M
91
3M
94
2M
92
Ref
100
1M
Speaker
84
5M
85
4M 3M
13Ft
87
92
2M
1M
85
89
0
1M
Speaker
87
84
84
2M
3M
4M 5M
13Ft
"Live" Conference Room, Pink Noise
20W Bass Guitar Amplifier(System)
On-Axis Data Analysis
Location Distance
Ratio
1M (3Ft)
Base
2M (6Ft)
2X
4M (13Ft)
4X
8M (26Ft)
8X
Measured
dB(SPL)
100
92
91
85
Theory
dB(SPL)
100
94
88
82
8 of 26
Difference
Percent
0 dB
2 dB
3 dB
3 dB
-2.1%
+3.4%
+3.7%
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
1950's Sound Design Textbook Recommendations for Many Small Speakers in Ceiling
(Standard 30-ft x 30-ft classroom or conference room = 900 sq-ft.)
Required
Area
Area per
Speakers Mounting Watts
60-deg Speakers Mounting Required per
Ceiling
Ceiling
centers, c- Watts per 90-deg per 1000 centers, per 1000
spkr, sq- per 1000
Height
Example Height
c-c, ft * sq-ft*
1000 sq-ft* spkr sq-ft
c, ft. *
sq-ft
ft
-4 ft
.*
*
7.0
3.0
9.0
166.0
2.5
28.0
52.0
4.4
8.0
4.0
16.5
91.0
3.3
50.0
30.0
5.8
9.0
5.0
28.5
52.0
4.4
78.0
20.0
7.1
classroom,
office
10.0
6.0
38.5
39.0
5.1
40.0
114.0
13.0
8.8
13.0
11.0
7.0
50.0
30.0
5.8
153.0
10.0
10.0
12.0
8.0
65.5
23.0
6.6
200.0
8.0
11.2
13.0
9.0
85.0
18.0
7.5
256.0
6.0
12.9
14.0
10.0
106.0
14.0
8.5
314.0
5.0
14.1
cafeteria
15.0
11.0
128.0
11.0
9.5
50.0
380.0
4.0
15.8
25.0
16.0
12.0
154.0
10.0
10.0
452.0
4.0
15.8
17.0
13.0
176.0
9.0
10.5
530.0
3.0
18.3
18.0
14.0
205.0
7.0
12.0
620.0
3.0
18.3
19.0
15.0
235.0
6.0
12.9
710.0
2.0
22.4
20.0
16.0
271.0
6.0
12.9
810.0
2.0
22.4
21.0
17.0
308.0
5.0
14.1
910.0
2.0
22.4
party
center
22.0
18.0
345.0
4.0
15.8
100.0
1020.0
2.0
22.4
50.0
* = column added to original table
The spectrum of the pink noise source used is reproduced below. This display indicates very
good representation of frequencies between 100 Hz and 14 kHz, with a missing frequency at
5600 Hz and 24dB drop-off at 14 kHz.
9 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
As with all electronic and acoustic measurements, the limitations of the test sources and test
measures must be well below the values of the variables being measured. The system is
almost always limited by the microphone and speaker selected, then by the space acoustics.
Very good microphones are available in the $50-100 range and usually identified as “music,
wide range” or “voice, low-frequency roll-off”.
Excellent speakers are available in the $100-400 range and can be identified for application by
which elements are inside the enclosure. No tweeter, no high frequencies. General-purpose
speakers all have mid-range speakers for voice. A woofer of 10” or 12” is required for low
frequencies to about 150 Hz. A sub-woofer is required to go down to about 40 Hz. One major
manufacturer advertises small speakers with low frequency response. The speaker has very
poor low frequency response but must be used with the manufacturer’s equalization system
that overdrives the small woofer to get sound levels. This fact is hidden in a footnote on page
three of the specification. The quoted response is with the equalization. They will not reveal
the direct-connect response.
Room acoustics are complicated, with limited range of ameliorative actions available. The
reason for this course being limited to cafeterias, auditoriums and small churches is that these
spaces is to avoid addressing the acoustics of large, complicated spaces. Small spaces
respond well to a number of speakers in the ceiling or around the perimeter. A microphone
with built-in low end roll-off eliminates room noise.
A last limitation of acoustic tests is digital recording of acoustic data for off-line analysis, as
with the spectrum analyzer. Microsoft .WAV files do not use compression, and record with the
accuracy of your equipment. Generic .MP3 codecs have adjustments that limit the accuracy
of the recording in order to reduce file size. Such limitations are at the limit of the audio
equipment and certainly do affect the observed results. Using an identical test setup assures
that artifacts not produced by the system under test are at least identical, if not minimal.
10 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
As discussed previously, the purpose of pink noise is to eliminate acoustic or electronic effects
that operate on narrow frequency bands. By having all available frequency bands (as does
music) the system “average” performance is tested.
Sound System Components:
The Sound System Graphic in the Introduction portion of the course indicates the components
which must be considered in designing a sound system. In a small sound system, several
components may share a common enclosure and carry a slightly different name.
Two completely self-contained sound systems are not being discussed here - the bullhorn
and portable PA system. The bullhorn is a hand-held device favored by protesters and SWAT
teams in TV movies.
The portable PA system is a more gentile version that has a short cord on the microphone and
is favored by Lolly-the-Trolly tour guides.
Sound Lectern: .
A sound lectern almost exactly matches the bass guitar amplifier used for acoustic testing. It
contains a good mid-range speaker, a good amplifier, a good preamplifier and adds a good
microphone. The enclosure is superb, usually costing more than the components. Cost is
$1,000-4,000. Outboard speakers are available, permitting use in very large spaces. Note
that security is excellent, as the unit can be rolled to a closet and locked up.
There are aesthetic objections to sound lecterns along with budgetary complications. A sound
reinforcement system is considered a capital expenditure. A sound lectern is considered
moveable furniture. They have different approval chains and cannot be interchanged.
Microphone: .
Microphone selection evokes considerable emotional response. All but the cheapest $1 mikes
that come with tape recorders work well. Some work better in critical situations. A key
question, however is directional vs non-directional. Directional is better for a single speaker.
Non-directional is better for a group. There is no price or performance difference.
Microphone descriptions identify directionality, sound element, output level and frequency
response. There is considerable overlap among types and some imaginative advertising. The
author’s collection of experience and catalog reading is summarized as follows:
11 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Microphone Summary
Name
Directionality
Element
Level
Response
Price
Comment
Crystal, hi-z
Omni
Cardioid
Crystal.
Crystal
-50dB
-50dB
300-6000Hz
300-6000Hz
$1-30
$1-30
Bad
Electret #1
Omni
electret
-55dB
300-6000
$1-30
No amp, Bad
Dynamic
Omni
dynamic
100-10,000Hz
$20-200
Cardioid
dynamic
100-10,000Hz
$20-200
Super
cardioid
dynamic
-70 to –
60dB
-70 to –
60dB
-70 to –
60dB
100-10,000Hz
$20-200
Wide range of
features
Wide range of
features
Wide range of
features
Electret #2
Omni
Cardioid
Electret
Electret
-60 to –45
-60 to –45
50-15,000 Hz
50-15,000 Hz
$50-200
$50-200
Req batteries
Req batteries
Condenser
Omni
Cardioid
Diaphragm
Diaphragm
-60 to -50
-60 to -50
30-18,000Hz
30-18,000Hz
$60-1000
$60-1000
Req DC pwr
Req DC pwr
Condenser
Omni
Cardioid
Ribbon
Ribbon
-60 to -50
-60 to -50
30-18,000Hz
30-18,000Hz
$200-1000
$200-1000
Req DC pwr
Req DC pwr
Off-axis pickup, front of microphone
-80dB
90 deg
-70dB
-55dB
n
so
e
v
r
cu
in
a
g
et
he
s
ta
da
of
ily
m
a
f
180 deg
0 deg
Reference, rated input produced rated output
Off-axis pickup, behind microphone
270 deg
Microphone Cardioid Pickup Pattern
12 of 26
PDH Course E126
www.PDHcenter.com
xi
s
www.PDHonline.org
45
de
g
of
fa
90 deg
0 deg
180 deg
Reference, rated input produced rated output
270 deg
Microphone Super-Cardioid
Pickup Pattern
90 deg
0 deg
180 deg
Reference, rated input produced rated output
270 deg
Microphone Omnidirectional
(non-directional) Pickup Pattern
13 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
It would be valuable to distinguish microphones by survivability. Unfortunately, all rate from
low to extremely-fragile. Condenser microphones are sensitive to moisture, in addition to
vibration and impact. A spit-screen is desirable for condenser mikes for performers.
Microphone Connectors: .
Three types of microphone connectors must be considered, along with myriad variations.
They are XLR-3, TRS, and mini. Many years ago Cannon and Amphenol made XLR
connectors for broadcast microphones. Today, Switchcraft makes them for broadcast,
commercial sound and a wide range of control applications. Mouser Electronics also lists five
new manufacturers. The “3” indicates a source conductor, a return conductor and a shield
conductor, connected internally to the shell of the connector. When used for unbalanced
inputs, the return and shield are connected to the same pin. XLR-3 connectors are available
as male cable connectors, female cable connectors, super-cheap plastic cable connectors,
chassis male, chassis female, super-cheap molded chassis connectors and printed circuit
mount connectors. The plastic connectors do not have shielding at the terminations and often
lack the positive lock on the female piece that grips a groove in the male piece. The following
graphic shows the various forms.
TRS are the phone plugs, where phone is short for telephone. Three-conductor ¼-in phone
plugs were used on the telephone switchboards of the 1920’s. Per telephone lingo, they are
TRS plugs, for Tip, Ring and Shell. This exact plug is used for balanced microphone and
high-level audio connections and works extremely well. Tip is supply; ring is return and shell
is shield. For two conductor versions, the return and shield are both connected to the shell.
As with XLR connectors, a metal shell is shielded; a plastic shell passes noise and hum.
Many, many versions of TRS plugs, jacks and adapters are available. A few are listed on the
following graphics.
14 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
On a recent High School auditorium renovation project, the Electrical Contractor asked for
clarification of the XLR part numbers specified. The following graphic was added to the
construction drawing set.
15 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
CONNECTOR SCHEDULE
SYMBOL
DESCRIPTION
XLR-3
CABLE
FEMALE
SWITCHCRAFT A3F
XLR-3
CHASSIS
FEMALE
SWITCHCRAFT D3F
XLR-3
CABLE
MALE
SWITCHCRAFT A3M
XLR-5
CHASSIS
FEMALE
SWITCHCRAFT D5F
XLR-5
CABLE
MALE
SWITCHCRAFT A5M
1/4-IN TRS
CHASSIS
FEMALE
TIP
RING
SHELL
SIMILAR TO
1/4-IN TRS
CABLE
MALE
SWITCHCRAFT 12B
SWITCHCRAFT 238
Mini-plugs (3.5mm) are not traditional professional audio connectors. However, panel space
is becoming a limiting factor on miniature electronics, as wireless microphone transmitter
packs. Therefore, the tiny connectors are being used. They are similar in every way to TRS
connectors except that the size almost precludes manual soldering.
Microphone Cable:: .
This is the time we must discuss the non-issue of microphone impedance. In theory (and in
practice 50 years ago) there are two very different connections available between audio
sources and audio loads - high impedance and low impedance. High impedance was an
almost direct connection to the grid of a vacuum tube. Low impedance was connection to a
step-up transformer which was almost directly connected to the grid of a vacuum tube. High
impedance had better frequency response, but was limited to short distances before noise
pickup became noticeable.
Today, we use field effect transistors (FETs) for input stages on audio equipment. They
provide a high impedance input (low current flow), but work equally well on high impedance
and low impedance sources. Theory requires that low impedance sources be terminated with a transformer or resistor. In rare cases, this is performed. Usually, however, the FET
input simply bridges the input and works equally well on high impedance or low impedance
sources.
16 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
[This discussion applies only to millivolt signal circuits. Impedance matching is still critical on
power circuits and high frequency circuits, as video.]
Another non-issue is balanced vs unbalanced lines. Every circuit has a source conductor and
a return conductor. In a balanced circuit, they are treated as equals. In an unbalanced circuit,
the return is considered second rate and must share its current path with the non-current
carrying shield. Unbalanced circuits came with high impedance vacuum tube grid input
circuits. Balanced inputs are now provided by push-pull inputs of FET’s. If an unbalanced
source is connected to a balanced input, the input tries very hard to provide a constant zero
signal through the grounded side. Works well. Many mixers, for example, permit selection of
balanced or unbalanced input from the same connector.
The question outstanding is, what microphone cable to specify for a sound reinforcement
system? Typically, the answer is two - a super flexible, rough-service shielded cable at the
mike, and an economic, well shielded cable in the wall to the mixer.
Old and abused mike cables “crackle” when bent, or when the performer walks around the
stage. They must be replaced. The author recommends an ethylene propylene rubber (EPR)
jacket over braided tinned copper shield with two #16 conductors. EPR hold up well, but can
be cut or crushed. Braided shield is a little stiffer than spiral, but has better shielding and
heals at broken strands. #16 is much larger than required for the signal, but provides the only
strength in the cable. There are many opinions on mike cable and all work well initially.
In-wall mike cable can benefit from modern technology. Broadcast studios use #22 solid or
stranded conductors in an aluminized Mylar shield with a tinned copper drain wire. The overall
jacket is some plastic. (The jacket must be Teflon if the cable will be open-strung in air
handling spaces.) This works because the cable is installed and left forever. Cost is 10 cents
per foot. On the school auditorium job used as an example earlier, the author was concerned
that the wire puller would overstress the cable. For this reason, #16 conductors were
specified. Cost is 30 cents per foot. The cable schedule from that job is reproduced below.
CABLE SCHEDULE (300V JACKET)
CONNECTOR
CABLE
SIMILAR TO
MIKE
2/C#16 SH
WEST PENN 294
DATA
4-PR CAT 6
WEST PENN 54568
F-FIDEO
RG-11
WEST PENN 1100
AUDIO
2/C#12
WEST PENN 227
DMX
5/C#18 SH
WEST PENN 3280
BNC-VIDEO
RG-59
WEST PENN 840
17 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Preamplifier:
The preamplifier device has become invisible in modern sound reinforcement systems. Look
for the term in the description of the mixer. A microphone has an output signal of millivolts,
about –60dB. A mixer is optimized for 1 volt, 0dB or 300mV, -6dB. A preamplifier is needed
between the microphone and the mixer. Because of the high amplification of a very weak
signal, early preamps contributed most of the noise heard on the speakers and deserved close
attention. Modern FET preamps have very, very low noise and get little notice. If you must
buy a preamp, the cost is $60-150.
As part of incorporating the preamp into the mixer, it is often joined with digital to analog
conversion and the first stages of digital processing. This appears complicated. It is
complicated. But, in today’s market place, it actually reduces the price of the components and
improves features and delivered sound quality.
The importance of the preamplifier in system specification is the microphone input count.
Each microphone requires one pre-amp. Depending upon how stereo is handled, the mike
input count goes to 24 channels very rapidly. For whatever reason, mixer manufacturers like
many high level inputs and few microphone inputs. One solution is a matrix switcher, so that
many mikes can use few inputs, but this becomes complicated for the user and inappropriate
for schools and churches. Mixer channels are cheap, but the boards become huge.
Mixer:
Many years ago, a mixer channel was an audio potentiometer and an isolation resistor. Six
channels meant six pots and six resistors. This primitive mixer sells today for $60.
A powered mixer is a primitive mixer combined with a power amplifier. It may also contain
some DSP functions. A powered mixer costs $100 to $1000.
A modern mixer must accept microphone and high-level inputs. It should have a master gain.
It must have enough amplification to provide zero insertion loss to the system. It is desirable,
almost essential to have multiple outputs, for monitor speakers, headphones, recording,
assistive listening and digital processing for multiple speaker channels. It should have LED
level indicators, preferably on each input. This describes the high-end mixer, introduced
below.
High-end Mixer:
Live music performers usually bring along their own sound system and technicians. This
clarifies responsibility for proper operation and provides uniformity and confidence to the
performers. Their choice is to have many, many microphones on stage, some wireless, many,
many wires (snake) to a massive high-end mixer in the first few rows of audience space, dead
center, and multiple audio returns back to the stage. The return channels are for performer
monitor speakers, to the power amplifiers, to multi-channel recording gear and to the assistive
listening system (sic).
The high-end mixer has an input channel for each microphone and source, always with
individual level control and usually with individual equalization and digital signal processing
(effects) controls and LED level meter. The processed signal from each input channel can be
directed to one or more master channels. The master channels have master level controls. A
moderate high-end mixer has 24 mike inputs, four master channels and costs $400-$5,000.
Inputs are XLR, TRS, RCA and proprietary connectors. DC power is available for condenser
mikes. Outputs are XLR, TRS, RCA and proprietary connectors. The simplest proprietary
connectors are high-density DIN separable terminals. The stripped conductors of a mike
cable are inserted into the high-density connector, the screws are tightened and the connector
18 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
is inserted in the mating strip. The remote control unit using these has 12 channels in a 1-1/2in rack space.
It is hard to get a high-end mixer that will fit in a 19-in audio rack. Many are six feet long.
(Remote control):
Current commercial audio technology does not support remote control, that is, digital
commands over a single circuit to operate the mixer functions. The function can be
approached, however, using a matrix-switch/digital mixer from Mackie Industrial ($1000) or the
system processor matrix switch/digital signal processor from Shure ($3000). With either unit,
inputs can be selected and master volume controlled by an expendable remote box. This is
extremely desirable for shared public venues. The cost is not high for a commercial system
expected to perform reliably for 20 years.
Signal Processor:
The signal processor is a digital signal-processing (DSP) chip controlled by a microprocessor
and a very sophisticated program. Anything that has ever been done by electronics can be
done and extended by DSP technology. The programmers have gone wild. Three functions
will be described here - feedback elimination, time alignment and equalization. Feedback,
as discussed in the acoustics section, is the natural resonant frequency of the space
magnifying any source content at that frequency. If the sound reinforcement gain is high, the
single frequency, magnified by the resonance, will cause squeal. A common DSP function is
to have the unit find that frequency and reduce the gain for that frequency only. Multiple
frequencies can be identified and muted, all automatically.
Time alignment is the constructive response to interfering echoes among multiple speakers.
The performer and speakers at the front release a sound. It takes 100 milliseconds for that
sound to travel 100 ft into the audience. If an adjacent speaker reproduces the sound when
made, then the audience receives two versions, real-time and 100mS delayed. The human
ear can detect 30 mS and loses coherence of the message with a loud 100mS interference. If
separate amplifiers drive the down front speaker and the 100 ft away speaker, a DSP delay of
100mS can be inserted in the 100 ft away output so that it arrives with the real-time version
from the front. Sounds match. Lips way out of sync.
Equalization is the idea that a good technician can tune the sound system to improve the
listening experience. A stand-alone equalizer is a row of +/-10 dB sliders, usually at ½ octave
centers, for very narrow filters. Each pass band can be boosted or diminished.
A low-end DSP preamp will do all this and much more, for $60. A high-end DSP requires a
week of training before use and costs $3,000.
Assistive Listening:
The author’s current interpretation of the Americans with Disabilities Act (ADA) is that places
of assembly (50 persons or more) must be equipped with an assistive listening system to aid
hearing impaired persons. An attorney or architect will be able to provide more competent
advice. The requirement appears to be 4% of audience capacity, that is, 4 receivers per 100
seats or authorized standing spaces.
As indicated on the sound system graphic, an assistive listening system is made up of a
transmitter, which accepts a line level input, multiple receivers, and chargers for the receivers.
Multiple channels are available. The graphic indicates a radio system. Infra red systems are
also available, with some serious problems.
19 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
An assistive listening kit costs about $800.
Power Amplifier:
There are two distinct products in the commercial sound reinforcement field which are called,
“Power Amplifiers”. The first is better termed a “powered mixer”. It contains rotary level
controls, rudimentary equalization (tone controls), sometimes a feedback filter, and a power
amplifier in the 20-1000w range. Well-known brand names are Bogen and DuKane. Cost is
$100-800.
The second category is a raw power amplifier. It usually contains a master volume control and
speaker protection. Some contain a slot for one or more microprocessor/DSP module. Brand
names are Crown, Rane and Electro-Voice. Cost is in the $400 –4000 range.
A power amplifier for a small installation is a non-issue. As shown by the field data, 1 w rms
provides very acceptable sound levels in a space 30ft x 30ft. A 100 w amplifier far exceeds
the need of any small space and is an economic price point in all manufacturer’s price list.
Because of the excellent noise characteristics of solid-state devices, there is no downside to
over sizing the amplifier.
At this point, we begin a concealed discussion of two arcane subjects, speaker damage and
speaker impedance. A 200w amplifier is connected to four 200w speakers. If the amplifier
rating is conservative, there is still plenty of headroom in the speaker ratings to permit quality,
safe operation. Right? Maybe.
The 200w rating on the amplifier is real. Manufacturers use a number of standards, but they
all include a rated input signal and rated rms power output into rated load for an extended
period with no damage. (RMS power is what you read with a good meter. A cheap meter may
read peak or average power.) To get continuous 200w power, the amplifier is inherently
capable of producing 400w, possibly 800w briefly. Music tends to contain brief, high intensity
passages. Politicians and rock groups tend to favor driving a system to its maximum
capabilities (turn it up all the way). At maximum gain, the amplifier distorts and produces high
magnitude, high frequency signals which exceed the speaker high frequency capability. One
manufacturer stated that his 100w full-range speaker is only capable of 20w at the tweeter. It
is sized for the frequency distribution of music and voice.
For this reason, speaker protection is a valuable feature in the amplifier. Also, speakers
should be sized conservatively. Another alternative is to put in an oversized amplifier so that it
does not distort on peaks. It will be capable of really massive overload signals if turned all the
way up.
Speaker impedance contains many nasty details that are better avoided. There are two
conclusions - identical speakers will work well together and series-parallel connections are
not difficult. Sound from 20 to 15,000Hz contains a lot of frequencies. Speaker manufacturers
have 100 years of tweaking up the mechanical and electrical designs to get linear response. It
works. A given speaker (tweeter, mid-range and woofer) can be expected to work well. In
addition, four of them or eight of them or even 20 of them and they will work well together.
Different speakers, however, have different compromises to produce net linear sound output.
Do not expect them to work well on the same output circuit. The will interact badly, though
possibly not noticeably to the audience. The solution is multiple power amplifiers if different
speaker types must be used together. This is one of the reasons for multiple send channels
on a mixer - to drive multiple power amplifiers. If you were studying the discussion of time
alignment in the DSP discussion, you noticed that multiple amplifiers are required for that also.
On the other hand, cafeterias, auditoriums and small churches are usually well provided by a
reasonable number of identical speakers. No problem.
20 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
There are three concepts of connecting speakers to the amplifier that we would like to avoid,
but cannot - matching nominal impedance, series-parallel connections to match the direct
output and transformer connections for a transformer output. Modern solid-state amplifiers are
rated at 8 ohms output, with 4 ohms and 2 ohms in fine print. Most speakers are rated at 8
ohms. The fine print in the speaker specification states that actual impedance ranges from 2-8
ohms, with midrange about 5 ohms. As long as we have plenty of watts in the amplifier, this
doesn’t matter. Otherwise, get more watts in the amplifier.
Connect an 8 ohm speaker to the 8 ohm amplifier terminal. Simple, easy no problem.
How to connect two right side speakers to the single 8 ohm out put, though? Put them in
series and two eight ohm speakers look like 16 ohms. Put them in parallel and two 8 ohm
speakers look like 4 ohms. This is where the fine print on the amplifier specification saves the
day. 4 ohms is OK. Four speakers in parallel look like 2 ohms and 2 ohms is OK. In addition,
any wiring errors are covered by the output protection in the amplifier.
What about more than four speakers on an output? What about trying to connect 8 ohms load
to a rated 8 ohm source? That is where series-parallel connections come in. (Series-parallel
connections are discussed in the equipment manuals and at the manufacturers’ websites,
also.) The graphic below illustrates common series parallel connections.
21 of 26
www.PDHonline.org
8 ohms
equivalent
8 ohms
PDH Course E126
4 ohms
equivalent
8 ohms
8 ohms
www.PDHcenter.com
2 ohms
equivalent
8 ohms
8 ohms
8 ohms
Parallel Connections
8 ohms
8 ohms
8 ohms
Exceeds
Amplifier
Rating
Series Connections
8 ohms
equivalent
16 ohms
equivalent
8 ohms
8 ohms
8 ohms
equivalent
8 ohms
8 ohms
8 ohms
8 ohms
8 ohms
8 ohms
8 ohms
8 ohms
4 ohms
equivalent
8 ohms
8 ohms
Series-Parallel Connections
Good design requires separate conductors from the amplifier to each speaker, with seriesparallel connections behind the amplifier. Simplified installation practice suggests daisychaining speakers, with connections in the field.
Transformer output is a good idea for the County Fair, where speaker runs are in the 1000’s of
feet. It is an economical use of equipment and cable and fairly simple to design and install.
Unfortunately, frequency response is not music quality and it is just marginally legal.
Direct output matches speaker impedance. It works well because transistor characteristic
impedance also matches speaker impedance. Hence, direct output. Unfortunately, 100w at 8
ohms is 28 volts at 3.5 Amps. Three point five amps cause a voltage drop (signal loss) over
1000 ft of wire, even over fairly large, #12 wire. Twenty-eight volts cannot afford much voltage
drop.
The solution of 50 years ago is to boost the voltage with a transformer. The line current is
proportionately reduced. Voltage drop is reduced and the base voltage doesn’t care as much.
The downside is the losses and frequency limitations of the transformer and the requirement
for a transformer at the speaker, with losses and frequency limitations.
22 of 26
8 ohms
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
Historically, sound installations have been performed after the Electrical Inspector left the
building. Voltages were low and any hazards were considered minimal. This is changing.
The 2002 National Electric Code includes low-voltage wiring and requires the same rules
apply as with 120V wiring. (This is because of the proliferation of data wiring, but still applies
to sound wiring.) Inspectors today are not checking sound installations, but they have the
authority and it is an individual decision. Eight ohm wiring meets the low voltage rules.
Transformer operated speakers wiring usually does not.
Speaker Wiring:
Two heavy stranded copper conductors with a 300-volt jacket are needed for each speaker.
Specify a Teflon jacket if open wiring is run through an air handling ceiling space. The 300volt jacket provides physical protection and stops the Electrical Inspector from asking
questions. Stranded is much easier to work than solid conductors and survives abuse better.
Heavy means #18, #16, #14 or #12 AWG wire size. #12 is best and not much more
expensive. Cost is 10-40 cents per foot.
Beyond these minimal requirements, many people have many opinions. The preferences are
not supported by double blind A-B tests, but it is a good idea to help the Client spend his
money as he wishes and not argue.
Eight-ohm speaker wiring can be run anywhere you can drill a hole or drive a staple. Ideally,
the wiring is concealed during construction and conduit or surface raceway are used. Sound
installers like to drill holes or poke through gypsum walls with a screwdriver and hang the wire
in space. Most of the workmanship is at ceiling level or behind the speaker.
Speaker:
Regardless of the project, cafeteria, auditorium or small church, the speaker requirements are
the same. A speaker in front of and on each side of the presenter are needed. Preference is
to hang them near the ceiling. After that, several speakers along the sides or distributed over
the ceiling.
Details of speaker selection are discussed in the Power Amplifier section. Good speakers
cost $100-600 each and should last 10-20 years.
23 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
SAMPLE CAFETERIA SOUND DESIGN:
A sound system plan drawing is usually sparse. That is, the locations of devices are indicated
roughly, for field fit into the ceiling grid, etc. The devices themselves are indicated very
generally, without detailed layout or wiring. The components may be precisely described on
accompanying specifications, but the specifications are ignored by the supplier and
Construction Manager. This is because each supplier has a close relationship with one or a
few manufacturers and gets substantial discounts on only these items. Also, the sound
installer has substantial incentive to install reliable components - the meetings are more
expensive than the components and the installation. A meeting with a principal of the sound
firm will cost $100 per hour. The focus of such meetings must be on up selling the customer
and positioning the firm for future jobs. Discussion of 5-watt or 10-watt ceiling speakers
distracts from the sound person’s purpose. He will personally guarantee proper operation and
go to lengths to deliver. Note that any sophisticated input control will be provided by portable
equipment with a microphone level output.
Some installers are following the National Electrical Code and using drive rings and bridle
rings for cable support.
WM
SC
SM = Flush ceiling speaker, school PA system
SC
W
SM
SC
SC
ind
ow
WM = Wall microphone input
SC
SC
= Flush ceiling speaker, cafeteria PA system
PA
= Mixer / Amplifier rack
SM
WM
SC
St
SC
ep
Floor
SM
SM
SC
Platform WM
Storage PA
RP1-37
SC
SM
SC
SM
SC
To Tray
Return
SM
From
Serving
Line
0 2
5
Scale - feet
10
Cafeteria Sound System Plan
SAMPLE AUDITORIUM SOUND DESIGN:
Like the cafeteria sound design, the auditorium plan drawing is sparse. The dimensions of a
high school auditorium approach the echo interference limit, where sound alignment
processing is required. The base design for this auditorium used two primitive mixers for
additional channels and a 100w powered mixer, driving six 400w speakers in series-parallel.
An alternate deleted the powered mixer and added a dsp/matrix switch and two 400w power
24 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
amps. The reason for the oversized speakers in the base was so that the upgrade could be
made later if the alternate was not accepted.
SAMPLE SMALL CHURCH DESIGN:
Like the previous sound designs, the small church plan is sparse. The amplifier was a 100-w
powered mixer. The speakers were music-rated 100-w corner enclosures. The rack contained
the transmitter for the assistive listening, including the antenna. The speaker volume control
has not been discussed previously because it is a high-maintenance item, to be avoided in
school applications. It is important to get an L-pad rated for the expected power level. A 5w
L-pad worked well here.
25 of 26
www.PDHonline.org
PDH Course E126
Choir
Room
www.PDHcenter.com
Main
Entrance
L-Pad, Speaker Volume Control
Sp
Supplementary Seating, Accordian Curtain
L
Sp
Choir
Assistive
Listening
Charger
Cardioid Dynamic Microphones
on 2-in stands into Pulpit
Minister
Audio Rack
inside Pulpit
Congregation Seating
Altar
Speaker
Cardioid Dynamic
microphoneSeating
Congregation
on 6-in stand into Lectern
Organ
Sp
Small Church Sound System Plan
0 2
5
Scale - feet
COMMON DESIGN ERRORS:
1) It is critical that the relative locations be as follows:
Performer or musician
Speaker(s)
Audience.
If the speakers are behind the performer, available gain is severely limited, so that
soft-voice
persons will not be heard. Increased gain will produce feedback.
Speakers behind the audience produce an eerie discomfort in the audience.
2) More small speakers are preferable to few large speakers.
A ceiling-full of flush mid-range speakers on 15-ft centers will produce uniform sound.
Four or six speakers along a wall can be driven at low level. Listeners close by hear
well; persons further away do not get interference.
3) Talk to the installer. Typically, the school or Owner has a sound supplier who has
done most of the work for them over the years. Use his accumulated knowledge on
how to make the Owner happy. Talk to him about the scope of the project and
features needed. Give him a chance to review and mark-up the design. Most sound
designs are created by the supplier and copied into the project set by the Engineer or
Architect. Asking for review and comments is much more ethical. Keep “or equal” in
the manufacturers list.
26 of 26
10
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
4) Serious acoustics are complicated, not intuitive. A medium-sized Ohio city has a
performing arts hall. For twenty years, the built-in sound system produced pure garble
in the balcony. Since those are the cheap seats, no action was taken. It is amazing to
some that a rented venue can survive with bad sound.
Attention to acoustics is obvious. There will be adjustable reflective and absorptive
surfaces (doors). There will be nothing prominent about the sound system except that
the audience experiences vivid clarity of speech and music. A trained, experienced
acoustician costs $1000-2000 per day for opinions, not a design. Acceptance testing
of the sound panel adjustments requires a full audience for the tests and recordings
rather than expert opinion.
5) There is value to simplicity. A school sound system must be operable by the Principal.
That means one switch and plugging in the microphone. A high-end mixer is
wonderful for a trained operator, but one of the channels, or a bypass, must be
dedicated to ad-hoc assemblies.
COMMON INSTALLATION ERRORS:
1) Installers want to install the brand they get the largest discount on. This provides
competition and benefits to the project budget and to the Owner. It is essential for the
designer to check the specifications of the materials offered. Specify the microphone
pickup pattern; check the microphone pickup pattern. Specify the microphone
channels; count the microphone channels. Specify the rms power rating; check the
rms power rating. Specify the speaker wire size; check the speaker wire size. Do not
accept, “This will work just as well.” Identical specs are required to work just as well.
2) Talk to the installer about workmanship before he starts on the job. Emphasize the
need for aesthetics, especially in the cable exits from the wall or ceiling. The right way
is a junction box or back plate fitting, and a cover plate with a gland fitting for cable
strain relief. Installers like to poke through a big hole with a screwdriver and let the
cable hang. This is a violation of the National Electric Code and really ugly when you
focus on it.
3) Terminations are the part of all electrical installations most prone to error. Usually, a
bad audio termination is obvious when you turn up the volume and listen for hum.
Wiggle the wires at the audio rack. Deafening clicks and even a little hum are
indications of poor terminations. The deafening clicks are destructive to speakers, so
have an installer technician do the wiggling.
4) Remember that an acceptance test requires a full audience. The space acoustics
change considerably.
PLAN AND RISER REPRESENTATION:
A sound reinforcement system plan diagram is a building plan with locations of sound system
devices shown by generally recognized symbols. At present, there is no consensus standard
for the symbols. Use symbols of your choice, but include a legend. The plan drawing should
always be accompanied by a Symbol Legend.
27 of 26
www.PDHonline.org
PDH Course E126
www.PDHcenter.com
The building plan, called a “background,” is provided by the Architect, who
also determines the use group, construction, surface treatments and provides
final acceptance of speaker locations. The designer takes on substantial
liability if he makes assumptions written direction or sign-off of preliminaries
from the Architect.
Three sound system plans were presented previously.
A riser diagram is a graphic Bill of Materials for the job. It shows all essential
components, essential specifications and calls out count. It should match the
specifications exactly. Often no one reads the specifications. The introductory
sound system graphic for this course illustrates a sound system riser diagram,
but lacks the ratings call-outs.
28 of 26
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising