USAFA-TR-74-6
EPA - 550/9-74-008
THE DESIGN OF A LOW COST
SOUND LEVEL METER
LT COL JOHN D. GRIFFITHS
DEPARTMENT OF ELECTRICAL ENGINEERING
USAF ACADEMY, COLORADO 80840
APRIL 1974
FINAL REPORT
APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED
DEAN OF THE FACULTY
UNITED STATES AIR FORCE ACADEMY
COLORADO 80840
-------
Editorial Review by Lt Colonel W. A. Belford, Jr.
Department of English and Fine Arts
USAF Academy, Colorado 80840
Prepared for
OFFICE OF NOISE ABATEMENT AND CONTROL
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
This research report is presented as a competent treatment of
the subject, worthy of publication. The United States Air Force
Academy vouches for the quality of the research, without necessarily
endorsing the opinions and conclusions of the author.
This report has been cleared for open publication and/or public
release by the appropriate Office of Information in accordance with
AFR 190-17 and DODD 5230.9. There is no objection to unlimited
distribution of this report to the public at large, or by DDC to the
National Technical Information Service.
This research report has been reviewed and is approved for
publication.
PHILIP,/ ERDLE, Colonel, USAF
Vice itean of the Faculty
Additional copies of this document are available through the National
Technical Information Service, U. S. Department of Commerce, 5285 Port
Royal Road, Springfield, VA 22151.
-------
Unclassified
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
REPORT DOCUMENTATION PAGE
READ INSTRUCTIONS
BEFORE COMPLETING FORM
1 REPORT NUMBER
USAFA-TR-74-6
2 GOVT ACCESSION NO
3 RECIPIENT'S CATALOG NUMBER
4 TITLE (and Subtitle)
THE DESIGN OF A LOW COST SOUND LEVEL METER
5 TYPE OF REPORT & PERIOD COVERED
Final Report
1 Jan 1972 - 1 Jul 1974
6 PERFORMING ORG REPORT NUMBER
7 AUTHORfs.)
LtCol John D. Griffiths, USAF
a CONTRACT OR GRANT NUMBER^;
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Dept of Electrical Engineering (DFEE)
USAF Academy, Colorado 80840
10 PROGRAM ELEMENT. PROJECT, TASK
AREA ft WORK UNIT NUMBERS
CONTROLLING OFFICE NAME AND ADDRESS
Office of Noise Abatement and Control
Environmental Protection Agency
12 REPORT DATE
April 1974
13 NUMBER OF PAGES
Washington, D.C. 2Q46Q.
1* MONITORING AGENCY NAME ft AD
AGENCY NAME a ADDRESSf/f different tram Controlling Ottice)
Dean of the Faculty
USAF Academy, CO 80840
IS SECURITY CLASS (ol thle report)
Unclassified
15a OECLASSIFICATION/OOWNGRADING
SCHEDULE
16 DISTRIBUTION STATEMENT (of thle Report)
Unlimited.
17 DISTRIBUTION STATEMENT (ol the abstract entered In Block 30, II different Iron Report)
18 SUPPLEMENTARY NOTES
19 KEY WORDS (Continue on reverse eide it necessary and Iden'lfy by block number)
Sound Level Meter
Sound Level
Noise Level
20 ABSTRACT (Continue on reverse elde II neceaeary and Identity by block number)
Conventional sound level meters generally use precision ceramic micro-
phones and discrete solid state circuitry. This report describes a design
using an electret microphone and integrated circuit operational amplifiers.
The advantages of this design are low cost, ease of manufacture, stable gain,
and low power consumption. Construction details are included to permit local
manufacture. The complete instrument is capable of meeting the Type 2, General
Purpose Sound Level Meter, requirements of ANSI SI.4-1971, American National
Standard Specification for Sound Level Meters.
DD ,
FORM
JAN 73
1473
EDITION OF t NOV 65 IS OBSOLETE
Unclassified
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)
-------
SECURITY CLASSIFICATION OF THIS PACEfHTim Data Enterod)
SECURITY CLASSIFICATION OF THIS PAGEftWian Dala Entered)
-------
PREFACE
The author would like to express his appreciation for the support
given by Dr. Alvin F. Meyer and the staff of the Office of Noise Abatement
and Control, Environmental Protection Agency, Washington, D. C. Thanks
are also due to Dr. Henning E. vonGierke and the staff of the Bionics and
Biodynamics Division, Aeromedical Research Laboratory, Wright-Patterson
Air Force Base, Ohio, for their assistance in the free-field calibration
of the microphones and sound level meter. Finally, I wish to acknowledge
the American National Standards Institute for their permission to quote
from American National Standard ANSI SI.4-1971, Specification for Sound
Level Meters.
-------
INTRODUCTION
The general purpose sound level meter has been the standard field
instrument of the acoustician and noise control engineer for many years.
Recent widespread general interest in environmental noise pollution has
created a need for a low-cost instrument that can be used by the average
citizen for assessing environmental noise. Unfortunately, most low-cost
"noise meters" fail to be sufficiently accurate to be of use.
There exists an American National Standard (ANSI SI.4-1971)* which
specifies the accuracy required for an instrument to be acceptable as a
sound level meter. Indeed, there are three different levels of accuracy
specified, depending on the conditions of use of the instrument. Further,
since measurement of noise pollution involves questions of legal
definitions of acceptability and unacceptability of noise levels, only
those instruments which are legally acceptable are likely to be of use.
In general, legal requirements necessitate an instrument which satisfies
the provisions of ANSI SI.4-1971.
At the time this sound level meter development was conceived there
was no low-cost sound level meter available which met ANSI SI.4-1971
requirements. Recent developments in integrated circuit and microphone
*ANSI SI.4-1971, American National Standard Specification for
Soun4 Level Meters, is available from American National Standards
Institute, Inc., 1430 Broadway, New York, N.Y. 10018.
-------
technologies indicated that it might be possible to produce a low-cost
sound survey sound level meter. ** Such a sound survey meter was
constructed, using an A weighting network only. *** This original
instrument considerably exceeded specifications, and, in fact* met most
of the ANSI Si.4-1971'requirements for a Type 2, or general purpose,
instrument (with A weighting only, -of course).
Because of the simplicity of the original instrument and the ease
with which it met design specifications, it was decided to see if the
design could be extended to a general purpose (Type 2) instrument with
A, JB, and
-------
This technical report is in several sections. First, the design
philosophy and its execution are covered in some detail. This section is
primarily intended for those who may wish to produce, modify, or improve
on the design. Second, a comparison of the performance of the instrument
with ANSI SI.4-1971 specifications is given. Finally, detailed
constructional information for the home builder or model shop is given.
GENERAL DESIGN
The sound level meter is basically a rather simple device. It
consists of a microphone to convert sound level to voltage levels, an
audio amplifier to amplify the voltage levels, some frequency compensation
networks, a calibrated attenuator to extend the range, and an indicating
detector and meter. To satisfy the requirements of ANSI SI.4-1971,
however, requires a fair degree of precision and reproducibility in the
actual sound measurement, which in turn necessitates close tolerances in
the internal components and design in the sound level meter.
The most critical component of a sound level meter is the microphone.
Normal high-fidelity microphones, and even those used commercially for
high quality sound recording, are generally inadequate for sound level
meter use. Type 2 and 3 sound level meters require microphone response
only over the range of 20 Hz to 10 kHz, but over most of that range the
microphone response must be uniform. In addition, the microphone must be
omnidirectional, even at the higher frequencies, which in turn physically
requires a small microphone and case. Finally, the sensitivity of the
microphone should be stable, both in time and under varying conditions.
-------
Sound level meter microphones are usually of the piezoelectric ceramic
variety, generally costing about $70 up for units that will meet Type 2
specifications. Capacitor microphones are capable of more linear and wider
frequency response than other types, but they have the disadvantages of
high unit cost, sensitivity to moisture, and the necessity for an internal
preamplifier and a power supply for the preamplifier and the microphone
polarizing voltage.
Recently the electret microphone has been developed, which is a
capacitor microphone with electrical charge permanently impregnated in
the dielectric between the capacitor plates, obviating the need for a
separate polarizing voltage power supply. Further, current manufacturing
technology uses semiconductor techniques to fabricate the preamplifier
as an integral part of the electret microphone. The cost of these units
can be quite low, and their performance rivals that of the traditional
capacitor microphone.
A number of foreign and domestic electret microphones were tested;
there was a low-cost domestic model which met the requirements for use
in a sound level meter. This model, the Thermo-Electron Corp. Model 814,
was selected and adapted physically so that the resulting unit would fit
in a standard microphone coupler or calibrator for subsequent sound level
meter calibration. The Model 814 units which were measured met Type 2
specifications. The roll-off of the frequency response at low frequencies
of the Model 814 also automatically satisfies one of the provisions of
ANSI SI.4-1971.
-------
The electrical design of a conventional sound level meter is
straight-forward, consisting of a high gain audio amplifier, an
attenuator calibrated in 10 dB steps, a number of frequency compensation
networks (A, ]J, £ filters), a detector, a meter calibrated in 1 dB steps
over a 15 dB range, and an overall gain control for setting overall
calibration. Current conventional practice has been to use discrete
transistors as the active gain elements.
Since the sound level meter must maintain calibration over a wide
range of conditions, this fact implies that the amplifier gain is stable
and constant under the same range of conditions. This stability is
usually accomplished through the use of inverse feedback, which in turn
requires a larger number of amplifier stages to achieve the same overall
gain, thereby raising the cost.
The use of integrated circuits, specifically operational amplifiers,
allows the use of very high gain, and therefore large amounts of inverse
feedback, at very low cost. When operational amplifiers (op amps) are
used and the amount of inverse feedback is sufficiently great, the
overall gain of the total circuit is essentially independent of the
actual gain of the operational amplifier.
Specifically, in the circuit below, K is the gain of the op amp, R
is the feedback resistor, and R. is the input resistor; the input
impedance of the op amp is assumed to be very high compared to R.. Then
the gain of the overall circuit is approximately -Rf/R., as long as
Rf/R.«K. : A^S 1
. i fS. j
., . r /\ i .x i
V in ^ _ Y
i
01
i v*j ' ru
i L^S uv
r ^
r\r\ arnr*
out
-------
There are a number of advantages in the use of op amps in a
circuit such as a sound level meter:
1. Cost: Each op amp has a gain, without feedback, of greater
than 10 and a cost of less than $1.00. Discrete components and wiring
costs to accomplish the same specifications would cost at least an order
of magnitude more.
2. Simplicity: A single component with two external resistors
replaces a large number of discrete transistors and resistors.
3. Stability: The high gain permits a large amount of inverse
feedback which results in stable performance and in insensitivity of
gain to external variations.
4. Power Consumption: The op amp operates at a low power level,
giving long battery life.
5. High input and low output impedances: these characteristics
allow multiple op amps to be cascaded and interfaced easily with other
circuits, such as frequency compensation networks, detectors, and
indicating instruments.
The use of op amps in place of conventional circuitry entails one
disadvantage, the necessity for a dual power supply. There are
techniques by which an op amp can be used with a single power supply,
but some of the margin of stability is sacrificed, and that approach is
not embodied here.
In order to change the gain of an op amp circuit, one needs to
merely change either Rf or R . In order to keep the overall input
impedance relatively constant (approximately equal to R ), Rf is changed
-------
to accomplish the variable attenuation. ANSI SI.4-1971 suggests 10 dB
steps in the calibrated attenuator, so R, is changed by multiples of
/To". (10 dB - 20 log /lb~.)
ANSI SI.4-1971 requires that the decade attenuator on other ranges
be accurate within 0.5 dB of the 80 dB SPL setting, for a Type 2
instrument. The design here uses two decade attenuators, alternately
changing the gain on one of the two variable op amp states by /10.
The feedback resistors are 2% components, except on the 80 dB SPL range
where they are 1% components. If we assume that the 1% resistors are
-1% in error and the 2% resistors are +2% in error, then the maximum
possible attenuator error, relative to the 80 dB SPL range is:
20 log1Qr"|?gj = 0.52 dB maximum error.
This error is slightly in excess of the maximum permissible error.
However, the probability of the error exceeding 0.5 dB is extremely small,
on the order of 0.1%. Further, the difference between 0.5 and 0.52 dB
is not measurable on the indicating meter.
[A Simple Type 3 instrument with a single attenuator would need only
5% feedback resistors, as the attenuator tolerance is 1 dB for Type 3
instruments.
20 log-Q( ' 1 = .87 dB maximum error].
The input resistor, R, , is a 5% component, for it affects only the
overall gain, not the relative gain between various attenuator positions.
The overall gain is adjusted by making the first Rf (R.,) variable and
adjusting the gain of the sound level meter so that its indicated sound
level agrees with a standard sound level in a calibrator. A difference
in microphone sensitivity can also be compensated by varying R_.
-------
Three op amps are used, the first one of which is operated at a
constant gain of approximately 20-30 dB. The dynamic range of this stage
is approximately 90 dB, thereby limiting the instrument to a maximum
range of 75 dB SPL. In this case the range has been selected to be
45 dB to 120 dB SPL.
The range of the instrument could have been widened by varying the
first op amp gain with the attenuator as well. However, that would
require a three-section attenuator, increasing cost, and also decreasing
the tolerance permissible on the attenuator resistors. Were 1% components
used throughout, / \3
20 log1Q( ' j = 0.52 dB maximum error.
The input network to the first op amp also contains the £ network
low frequency compensation. This network consists of the source
impedance of the electret microphone internal amplifier, the input
resistor R2 to the first op amp, and the series capacitor C2. This
network attenuates the low frequencies at a rate of 6 dB for each
halving of frequency (6 dB/octave), beginning at 31.6 Hz, where the
attenuation is 3 dB. (The £ network compensation is also part of the A
and IJ compensation, as required by ANSI SI.4-1971.) The variable
feedback resistor R_ changes the gain of the first stage to permit overall
calibration of the instrument with reference to an external acoustic
calibrator.
The gain of the second stage is controlled by the decade attenuator.
The decade attenuator adjusts the gain of the second stage between 0 dB
and 30 dB in 10 dB steps.
At the output of the second stage is a switch, selecting _A, 15, or
£ frequency compensation. In the case of £ compensation, the networks
are always in the circuit and are located elsewhere in the instrument, so
10
-------
the switch selects only a resistor R9 which attenuates the gain by
approximately 1.8-1.9 dB. ANSI SI.4-1971 requires this attenuation for
J3 and C_ compensation relative to A compensation. [This relative
attenuation refers to high frequency gain, above 5 kHz.] C4 has no
appreciable effect on the frequency response above 20 Hz.
When the switch selects JJ compensation, this resistor is retained,
and capacitor C7 is switched in, forming a 6 dB/octave low frequency
attenuation network. The gain of this network, which includes R12, is
down 3 dB at 158.5 Hz.
Finally, when A compensation is selected, the attenuation resistor
R9 and the ]J network R9-C7 are switched out, and a pair of cascaded
6 dB/octave resistor-capacitor low frequency compensation networks,
R10-C5 and R11-C6, are connected. These networks interact, so that the
instrument response is down 3 dB at 738 Hz, is attenuated 6 dB/octave
approximately between 738 Hz and 107.7 Hz, and is attenuated 12 dB/octave
approximately below 107.7 Hz. The input resistance R12 of the following
op amp stage also forms part of these compensation networks. The gain
of the third and final op amp stage is also controlled by the decade
attenuator, and varies from 0 dB to 30 dB in 10 dB steps.
Following the third op amp is the high frequency C;network, R17-C9,
which attenuates at 6 dB/octave above 7943 Hz and is down 3 dB at that
point. In addition, the high frequency roll-off of the 814 microphone
helps satisfy the requirement that above 20 kHz the attenuation approaches
12 dB/octave.
The A, JJ., and £ networks are formed of resistor-capacitor pairs,
as described in ANSI SI.4-1971. Capacitors with 10% tolerance are
11
-------
available, and have been selected for use here. The frequency response
tolerances allowed by ANSI Si.4-1971, given later in the report, are
such that the attenuation due to a single R-C pair .may b.e in-error by + 1 dB
for a Type 2 instrument, this error being in addition to the basic + 1, dB
tolerance of the instrument. An analysis of an R-C network-'reveals that
the total error of the two components is approximately 12% for a 1 dB
error in response. Therefore, 2% resistors as well as 10% capacitors
satisfy this requirement.
If commonly available capacitors with wider tolerances, such as
-10 + 100%, are used, then the pertinent resistors (R9, RIO, Rll, R17)
will have to be selected so that the instrument- has the proper electrical
response as in Fig. 1. In this case the resistors can be selected from a
common bench stock, and will, in general, have lower nominal values than
those specified in Fig. 14. Methods of verifying the electrical response
are described elsewhere in the report.
At the output of the high frequency C_ network following the third
op amp, the signal branches to the output jack to drive an external
analyzer, and also goes to the detector, and thence to the indicating
meter.
Germanium diodes are used in the detector, since silicon diodes
cause upper meter scale expansion and lower meter scale compression due
to their transfer characteristics. That is, if silicon diodes are used,
the upper 5 dB range occupies nearly the entire meter scale, and the lower
10 dB range is compressed into a small scale increment.
A SPST switch selects a large load capacitor for SLOW response.
12
-------
te.
u
_ LU
C. i
u
31
or
2
£O
5 67897
100
507891
IOOO
FREQUENCY IN CYCLES PER SECOND
5 0
7 a » i
10000
2
2OOOO
Figure 1
Sound Level Meter Random-Incidence Relative Response Level as a Function of Frequency for Various Weightings.
-------
The function switch has a position where the two batteries are
placed in series with the meter and an attenuating resistor in order to
indicate battery condition.
Calibration is accomplished by use of an external standard
calibrator.
14
-------
VALIDATION AND CALIBRATION AS A TYPE 2
GENERAL PURPOSE SOUND LEVEL METER
In general, this section follows the sequence in ANSI SI.4-1971 and
addresses Itself to those requirements. Only those requirements for Type 2
instruments will be given, and are to be assumed as having been extracted
from ANSI SI.4-1971 without further reference to that document.
Tolerances
The required frequency response is given below and in Fig. 1 for the
various weighting networks.
Sound Level Meter Random-Incidence Relative Response Level
As a Function of Frequency for Various Weightings
A Weighting B Weighting C Weighting
Frequency Relative Response Relative Response Relative Response
Hz dB dB dB
10
12.5
16
20
25
31.5
40
50
63
80
100
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10000
12500
16000
20000
-70.4
-63.4
-56.7
-50.5
-44.7
-39.4
-34.6
-30.2
-26.2
-22.5
-19.1
-16.1
-13.4
-10.9
- 8.6
- 6.6
- 4.8
- 3.2
- 1.9
- 0.8
0
+ 0.6
+ 1.0
+ 1.2
+ 1.3
+ 1.2
+ 1.0
+ 0.5
- 0.1
- 1.1
- 2.5
- 4.3
- 6.6
- 9.3
-38.2
-33.2
-28.5
-24.2
-20.4
-17.1
-14.2
-11.6
- 9.3
- 7.4
- 5.6
- 4.2
- 3.0
- 2.0
- 1.3
- 0.8
- 0.5
- 0.3
- 0.1
0
0
0
0
- 0.1
- 0.2
- 0.4
- 0.7
- 1.2
- 1.9
- 2.9
- 4.3
- 6.1
- 8.4
-11.1
-14.3
-11.2
- 8.5
- 6.2
- 4.4
- 3.0
- 2.0
- 1.3
- 0.8
- 0.5
- 0.3
- 0.2
- 0.1
0
0
0
0
0
0
0
0
0
- 0.1
- 0.2
- 0.3
- 0.5
- 0.8
- 1.3
- 2.0
- 3.0
- 4.4
- 6.2
- 8.5
-11.2
15
-------
The total allowable tolerance limits, including microphone random
response, are as follows:
Total Tolerance Limits for Sound at Random Incidence
for Type 2 Sound Level Meter
Frequency
Hz
20
25
31.5
40
50
63
80
100
125
160
200
250
315
400
500
630
800
1000
1250
1600
2000
2500
3150
4000
5000
6300
8000
10000
A Weighting
dB
+5.0,-°°
+4.0, -4. 5
+3. 5, -4.0
+3.0, -3. 5
+3.0
+3.0
+3.0
+2.5
+2.5
+2.5
+2.5
+2.5
+2.0
+2.0
+2.0
+2.0
+1.5
+2.0
+2.0
+2.5
+3.0
+4.0, -3. 5
+5.0, -4.0
+5. 5, -4. 5
+6.0, -5.0
+6. 5, -5. 5
+6. 5, -6. 5
+6 . 5 , -°°
B Weighting
dB
+4.0,-°°
+3.0, -3. 5
+2. 5, -3.0
+2.0, -2. 5
+2.0
+2.0
+2.0
+2.0
+2.0
+1.5
+1.5
+1.5
+1.5
+1.5
+1.5
+1.5
+1.5
+2.0
+2.0
+2.5
+3.0
+4.0, -3. 5
+5.0, -4.0
+5. 5, -4. 5
+6.0, -5.0
+6. 5, -5. 5
+6. 5, -6. 5
+6 . 5 , -°°
C Weighting
dB
+3.0,-~
+2.o!-2.5
+1.5, -2.0
+1.0, -1.5
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.0
+1.5
+1.5
+2.0
+2.5
+3. 5, -3.0
+4. 5, -3. 5
+5.0, -4.0
+5. 5, -4. 5
+6.0, -5.0
+6.0, -6.0
+6.0,-°°
The measured electrical response of the USAFA/EPA sound level meter
is given in Fig. 2. To this response must be added the random incidence
response of the microphone, given in Fig. 3. The sum of Figs. 2 and 3 meet
the requirements of Fig. 1 within the above tolerance limits.
16
-------
att
H:
1
—
—
IT"
1
m
Q-IO
z
iii.
aJf
fi
n
CM
I
LI
§-
20
j- L*J
S3
IT £
o.l
O
D.
S-30
£
-40
-50
1
fit
T 4
IOO
587881 2
1000
FREQUENCY IN CYCLES PER SECOND
' * 8,oooo
20000
Figure 2
Electrical Relative Response Level of USAFA/EPA Sound Level Meter.
-------
20
4 6 e 7 a 8 i
too
.66788?
1OOO
FREQUENCY IN CYCLES PER SECOND
S 67891
1OOOO
O) _i
^ LlJ
CM >
20000
Figure 3
Random-Incidence Relative Response Level of Thermo-Electron Corp Model 814 Microphone.
-------
The attenuator must have the following tolerance for its settings,
with respect to the 80 dB setting:
Within + 0.5 dB 63 to 2000 Hz
Within + 1.0 dB 22.4 to 11,200 Hz.
The attenuator satisfies the requirement. (See preceding section for
analysis of attenuator design and component tolerance).
Internal Noise and Distortion
On all attenuator settings, except the three most sensitive, the
internal noise of the instrument must be at least 40 dB below the maximum
scale reading, with an acoustically shielded microphone in place.
When the shielded microphone is replaced by equivalent electrical
impedance, the noise level must be at least 5 dB below the lowest sound level
the instrument is intended to measure.
When the output meter is replaced by its equivalent impedance, the sine
wave response from 22.4 to 11,200 Hz must be linear within 1 dB up to 10 dB
above maximum scale reading.
Omnidirectional Responses and Tolerances
The random incidence response of the 814 microphone, as mounted on the
sound level meter, is given in Fig. 3. With respect to the random incidence
response, the free-field response at various angles may not deviate more than
the following tolerances:
19
-------
Maximum Allowable Deviation of Free-Field Relative Response
Level with Respect to Random-Incidence Relative Response Level
When the Angle of Incidence is varied from 45° to 90° from the
Axis About Which the Response is Most Nearly Cylindrically
Symmetrical.
These allowances are added arithmetically to the respective
tolerance limits previously given.
Frequency Type 2
Hz dB
31.5 to 2000 +2
2000 to 4000 +2.5
4000 to 5000 +3
5000 to 6300 +3.5
6300 to 8000 +4.5
Maximum Allowable Deviation of Free-Field Relative Response
Level for Sounds Arriving at Any Angle of Incidence with
Respect to Random-Incidence Relative Response Level for Any
Angle of Incidence.
These allowances are added arithmetically to the respective
tolerance limits previously given.
Frequency Type 2
Hz dB
31.5 to 2000 +3
2000 to 4000 +3,-4
4000 to 5000 +4,-6
5000 to 6300 +5,-8
6300 to 8000 +7,-9
The Model 814 microphone on the USAFA/EFA sound level meter satisfied these
requirements.
Should one desire to make free-field measurements, the most accurate
response is obtained when the angle between the sound source and the
sound level meter is approximately 75° to 90°, for this sound level meter.
O O
• Sound Source ' Sound Source
75° . -
90e
20
-------
The correction for normal incidence (0° - microphone pointing at sound
source) is given in Fig. 4.
Indicating Instrument Characteristics
The scale is calibrated in 1 dB steps over a 15 dB range, from
+10 dB above the attenuator setting to -5 dB below. The scale is
accurate within 40.2 dB full scale to within ±0.65 dB at the lower limit.
The indicating instrument is of the square law type and satisfies
within 0.5 dB the rule of combination measurement procedure as specified
in Appendix B, ANSI SI.4-1971.
When the instrument is in "FAST" response, an applied 0.2 second
pulse of 1000 Hz sinusoid results in a reading 0 to 4 dB less than that
for a continuous signal of the same frequency and amplitude. A suddenly
applied and maintained signal between 125 and 8000 Hz results in less
than 1.1 dB overshoot. The steady reading is 4 dB less than full scale.
When the instrument is in "SLOW" response, an applied 0.5 second
pulse of 1000 Hz sinusoid results in a reading of 2 to 6 dB less than
that for a continuous signal of the same frequency and amplitude. A
suddenly applied and maintained signal between 63 and 8000 Hz results in
less than 1.6 dB overshoot. The steady reading is 4 dB less than full
scale.
At any frequency between 31.5 and 8000 Hz, "SLOW" and "FAST" readings
differ by less than 0.1 dB for a steady input.
There is a battery indicator to indicate adequate battery voltage.
The instrument is calibrated by means of a standard acoustic calibrator.
If a resistive load impedance of 10,000 ohms or greater is connected to
the output, the reading is affected by less than 0.5 dB over the frequency
range 22.4 to 11,200 Hz. The nominal output voltage is 0.5 volts rms
across 10,000 ohms for a full scale reading.
21
-------
at
K UJ
2S
Ox
[> (J O
CM u Q
^CC y
ffl_I I
J. U h
N > °
l"i
ty 5
5 i z
^5
U oe
U>
3
5 4
FREQUENCY
IN
e 7 e e ;
IOOO
CYCLES PER SECOND
2
6878
20000
Figure 4
Correction for Free-field Perpendicular-Incidence Response for TEC 814 Microphone, to be added to Random-
Perpendicular- Incidence
Incidence Response.
Microphone
-------
Specific tests were not made to determine compliance with the temperature,
humidity, vibration, airborne noise, and magnetic and electrostatic fields
provisions of ANSI SI.4-1971. Whether or not a completed instrument meets
these provisions would also be a function of the instrument layout and
method of manufacture and is not inherently part of the electrical and
acoustical design. The microphone and operational amplifiers as separate
devices easily meet the specification requirements.
Methods of Verification
The electrical response of the sound level meter was determined as follows:
G-R 913C
Beat Frequency
Oscillator
Attenu-
ator
X
Equivalent
Microphone
Impedance
USAFA/EPA
Sound Level
Meter
-^
G-R 1521-A
Graphic Level
Recorder
Mechanical coupling
The microphone calibration between 20 Hz and 1000 Hz was determined as follows:
G-R
913C
Oscillator
s
S
G-R 1559
Microphone
Reciprocity
Calibrator
Acoustic^
s
Coupling
USAFA/EPA SLM
(with microphone) .
C_ network out
>^.
s
G-R 1521-A
Graphic Level
Recorder
_. _ _ _i
Above 1000 Hz the microphone, mounted on the sound level meter, was free field
calibrated in an anechoic chamber at the angles of incidence suggested in
Appendix A ANSI SI.4-1971. The calibration was by comparison with the response
of a B & K Model 4145 secondary standard microphone to a known acoustic source.
The calibration was at the 1/3 octave frequencies specified in ANSI SI.4-1971
23
-------
between 1 kHz and 20 kHz. As a check, the response of the sound level
meter was compared to that of a G-R Precision Sound Level Meter with a
Model 1560-P7 Precision Microphone which had been calibrated. The
1559-B Microphone Recriprocity Calibrator was used as a sound source
between 1 kHz and 8 kHz.
The "FAST" and "SLOW" response tests used a multivibrator with
different "on" and "off" times to trigger a 1 kHz oscillator as a signal
source. The remainder of the calibration verification used normal
standard laboratory procedures.
24
-------
CONSTRUCTION DETAILS
The basic instrument detailed here uses the Thermo-Electron Model
814 microphone and is housed in a 2 1/8 x 3 x 5 1/4 inch minibox. The
unit which was tested for omnidirectional response and for which the
earlier curves were derived was housed in a 3 x 4 x 5 inch minibox.
Since the smaller 2 1/8 x 3 x 5 1/4 box results in less sound diffraction,
improved high frequency response should result, and the resulting random
incidence response should be somewhere between those of Fig. 3 and Fig. 4.
The earlier model is shown in Fig. 5, and the current model in Fig. 6.
If it is desired to duplicate the earlier model (to achieve the
response of Fig. 4), the microphone mount cover, Fig. 7, will have to be
enlarged so that the base is 3" x 4". The heights of both the microphone
mount and the cover should be increased by 3/8".
The details of the 814 microphone mount are shown in Fig. 9. The
microphone is cemented in the hole so that the face plate of the
microphone is flush with the top of the mount. RG-174/U coaxial cable
or tonearm phono cable is used for the signal and ground connections, and
26 gauge wire is used for the power supply connection. The wires pass
through a hole in the top of the minibox. The microphone mount is
attached to the minibox via three 6-32 screws.
25
-------
- ••
I 0 2 4 6 8
i I i I i
UNITED STATES AIR FORCE ACADEMY
ENVIRONMENTAL PROTECTION AGENCY
0 SOUND LEVEL
METER A
eo M-4
Figure 5
Sound Level Meter, 3x4x5 inch case
26
-------
SLOW
FAST
N
wr
IT ED STATES AIR FORCE ACADE
ENVIRONMENTAL PROTECTION AGENCY
SOUND LEVEL METER
Figure 6
Sound Level Meter, 21/8x3x5 1/4 inch case
27
-------
1/8
2-1/4
2-1/8
MATERIAL-BALSA WOOD OR PLASTIC
FIGURE 7
MICROPHONE MOUNT COVER
28
-------
'
II
2-3/4
1/8"
*
1
1
n
1 1
1
I
I
1
1
1
I
1
i
1
1
1
1
i
i
i
i
t
-*•) H-5/32"
ZJ 1
^ |
_^ta
' k- 1-3/4" DIA.
.^ r» oir»" r»i A
.^ r» CKC" m A
DRILL AND TAP 6-32.3HOLES
ON 1-7/16" CIRCLE
MATERIAL-ALUMINUM
FIGURE 8
MOUNT FOR MODEL 814
MICROPHONE
29
-------
O
MTED STATES AIR FORCE A CADE I
ENVIRONMENTAL PROTECTION AGEN<
SOUND LEVEL METER
Figure 9
Microphone Mount, Close-Up View
30
-------
Shown in Fig. 10 is the interior of the minibox, with the meter at
the top and the two batteries at the bottom. The switches are on the
other side of the printed circuit board.
A full size diagram of the printed circuit board, foil side, is
given in Fig. 11, and may be used to photo-etch a duplicate board via
normal techniques.
The parts layout and interconnections to the pc board are given in
Fig. 12. This is, of course, the opposite (non-foil) side of the board.
The original board used foil on both sides, but in the interests of cost,
simplicity, and duplication it was re-designed into a one-sided board.
This procedure necessitates a number of jumper connections, however, also
shown in Fig. 12.
The full schematic diagram is given in Fig. 13, and the parts list
in Fig. 14. All parts are mounted on the pc board except for the
microphone, switches, output jack, batteries, meter, and R19, which is
mounted directly on S2.
Templates for drilling the front, top, and bottom panels are given
in Fig. 15 and Fig. 16.
31
-------
Figure 10
Interior of Sound Level Meter
32
-------
Figure 11
Printed Circuit Board, Foil Side
33
-------
SIB- no
SIB-90,100
SIB-70.8O
SIB -5O.60
SIB-POLE
SIA-IOO.MO
SIA-80.9O
SIA-6O.7O
SIA-50
SIA -POLE
0
10®
9e
8 o
7 o
6 o
B
5®
4o
3 °
2 o
o
A
r— r
u
e
•
4
*.
IRJ'S
u— — -- »_
IRIS
IRI4
IRI3
[R8
cm
1 R6
1 R5
^
•»- ^^^
R3
J
|
|
!
U
i
J-
^
\
^
—
'
• —
»•
• — — -
c
[RjT
fRTcT
C 1
-
3
n
n
c!
!
-•
~[
|
f
J
#•
2
.
IC3
C2
— .
!
1
C 7
^
i
I
l_
C
c
V.
c
(-
i
i
i
E
i
]
4
*
i
i
t
\
*}\
i
J
h
» _
r«i
CjB
r
L
—«
Cl
C2
-t<-
C 1
c
c
C
<
1
>
«
(
R
"F
^
. ^
•
i
*^»
c
~2
D
7
7
»
_ -
5
i— — .
a
Jl
"
r
1
—
•
*
1
1
CIO
•
;
* !
CI2!
*
,
i
i
0 '
I
._ !
)
1
o,2
i
i
«IO
09
•
^8
3 f '
fi.5 ,
1
04 !
°3
1
0^
,
91
,
GND-BATT
S2D-A.B.C
IS3-POLE
i
j S2C-AtB,C
S3-SLOW
S2B-A.B.C
OUTPUT
S2A-POLE
,S2A-C
t
i
! S2A-B
S2A-A
MIC- S
MIC- -h
MIC-G
DOTTED LINES ARE JUMPERS - CONN ECT BEFORE COMPONENTS.
FIGURE 12
P C BOARD COMPONENT LAYOUT
34
-------
SI
MIC
(ft
S2
°A
OB
oc
OB ATT
°OFF
S3
OSLOW
°FAST
SWITCH CODE
OUTPUT
FIGURE 13
SCHEMATIC DIAGRAM
-------
Bl,2 8.4V mercury battery (RCA VS 146X), or 9V transistor radio battery
Cl lOyF
C2 luF 10%
C3 10yF
C4 10yF
C5 luF 10%
C6 lyF 10%
C7 .81yF 10% (.33uF and .47yF in parallel)
C8 10yF
C9 .047yF 10%
CIO lOOyF
Cll 5yF
C12 5yF
CR1,2 1N34A
IC1,2,3 yA741C, "minidip" 8 pin DIP package, integrated circuit operational
amplifier
Ml 0-100yA meter, Radio Shack Micronta Type 22-037
MIC Thermo-Electron Corp Model 814 electret microphone
Rl 47kJ2
R2 3.3k«
R3 SOkiJ variable, "Trimpot" type, to fit p.c. board
R4 Ikn
R5 31.6kfl 2%*
R6 lOkn 2%
R7 3.16kf2 1%
R8 lk« 2%
R9 240ft 2%
RIO 560S2 2%
Rll 1.3kfl 2%
R12 IkJJ 2%
R13 31.6k« 2%*
R14 lOkfi 1%
R15 3.16kfl 2%*
R16 IkJi 2%
R17 430n 2%
R18 8.2kn
R19 200kfi
SI 2P7T wafer switch, minature rotary, Centralab PA-1005
S2 5P5T wafer switch minature rotary, Centralab PA-1021
S3 SPST toggle or slide switch
All resistors 5%, 1/4 watt, unless otherwise noted.
All capacitors -10+100%, 10V, unless otherwise noted.
Misc: Minibox
Battery clips and connectors
Shielded phono cable (tonearm)
p.c. board
* Note: R5, 13, 15 can be formed by series or parallel connecting 2% resistors,
such as a 60k£2 and a 63kft in parallel for R5, or by using a 1%
resistor, depending on local availability and costs. The necessary
values for R5, 13, 15 are not readily available in 2% resistors.
Figure 14
Parts List
36
-------
TOP
(D
V
T
CO
i?5
"to
ro
oo
in
-f
"oo
X
m
JL
— 0
o
O;
O
O
1-3/16"-
1-1/2"
O
1-1/2 Dl A.
l/8"DIA.-4 EA.
l/l€"OIA.-4 EA.
|/4" 01 A.-4 EA.
FIGURE 15
FRONT PANEL
37
-------
TOP
BOTTOM
<0
CO
X
I
CM
5-L
T
r
l/8"DIA,-3 EA. ON 1-7/16" CIRCLE
1-1/2
FRONT
o o o
3/8 DIA.
1/8" DIA.-4 E A.
FIGURE 16
TOP ft BOTTOM PANELS
38
-------
The faceplate of the lOOyA meter needs to be recalibrated in terms
of dB SFL. A conversion table is given below:
+ 10 dB SPL lOOyA
+ 9 dB SPL 88uA
+ 8 dB SPL 77yA
+ 7 dB SPL 67yA
+ 6 dB SPL 56yA
+ 5 dB SPL 48yA
+ 4 dB SPL 41yA
+ 3 dB SPL 35yA
+ 2 dB SPL 30yA
+ 1 dB SPL 26yA
0 dB SPL 22yA
- 1 dB SPL 18yA
- 2 dB SPL 15yA
- 3 dB SPL 12yA
- 4 dB SPL 9yA
- 5 dB SPL 7yA
The above calibration chart should be checked on the completed
instrument by replacing the microphone input with a variable amplitude
signal, such as a 1 kHz sine wave from an oscillator, variable over a
20 dB range, with a maximum output of less than a volt. An A. C.
voltmeter, calibrated in dB, either connected across the sine wave
oscillator or connected to the output of the sound level meter, would
then determine the proper meter calibration for that particular meter
and the particular diodes used in the detector. The meter scale can be
expanded or compressed somewhat by means of the meter zero adjuster.
39
-------
Alternative Microphones
The Model 814 specified herein is available from Electronic
Enterprises, 3305 Pestana Way, Livermore, CA 94550, for $32 and from
Thermo-Electron Corp., 85 First Avenue, Waltham, MA 02154, for $25 each
in lots of two or more. The specifications of the Model 814 are as
follows: Sensitivity -68 dB V (Re: IV/ybar)
Noise Level 25 dBA
Distortion (10% THD) 120 dB SPL
Vibration Level 80 dB SPL/g
Hum Level 10~7V/g
If only a Type 3, Sound Survey, instrument is desired, the Thermo-
Electron Model 5336 microphone may be a lower cost substitute. A number
of them were tested, and indeed a few met Type 2 specifications, but in
general the low frequency response was inadequate on several of them to
satisfy the tighter Type 2 requirements. The specifications are generally
the same as those for the Model 814, except that the Model 5336 is more
sensitive by approximately 7-10 dB. It is smaller physically and will
require modification of the microphone mount, but it is electrically
compatible with the circuit. It is available for $10 each in lots of
five or more from Thermo-Electron Corp.
Another alternative is the Model XL-9076 of Knowles Electronics, Inc.,
3100 North Mannheim Rd., Franklin Park, IL 60131. It is an adaptation of
their Model XD-992, a ceramic microphone with integral field effect
transistor amplifier (as in the electret microphones) which is being used
for Type 2 sound level measurements at present. The XL-9076 is an XD-992
mounted in a 1/2" dia. case so that the resulting unit satisfies Type 2
specifications. It is electronically compatible with this sound level
meter if R2 is changed to IkR to provide the proper low frequency
attenuation. The microphone mount would become a simple hollow post on
40
-------
which the microphone case would clamp. The price should be competitive
(not yet determined at this time), and the unit has the advantage that
its intended use is Type 2 sound level measurements. The other
specifications are similar to those of the Model 814 above. Although
1/2" dia. acoustic calibrators are becoming common, a sleeve, 1/2" inner
dia., 0.910" outer dia., would be necessary to adapt the XL-9076 to older
model calibrators.
If the constructor has access to a precision low frequency sound
source, such as a General Radio Type 1559-B Microphone Reciprocity
Calibrator, the value of R2 can be adjusted so that the sum of the
microphone low frequency roll-off and the £ network roll-off equals the
desired £ weighting.
If a particular Model 814 microphone has inadequate sensitivity,
the gain of the first op amp can be increased by changing R2 to Ikfl.
C2 must then be changed to 2yF to preserve the £ network low frequency
attenuation.
41
-------
CONCLUSIONS
The original charter was to determine experimentally the feasibility
of a low-cost sound level meter which would meet the minimum requirements
of ANSI SI.4-1971 and which could be reproduced by a reasonably competent
electronics manufacturer without recourse to a fully equipped acoustical
laboratory. The instrument would be used for sound surveys in an
unsophisticated manner by personnel in the field. The instrument was not
intended to replace the standard sound level meter in the hands of a
trained acoustician or noise control engineer.
The evolved design has been shown capable of meeting the requirements
of ANSI SI.4-1971 for a general purpose sound level meter. The parts and
manufacturing costs remain low, and may be reduced further by an adept
manufacturer.
The completed instrument can be calibrated by standard sound level
meter calibrators such as those available in the EPA field offices
throughout the U.S.
Although instruments built to this design have met ANSI SI.4-1971
requirements, there is no representation that all such instruments will
also meet the requirements. For a manufacturer, some sort of end item
testing would be necessary to ensure "certification." There is also no
representation that features of the design and circuit are not covered
elsewhere by patents.
Mention of particular models, manufacturers, and suppliers in this
report should not be construed as endorsement by either the United States
Air Force or the Environmental Protection Agency. This report summarizes
a research study, not a production contract design.
42
------- |