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

-------