Report of Investigations 8508

Airblast Instrumentation
and Measurement Techniques
for Surface Mine Blasting

By Virgil J. Stachura, David E. Siskind,

and Alvin J. Engler

US Department of Interior
Office of Surface Mining
Reclamation and Enforcement
Kenneth K. Eltschlager
Mining/Blasting Engineer
3 Parkway Center •
Pittsburgh, PA 15220

Phone 412.937.2169
Fax 412.937.3012
Keltschl@osmre.gov

UNITED STATES DEPARTMENT OF THE INTERIOR

James G. Watt, Secretary
BUREAU OF MINES
This publication has been cataloged as follows:

Stachura, Virgil .J
Airblast instrumentation and measurement techniques for
surface mine blasting.

(Report of investigations ; 8508)

Bibliography: p. 32•34.
Supt. of Docs: I 28.23:8508.

1. Strip mining. 2. Blasting. 3. Blast effect-Measurement. I. Siskind,

D. E., joint author. II. Engler, Alvin J., joint author. III. Title. IV.
Series: United States. Bureau of Mines. Report of investigations
8508.

TN23.U43 [TN29l] 622s [622' .31] 80-607860

 CONTENTS

Abstract ................................................. o •••••••••• 1

Introduction11 .....••••••••. 1
Acknowledgments •••••••••••• 4
Previous investigations •••••.••••.••••••• 4
Sound measurement ••••••••••••••••••• 7
Instrumentation ••••.••••••••••• 7
Determination of frequency response needed......................... 10
FM tape recorders................................................ 11
Sound level meters .•.••. IU........................................ 12
Electronic frequency response.................................. 14
Effects of wind and windscreens.................................... 14
Sound exposure level (SEL)......................................... 15
Calibration instruments and methods..................................... 16
Calibrations and experimental results................................... 20
Conclusions and recommendations ......•••••..•..•...•••....••.. ~····••••• 30
References .••.•...... .,.................................................. 32
Appendix A.--Characteristics of 14 typical airblasts.................... 35
Appendix B.--Data for measurement technique comparisons................. 41
Appendix C.--Equipment used in tests.................................... 43
Appendix D.--Low-frequency rolloff of tested equipment.................. 46
Appendix E.--Microphone types suitable for airblast measurement......... 47
Appendix F.--ru1S processing system...................................... 48

II:.LUSTRATIONS

1. Coal mine blast showing some airblast sources •• 2

2. Sound level conversion graph ........................•............. 3
3. Sonic boom microphone carrier system •••••.•••••••••••••••••••••••• 8
4. Differential pressure transducer with a carrier demodulator ••••••• 9
5. Relationship between time constant and relative pressure •••••••••• 10
6. FM tape recorder, 7 channel . ................•..................•.. 11
7. Precision sound level meters ••••••••••••••••••••••••••.••••••••••• 12
8. Standard sound measurement weighting scales •••••••.••••••••••••••• 13
9. Typical response of SLM input electronics •••••••••••••••••••• 14
10. Wind noise as function of wind speed in the range of 20 Hz
to 20 kHz ........................................................ . 15
11. Wind noise spectrum, flat weighting ••••••••••••••••••••••••••••••• 16
12. Pistonphone and sound level calibrator •••••••••••••••••••••••••••• 17
13. High-pressure, low-frequency calibrator •••••••••••••.••••••••••••• 18
14. Piston chamberphone ........•....•...••.......••.•........•.•..•.•. 19
15. C-Slow compared with CSEL, normalized to 1 sec •••••••••••••••••••• 20
16. Frequency response: B&K 2631, B&K 2209, Dallas Instruments AR-2,
and VME Noisetector ••••••••••••••••••••••••••••••••••••••••••••• 21
17. Frequency response: B&K 2631, B&K 2209, and Safeguard Seismic
Unit II. • . . • . • . . . . . . . . . . • . • • • . . . . . . . . • . . . . . . . . . • . . • . . . . . . . • . . . . . 21
18. Frequency response: GR 1933, GR 1551-C, VME Model F,
and B&K 2209 •• ·................................................... 21
ii

ILLUSTRATIONS--Continued

19. Frequency response: Vibra-tape 1000 and 2000 series, Dallas

Instruments ST-4, DP-7 transducer ••••••••••••••••••••••••••••••• 22
20. Type I airblast measured four different ways •••••••••••••••••••••• 22
21. Spectra of Type I airblast measured four different ways ••••••••••• 23
22. Type II airblast measured four different ways ••••••••••••••••••••• 24
23. Spectra of Type II airblast measured four different ways •••••••••• 25
24. Vibration response of DP-7 sound measurement transducer ••••••••••• 26
25. Sound levels for coal mine highwall shot,.s. ~ ••••••••••••••••••••••• 27
26. Sound levels for coal mine parting shots •••••••••••••••••••••••••• 27
27. Sound levels for coal mine assorted shots ••••••••••••••••••••••••• 28
28. Sound levels for quarry shots .••••••.••••.••••••.•.••.•••••••••••• 28
29. Sound levels for metal mine shots ••••••••••••••••••••••••••••••••• 29
30. Sound levels for all shots .•..•.••••••••••••.•.•••.•.••••.••••••.. 29
A-1. Quarry shots, 95-ft highwall: Type I airblast and reflected
Type II airblas t. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
A-2. Quarry shots: airblast, initiation towards gage station; Type II
airblast; airblast initiation away from gage station; and Type I
airblast........................................................ 37
A-3. Metal mine shot, very long duration airblast; coal mine shot,
Type II airblast, large holes; and coal mine shot, long-duration
airblast, six decks............................................. 38
A-4. Quarry shot, Type II airblast initial arrival, Type I airblast
when reflected; coal mine shot, parting, Type I airblast; and
metal mine shot, Type II airblast ••••••••••••••••••••••••••••••• 39
A-5. Coal mine shot, airblast from a blowout, and metal mine shot,
airblast from a partial misfire, exposed detonating cord •••••••• 40
F-1. RMS detection system •••..•••...•...••.•••..•.•.......•••..•••••••• 48
F-2. RMS detection system block diagram •••••••••••••••••••••••••••••••• 49
F-3. Amplifier modification for Multifilter •••••••••••••••••••••••••••• 50
F-4. Typical RMS output from an airblast ••••••••••••••••••••••••••••••• 52
F-5. Expanded output of rms detector ••••••••••••••••••••••••••••••••••• 53

TABLES

+,. Sound measurement descriptors ••••••••••••••••••••••••••••••••••••• 6

A'B~l.
f*"· Test airblasts: field and laboratory measurements ••••••••••••••••
C-Slow to CSEL comparison data ••••••••••••••••••••••••••••••••••••
35
41
B-2. Comparison of measurement techniques to 0.1 Hz linear peak
airblasts . •........•..............•..•..........•........•.•.•.. 42
C-1. Precision sound level meters tested by the Bureau of Mines •••••••• 43
C-2. Seismographs with airblast channels tested by the Bureau of Mines. 44
C-3. Long-term monitors tested by the Bureau of Mines •••••••••••••••••• 45
C-4. Wide-band, research-type instrumentation tested by the Bureau
of Mines .. ••••..••••.•...•••....••.•••.••.•••••.•••..••••.•••..• 45
D-1. Low-frequency rolloff of tested equipment ••••••••••••••••••••••••• 46
F-1. Integration-time parameters for GR 1926 RMS detector •••••••••••••• 51
 AIRBLAST INSTRUMENTATION AND MEASUREMENT TECHNIQUES
FOR SURFACE MINE BLASTING

by

Virgil J. Stachura, 1 David E. Siskind, 1 and Alvin j, Engler2

ABSTRACT

The Bureau of Mines has investigated techniques and instrumentation that

measure accurately the airblast overpressures from surface mine blasting. The
results include equivalencies between broadband research instrumentation and
commercially available impulse precision sound level meters measuring: root-
mean-square, peak, fast, slow, impulse, A and C weighting, C-weighted sound
exposure level (CSEL), and "linear" (flat) response. These values were
obtained from field measurements and broadband FM tape recordings of produc-
tion blasts at area and contour coal mines, limestones quarries, and iron
mines. Frequency response was determined for 14 commercial systems.

INTRODUCTION

The surface mining industry has seen extensive regulation of blast

effects, which has caused a need for uniform instrumentation and measurement
techniques. Airblast is particularly hard to regulate because it varies
widely in generation, propagation, and effects on humans and structures.
Abnormal levels of airblast sometimes occur far from a surface mine, and so
they can involve a much larger area than is usually associated with ground-
borne vibrations. The weather conditions can cause anamalous airblast propa-
gation through focusing caused by temperature inversions and intensification
from wind (I). 3 The level and character of an airblast are also strongly
affected by the degree of explosive confinement afforded by the burden, stem-
ming, and geologic conditions.

The general airblast can be characterized as an impulsive noise primarily

in the infrasonic range. Most of the energy in an airblast is inaudible,
because its frequency content is below the range of human hearing (20 Hz to

Geophysicist.
2Electrical engineer.
All authors are with the Twin Cities Research Center, Bureau of Mines, Twin
Cities, Minn.
3 Underlined numbers in parentheses refer to items in the list of references
preceding the appendixes.
2

20 kHz). Airblast level can be expressed in decibels, with the following

equation for sound pressure level (SPL):
p where P 0 is the reference pressure 20 x 10-6 N/m2
SPL 20 log

or 2.9 x 10- 9 psi, and P is the overpressure in N/m 2 or psi.

The reference pressure has been experimentally determined to be the threshhold

of hearing for young listeners, at 1,000 hz. This corresponds to 0 dB. 11any
people can hear levels as much as 10 to 20 decibels lower in amplitude. Ini-
tial discomfort and pain thresholds for steady-state sounds are 110 and 140
dB, respectively (23).

Some of the sources of airblast can be seen in figure 1. The white plume
just left of center is the source of stemming release pulses. To the right of
center, a hole has "cratered," transmitting more energy into the atmosphere
than into breaking rock. These gas releases, combined with the stress wave
energy transmitted from the rock and the heaving motion around the collar,
contribute to the higher frequency portion of airblast. A general swelling of
the shot area, including the free face, produces the "piston effect," a low-
frequency component of airblast. The high-frequency component is generally
above 5 to 6Hz, while the low-frequency portion is in the 0.5-to-2-Hz region
(28). The phenomenon of airblast generation has been studied extensively by
fifiss (39) and by Snell and Oltmans (33).

FIGURE 1. - Coal mine blast showing some airblast sources.

 3

10.00
.100

1.00
.010
(\J
c:
.c
'
.c
...
E
.100 w...
w .001 a::
0::: :::::>
:::::> (/)
(/) (/)
(/) w
w a::
.010 a..
0:.::
a.. 10- 4 a::
0:.:: w
LLJ >
>
0
0

.001
10- 5

10- 5
1o- 7 ~~~~~~~~~~~~~~~~~~~~~

50 70 90 110 130 150 170

SOUND LEVEL, dB
FIGURE 2. - Sound level conversion graph.
Airblast can be separated into two types, which are identified by their
frequency content. Type I airblast has considerable more energy above 6 Hz
than the Type II airblast. Both types are dominated by low-frequency energy
(below 2Hz), but the former has a secondary band of frequencies (over 6Hz),
4

which is less than 15 dB below the low-frequency energy level. The Type I
airblast is more troublesome because of its energy in the resonant frequency
range of structures (28). Efforts to document the environmental effects of
this acoustic energy require highly specialized instrumentation, which takes
into account the frequencies and amplitudes generated by the source.

In this report, sound pressure levels are expressed in decibels and over-
pressures in pounds per square inch (psi). Other units used in acoustics are
millibars and Newtons per square meter (N/m 2 ), also known as Pascals (Pa).
Sonic boom levels are often expressed in units of pounds per square foot
(psf). A conversion chart is shown in figure 2. The two overpressure scales
are slightly offset and not symmetrical.

1 psi 0.000145 N/m2 or Pa

1 psi 0.006944 psf
1 psi = 0.0145 mb

ACKNOWLEDGMENTS

The authors acknowledge the generous cooperation of the coal companies,

iron mines, and quarries that assisted us in obtaining the data presented in
this report. Special thanks are expressed to A. B. Andrews of E. I. du Pont
de Nemours & Co. for helpful suggestions, and to George w. Kamperman of
Kamperman Associates, Inc., for technical assistance. Additional thanks go to
Vibra-Tech Associates, Inc., VME-Nitro Consult, Inc., and Dallas Instruments,
Inc., for supplying instruments for frequency response testing.

PREVIOUS INVESTIGATIONS

The Bureau has studied the problems of airblast and instrumentation,

starting as early as 1939 (~, ~. 12-38). Most of this work involved uncon-
fined or poorly confined blasts that were dominated by acoustic energy in the
audible range (20Hz to 20 kHz) and that could be measured by standard com-
mercial sound measuring systems. In 1973, Siskind and Summers (32) surveyed
airblast noise from conventional quarry blasting, using instruments with a
variety of frequency responses. It was evident that much low-frequency energy
(less than 2 Hz) existed but that the instruments produced distortion and
"ringing" from insufficient microphone low-frequency response. A sound system
that could respond at 0.1-Hz low-frequency was then built for subsequent stud-
ies. Airblasts could be accurately captured to analyze structure response,
damage, and annoyance potential (28-29, 35). In an interim report (~),
Siskind and Summers recommended that instruments have a frequency response of
5 Hz or lower. This recognizes that houses have natural frequencies in the
range that will respond to infrasonic vibrations, and that such vibrations are
the most serious airblast problem in surface mining. Prior to this, many
measurements, made with 20-Hz systems, could not be correlated to complaints
or damage.

The main transient noise sources that cause annoyance and damage are
sonic booms, surface blasts, artillery, explosive testing, nuclear blast simu-
lation, accidental explosions, and partially confined blasts (mining,
 5

quarrying, ditching, construction, and excavation). To analyze these sources,

a variety of sound descriptors have been developed or adopted from methods
that characterize steady-state noise (table 1). Some are quite complex, in an
attempt to be all inclusive. Others involve unproven simplifying assumptions
so they can be applied to transients in general, and blasting in particular.

Kryter (18) examined sonic boom effects on structures based on peak over-
pressure levels, and also determined severity equivalencies between peak over-
pressures and "perceived noise levels" (Lpn• also labeled PNdb) from subsonic
jets. Lpn levels are rms values calculated from the "noy" values of highest
noise level in each octave (1/3 octave) band. The noy was derived from judg-
ment tests of perceived loudness conducted in a laboratory. Noy values cannot
be directly measured, so Schomer (27) described their involved calculations.
A later study by Kryter (19) involved a more complex "effective perceived
noise level" (Lepn) based-on the largest Lpn calculated from band measurements
every 0.5 sec, including a tone correction for turbine whine. Schomer (27)
summarized Kryter's studies and also proposed "composite noise ratings" (eNR),
a 24-hour integration with a 10-dB nighttime penalty. Schomer stated that the
fear of property damage is related to complaints, both of which are different
from psychological annoyance. The distinction is significant for the mining
industry.

A 1978 review by Schomer (~) gathered the results of a variety of sonic

boom studies, including those of Kryter. With Lpn the most widely analyzed
descriptor, Young (40) studied annoyance effects from unconfined impulsive
sources (for exampl;:- artillery) and utilized the concept of "sound exposure"
levels, (Lg), which are weighted rms values integrated over the duration of
the event and normalized to 1 sec. Unfortunately, labeling among studies
using Lg has not been uniform. The e-weighted sound exposure level has been
identified as Ee (40), LeE (36), or simply eSEL, with a recently recommended
standard notation of Lse (34). The preferred standard symbols for A-weighted
and e-weighted sound exposure levels are LAE and LeE• with abbreviations ASEL
(or SEL alone, which implies A-weighting) and eSEL. The sound exposure analy-
sis methods appear to offer distinct advantages over previous efforts to char-
acterize· impulsive noises. They introduce weighting, which allows selection
of the frequency bands of most concern to the noise receivers (either struc-
tures or people). Unfortunately, the standard weighting bands (A, B, e, D)
are not ideal for the troublesome responses. Sound exposure methods also pen-
alize excessively long events and tolerate shorter ones (by the 1-sec normali-
zation), recognizing that the former are more serious. This problem applies
more to annoyance than structure response, since the latter has not been
related to integrated airblast energy.
6

TABLE 1. - Sound measurement descriptors

-··------;::-De-s-cr-;i~plto-r----·-,--------------~---------..,----------­

Symbols Abbrevili:...'T--:E'_x_p-:l:--a_n_a_,t-i=-o-n--1 Equation(s) and characteristics Applications

tion
SPL Sound pressure Lp = 20 log 10 P /P 0 Standard sound level
level, in where as measured by com-
stated band. P0 = Reference pressure mercial sound level
= 20 x 10-6 N/m2 meters. Converts
= 2.9 x 10-9 psi pressures to sound
= 4.18 x 10-7 psf levels.
and
P = Sound pressure in an
unidentified bandwidth.
PNdB Perceived RMS values computed from the noy Designed for aircraft
noise level. values of the highest noise level and nonimpulsive
in each octave (or l/3 octave) band, sources ..
based on the 40-noy D-weighted
scale; maximum integration is
1/2 sec (18 ).
EPNdB Effective per- As Lpn except utilized a 1/2-sec time Do.
eel ved noise correction and pure tone penalty of
level. up to 10 dB (for turbine whine) (19).

NA Equivalent
sound level.
LAeq = 10 log1r T c [!It.
10LpA(t)/10dt-J
Running time average;
any time duration
where can be used; good
LAeq = A-weighted equivalent sound level for long-term noise
T = Time interval for average impact analysis.
LpA = A-weighted sound level during
time T (any weighting can be nsed)

NA Equivalent 1 f It(O?oo) [T-A(t)+l0]/10 As above, except

day-night = 10 1 --
oglO 86400
L ,(oooo) 10 -p dt required at least a
sound leveL 24-hour integration
+ J'(aaoo) LpA(t)/10 and includes a 10-dB
t(0?00)10 dt nighttime.

>(a4oo) [L-A(t)+l0]/10 }
+
J>(aaoo) 10 " dt

= A-weighted day-night average sound level

A-weighted sound level within time
interval (any weighting can be used)
t = Time interval, sec, within hours indi-
cated, over which intergration takes
place
LeE• LAE CSEL, SEL C-, A-weighted Measure of acoustic
(preferred); sound expos- energy within an
also Ls, sure levels. event; normalization
Lsc • Ec, to 1 sec gives a
LsA• EA. where penalty to longer
LeE • c-weighted SEL events (3 dB per
Pc • C-weighted sound pressure (any doubling of time);
weighting can be used) it also tolerates
P0 • Standard reference pressure higher SPL for dura-
• 20 x 10-6 N/m2 tions shorter than
tl, t., • lleginning and ending times of event 1 sec.
t0 • 1 sec

Alternate form:
LAB • lO loglO -
1
to tl
t.,
J
1 (t)/lO
10 pA dt

where
LpA • A-weighted sound level during time
interval t 1 to t., (any weighting
can be used) r:,
NA PLdB Perceived PLdll • 55 + 20 log10 T p Developed to charac-
level, dB. terize the sharpness
where and loudness of
6P = Pressure change, psi sonic booms ..
T ~ Rise time sec. corresponding t()_::A:..P_ _ _ __, ___________~
NA Not available.
 7

The CHABA Committee Report presents a summary of averaging and SEL meas-
urernent methods, including average sound level (Leq or LT), which runs aver-
ages over any desired time periods, and day-night average (Ldn), which is a
24-hour Leq with a 10-dB nighttime penalty. Schomer (25) also uses Ldn,
except that he introduces C-weighting. Both Schomer (25) and The CHABA Report
(36) examine LeE and LAE and relate them to annoyance potential of blasts.
Generally, the sources used to develop these levels--steady state (aircraft),
sonic booms, and unconfined surface blasts--are quite different from mine pro-
duction blasts. In particular, Schomer's (25) comparisons of artillery LAE's
to blasting do not consider the vast difference between amounts of A-weighted
energy present in the two sources. A more serious problem is the tendency to
include transients such as blasts in long-term (that is, 24-hr) sound aver-
ages which leads to anamalous situations where the relatively coarse Ldn
values do not accurately characterize the annoyance and damage potential of
infrequent events.

Higgins and Carpenter (11) introduce the concept of perceived level

(PLdB), based on the actual characteristics of the overpressure. This method
uses pressure changes and corresponding rise times, and, although developed
for sonic booms, could be applied to airblasts.

The Bureau (31) recently serveyed these noise descriptors and their
applicability to blasting, specifically comparing LeE's, special weightings,
and various linear sound levels with actual structure responses.

SOUND MEASUREMENT

Instrumentation

The measurement of frequencies below the audio range requires specialized

instrumentation such as microphone carrier systems, one of which is shown in
figure 3. The Bruel and Kjaer 4 (B&K) utilizes a F}1 carrier-demodulator system
enabling it to operate down to de response or to the lower limiting frequency
of the microphone. The system shown has a l-inch microphone with a time con-
stant of 1.6 sec, and a frequency response that is flat from 0.1 Hz to
8kHz + 3dB. The tolerance in the dynamic range of + 3 dB is the relative
deviation from the nominal value and is called "flat." This type of system
has measured sonic booms and other acoustic transients that contain energy at
very low frequencies. An auxiliary storage system must be used to capture
transient overpressures. This can be a light beam oscillograph, oscilloscope
with camera, storage oscilloscope, waveform recorder, FM analog magnetic tape
recorder, or similar device with a frequency response that is flat (±3 dB)
down to at least 0.1 Hz.

4Reference to specific brand names is made for identification only and does
not imply endorsement by the Bureau of Mines.
8

FIGURE 3. - Sonic boom microphone carrier system.

An alternate system with a differential-pressure transducer (fig. 4) has
been adopted by the Bureau (~). The frequency range of this transducer is
more limited at higher frequencies than the sonic boom microphone carrier sys-
tem, owing to the dimensions of its air passages, but it does effectively
cover the blast-generated frequency range. A precision needle valve adjusts
the time constant, which controls the low-frequency response.
 9

FIGURE 4. - Differential pressure transducer w.ith a carrier demodulator.

In both carrier systems, the low-frequency limit is determined by the

following equation:
1
FR. =

where FR. = lower limiting frequency (Hz)

and < time constant, (sec), for any given pressure P0 to drop to (1/e) P 0
(see • 5),

with~= 3.14159 •••

and e = 2.71828 •••

This equation calculates the low-frequency limit (-3 dB) from the rate of
pressure equalization between the positive and negative sides of the trans-
ducer diaphragm. A slow leakage is needed to maintain ambient atmospheric
pressure in the cavity behind the diaphragm for the duration of a time con-
stant. This pressure becomes a reference in making a differential measure-
ment. A totally sealed cavity would give a 0-Hz (de) lower limiting
10

frequency, but temperature expansion and contraction of the sealed air volume
and barometric pressure changes could damage the diaphragm and cause
drift (29).

Determination of Frequency Response Needed

The frequency response selected for airblast measurement may meet one of
several criteria. One rule-of-thumb is that gages should respond to four
times the frequency of interest (24). Using this rule, previously measured
airblasts (28, 31-32, 35, 39) suggest a range of 0.1 to 200Hz. Reed defined
a second criterion-ror~he~igh-frequency response based on attenuation of

Po

w
0:::
~
(j)
(j) T= time constant
w
0:::
0...
.Lpo
w 2
>
1-
<(

GJ ..!.. Po-
o::: e

0 2 3 4 5 6 7 8 9 10
T TIME, sec
FIGURE 5. - Relationship between time constant and relative pressure.
 11

explosion waves from unconfined high explosive and nuclear blasts. He stated
that there appeared to be no need for greater response greater that 1 kHz,
except at locations very close to small explosions. For environmental moni-
toring at levels of 114 to 144 dB at distances greater than a kilometer, 1 kHz
would be adequate, if not an order of magnitude on the conservative side (24).
Confined mine production blasts would not generate the high frequencies Reed
observed. A third criterion is based on a theoretical study of Crocker and
Sutherland, which resulted in formulas and curves for calculating upper and
lower limits of frequency response (6). For ±10 pet accuracy in peak pressure
and positive phase durations, a range of 0.05 Hz to 1 kHz would be required.
The first criterion would be sufficient for confined mine production blasts,
while unconfined surface blasts would be best measured under the second and
third.
FM Tape Recorders

A typical FM tape recorder, shown in figure 6, is a portable 7-channel

instrument with tape speeds from 15/16 to 60 in/sec. The frequency response
is from de (0 Hz) to a maximum determined by the tape speed selected (for
example, 2.5 kHz at 7.5 in/sec. An AM tape recorder has insufficient response
at low frequencies for airblast recordings. Such recorders are designed for
use in the audio range (20 Hz to 20 kHz), rather than the infrasonic range
where most airblast is found (21).

FIGURE 6. - FM tape recorder, 7 channel.

12

Sound Level Meters

Commercial sound level meters (SLM's) fall into four categories:

Type I - Precision; Type II - General Purpose; Type III - Survey; and
Type IV - Special Purpose. The tolerances for these categories are defined by
the American National Standards Institute (ANSI) for A-, B-, and C-
weighting (2). In figure 7 two Type I meters are shown with their respective
calibrators-and windscreens.

The Type I precision meter is available with an impulse response that

becomes more accurate for transient noises with sharp rise times (14-15).
Standard (nonimpulse) sound level meters can only resolve signals having crest
factors of up to 10 dB. The type I impulse precision sound level meters con-
tain squaring circuits that can resolve signals having 20-dB crest factors
even when used with fast or slow response. Crest factor is the ratio of peak
to rms and is a measure of "peakiness" of a signal.

FIGURE 7. - Precision sound level meters.

 13

SLM's have selectable response times: slow, fast, impulse, and peak.
Slow, fast, and impulse have time constants of 1,000, 125, and 35 msec,
respectively (1-2). Peak response does not have a standardized time constant
but should be about 10 ~sec for rise times in the upper audio range. Storage
of the highest peak value is an important feature (peak hold) when measuring a
short, unexpected event.

Slow response is used to measure a random noise source, since its averag-
ing effect makes reading easier. Fast response has the same effect, but to a
lesser degree. Slow and fast produce rms readings only. Impulse response was
designed for short-duration, single-pulse events. The 35-msec integration is
based on the mean averaging time of the human ear. This setting should not be
used for correlation with structure response or damage, because the response
of the ear and structures are not related. The impulse setting has a 3-sec
decay time constant, which can introduce errors from successive pulses such as
those found in a delayed mine blast (~).

A-, B-, and C-weightings simulate human sensitivities electronically

for different sound levels. A-weighting approximates the inverse of the
equal loudness contour at a sound level of 40 dB (about 20 x 10- 6 Nm 2 ); B-and
C-weightings approximate this contour at higher sound levels. Frequency
responses of these three weightings are shown in figure 8. All three weight-
ings attenuate the lower frequencies strongly present in airblast. c-
weighting, which attenuates low frequencies the least, is 3 dB down at
31.5 Hz. Flat or linear specification, found on SLM's indicates broader fre-
quency response than C-weighting. The frequency response is controlled either
by the input electronics or by the lower limiting frequency of the microphone.
A typical flat and linear response is from about 5 Hz to more than 10 kHz and
can vary considerably among manufacturers.

+10

0
OJ
"0 -10
w
0-20
::::>
1--
::i -30
a..
:E -40
<t:

-50

-60

-7o~~--~~~~--~~~~~--~~~~~--~~~~~~~~~~~

I 10 100 1,000 10,000 100,000

FREQUENCY, Hz
FIGURE 8. - Standard sound measurement weighting scales.
14

The most important features on SLM's for blast measurement are linear or
flat-peak mode of operation, C-slow mode of operation for approximating sound
exposure level, and peak-hold or sample-hold to retain the highest reading.

Electronic Frequency Response

The frequency response of typical SLM input circuits is shown in figure 9

(5). This is a second factor determining how the system works overall. IEC
179 is an international specification that does not define frequency response
below 20Hz, except for the upper limit, which extends to 10Hz. The degree
of frequency response rolloff must be documented when wideband recordings are
played back into the input circuitry. Even if the frequency response of a
microphone is extended, the whole system may not improve because the input
preamplifiers have insufficient frequency response. A different microphone
may change the input impedance matching so that overall response at low fre-
quencies decreases (21). The rolloff of a condenser microphone may be altered
by the introduction of a series capacitor as illustrated by the 2-Hz and 10-Hz
curves in figure 9 (5). The direct curve is for an impedance matched input
from a tape recorder7

Effects of Wind and Windscreens

Wind blowing across a microphone can generate turbulence, which can cause
a noise measurement to be erroneously high. Figure 10 shows typical wind
6

0
(I) - 2
"0

w
(/) 4
z
0
a.. - 6
(/)
w
0: 8
w
> -10
1- KEY
<(
_J
w 12 2 Hz lower limit 10Hz lower limit
0:
-14
- - - - l-in. microphone ------I- in. microphone
- - - - lt2 in. microphone in. microphone
- - - - - 1t2
16 ----Direct -----Direct

5 10 20 50 100
FREQUENCY, Hz

FIGURE 9.- Typical response of SLM input electronics (2,).

 15

II 0

ro
-.:::100
..J
w
>
w
..J
0
z
::J 90
0
(J)

KEY
a Wind parallel to diaphragm
80 b Wind perpendicular to diaphragm
cAs b with windscreen
d As a with windscreen

d
70~----~------~------~------~------~------~------~------~
0 10 20 30 40 50 60 70 80
WIND SPEED, mph
FIGURE 10. - Wind noise as function of wind speed in the range of 20Hz to 20 kHz Q).

noise for microphones with and without windscreens (3). For example, a wind
about 25 mph parallel to the diaphragm (a and d) gen;rated about 99 dB without
a windscreen and 73 dB with one. The windscreen reduced this noise by 26 dB.
Figure 11 illustrates the effects of wind perpendicular to the diaphragm
with and without a windscreen (23). The windscreen reduces wind noise by 17
to 28 dB in the 25 to 6,000 Hz range. The microphone should be protected from
direct exposure to wind; if not possible, a windscreen should be used.

Sound Exposure Level (SEL)

The Committe for Hearing, Bioacoustics, and Biomechanics (CHABA) Working

Group 69 has recommended that the U.S. Environmental Protection Agency adopt a
C-weighted sound exposure level (CSEL or LeE) method to regulate noise from
booms, quarry blasts, or artillery fire (36). The equation for and character-
istics of this level are given in table 1-.- One advantage of using CSEL for
blast regulation is that a standard sound level meter may be used to approxi-
mate values. The meter must be set on "C-weighting" and "slow response."
16

Errors for shot durations of

0.5, 1.0, and 2.0 sec will
25 mph (40 km/hrl be 1, 2, and 3.5 dB, respec-
tively (16). A Type I pre-
90
cision SLM may be preferred
because better low-frequency
response (below 20Hz) is
eo required (2). Unfortu-
nately, in-tests to date,
use of CSEL has not appreci-
70 able improved prediction of
damage over the previously
recommended linear-peak
60 measurements (~, 28, 30-31,
00 12)·
"'
..J
LIJ
> 50
CALIBRATION INSTRUMENTS AND
LIJ
...J METHODS
0
z
<t
00 Sound level meters may
40
be calibrated dynamically
with a sound level calibra-
tor (SLC) or a pistonphone
30 (fig. 12). Dynamic calibra-
tion should be traceable to
the National Bureau of
20 Standards (NBS), since this
assures that the proper sen-
sitivity is being main-
10 tained. Pistonphones are
slightly more accurate but
are mechanically limited to
lower frequencies--in the
OL----L--~~--~----~----~--~----~----~
25 50 100 200 400 1,000 2,000 4,000 10,000 case pictured, to a single
FREQUENCY, Hz (lt3 octove) frequency of 250 Hz. The
SLC illustrated has select-
FIGURE 11. - Wind noise spectrum, flat weighting (23).
able frequencies of 125, 250,
500, 1,000, and 2,000 Hz.

Calibration in the frequency domain is more difficult and requires

specialized equipment and techniques. Three types of pressure amplitude
calibrations will be illustrated here: Change of height; a commercially
available high-pressure, low-frequency calibrator; and a simply built
piston-chamberphone. An additional method is described by Hunt and Schomer
that locates the -3 dB point (11).

The "change of height" method can determine the -3 dB response point with
the equation on page 9 and the dynamic calibration can be determined by the
following equation:
 17

FIGURE 12. - Piston phone (left) and sound level calibrator.

(T + 273)
h = PC 4.548 P0

where h = change in altitude (m)

Pc = calibration pressure (N/m 2 )

T = temperature ( 0 c)

and Po = atmospheric pressure (mm of Hg)

This method is only effective for microphones or transducers with good low-
frequency response (long time constants), e.g. below 0.05 Hz. This method
18

measures the change in atmospheric pressure over a measured altitude shift

(for example, a change of approximately 5.2 feet would produce 120 dB re 20 x
10-6 N/m2 at 0° C. and 760 mm of Hg) (29, 35). This method works best for
instruments with long time constants and produces one calibration point, the
-3 dB level. A commercially available high-pressure, low-frequency micro~hone
calibrator (fig. 13) may be used over a continuous frequency range of 10- Hz
to 1 kHz and can generate pressures up to 172 dB. This is a constant force
rather than a constant displacement pistonphone and is driven by a miniature
electromagnetic shake table. When the pressure changes are turning from adia-
batic to isothermal processes, the error is minimized by this method. Changes
in gas compressions can be adiabatic or isothermal depending on volume, shape
of the chamber, and the rate of change (frequency). This calibrator was modi-
fied to accommodate differential pressure gages and 1-1/8-inch microphones
with a minimum of internal volume changes to the pressure chamber. The fre-
quency response curves in figures 21 to 24 were obtained with this calibrator
(3?_, E_).

The third method of calibration is the piston-chamberphone (fig. 14). It

generates a sound pressure of 0.0005 psi (125 dB peak SPL) by a gas model-
airplane engine, a clear plastic chamber, and a variable-speed electric motor
(29, 35). The ratio of specific heats of gas (1.30 to 1.41), which must be
included when calculating the sound pressure within the chamber works out to
about a 3-dB change in the pressure. A correction factor is then applied for
nonadiabatic compression in the cylindrical volume·for frequencies below

FIGURE 13. - High-pressure, low-frequency calibrator.

 19

FIGURE 14. • Piston chamberphone.

20

3 Hz (4). On this particular system, the greatest correction was 0.59 dB at

0.1 Hz~ The effective frequency range was 0.1 to 100Hz. Sealing was a
problem at low frequencies, and vibration and lubrication at the high
frequencies. Some of these problems may be minimized by making the volume
large compared with the surface area of the calibrator chamber (~, ~).

CALIBRATIONS AND EXPERIMENTAL RESULTS

The calibration and comparison tests in this section were used to deter-
mine the most effective instrument and measurement technique for surface mine
blasting. A comparison of C-slow and CSEL normalized to 1.0 sec is presented
in figure 15. Standard commercial SLM's were used to obtain the C-slow

117

3
0
(/)
I
(.)
~

__J
w
> Ill
w
__J

0
z
:::)
0
(/)

105

99

93
87

87 93 99 I05 I II 117
SOUND LEVEL, CSEL, normalized to 1.0 sec
FIGURE 15. • C-Siow compared with CSEL, normalized to 1 sec.
 21

KEY
10 - - B 8. K 2631 l-in. microphone
--Dallas Instruments AR-2 11;8 -in. microphone
co 5 ----- B 8. K 22091;2-in. microphone, 10-Hz lower limit
"0
- - - VME noisetector (prototype)
~ 0
=>
1--
__J -5
a..
~
<(- 10

-15

-20 L--~--L~-~~LUL__~-~-L~~_u~--~~-~-L~~---L-~~~~~
0.01 0.1 1.0 10.0 100.0
FREQUENCY, Hz

FIGURE 16.- Frequency response: B&K 2631, B~K 2209, Dallas Instruments AR-2, and
VME Noisetector.
15

10

co 5
"0

w
0 0 .... --------
,/"
...- -;/ ....-~-

L:
=> /
I- /
/ -""/
-5 KEY
/ "
__J I /
I
a.. I
/
/
/
~ 1;2 -in. microphone I
<( -10 I
I /
--Safeguard seismic unit Il ' /
-15 ----B 8. K 2209 l-in. microphone,2-Hz lower limit
---B e. K 2209 1-in.microphone,IO-Hz lower limit
-20
0.01 0.1 1.0 10.0 100.0
FREQUENCY, Hz

FIGURE 17.- Frequency response: B&K. 2631, B&K 2209, and Safeguard Seismic Unit II.

10 KEY
- - - GR 1933 1;2 -in. and l-in. electret microphone
ID 5 and l-in. ceramic microphone ·
"0
- - GRI551-C with impact analyzer
~ 0
----- VME Model F with l-in. ceramic
~- -- ----./-=::-::--::::-:::::-;:;.-=~------1
=> I , -
1-- microphone I 1/ //
...J -5 / I /
a.. --- B 8. K 2209 1;2 -in. microphone, 2-Hz I I I
:::E lower limit
<( -10 / I
-15

-20
0.01 0.1 1.6 10.0 100.0
FREQUENCY, Hz

FIGURE 18.- Frequency response: GR 1933, GR 1551-C, VME Model F, and B&K 2209.
22

readings, and a GR-1926 rms detector was used for the CSEL values. The corre-
lation coefficient of C-slow versus CSEL is 0.986, and standard deviation is
0.0000628 psi with a best-fit line of Y ~ 0.968 X+ 0.0000169 psi. Individual
data points are listed in table B-1. The high degree of correlation suggests
that C-slow may be an approximation for shots of 1.0-sec duration or less.
Longer shots require an SLM modified to integrate for longer than 1.0 sec or
the correction factors given by Kamperman. His corrections for shots of 0.5,
1.0, and 2.0 sec were 1, 2, and 3.5 dB, respectively (~).

The frequency responses of all airblast instruments used or tested by the

Bureau is shown in figures 16 to 19. A modified B&K 4221 high-pressure micro-
phone calibrator was used to obtain the overall response from 100 Hz down to
the frequency at which the response dropped 3 dB.

Figures 20 to 23 illustrate the effect of various frequency responses on

amplitude for typical Type I and Type II airblasts. The best response is on
top and the poorest is on the bottom, with the vertical scales relative in
size.
15
KEY
10 - - Vibro-Tape 1000 series and nominally
the 2000 series
~ 5 Dallas Instruments ST-4

~ 0
::J
t-
::i - 5
Q_
2
DP-7 transducer

//
,//
./' -- _...-------

/
/
<1-10 // /

~~> -~~.,.d ~ ~i-J_J._.l.J_L__ ------L-J__~LLLLj

-15

-20 _ ____[_ L___j__l___L-'-'LLJ.........

0.01 0.1 1.0 10.0 100.0

FREQUENCY, Hz
FIGURE 19.- Frequency response: Vibra-tape 1000 and 2000 series, Dallas Instruments ST-4 1
DP-7 transducer.

DP-7 No. 10
133 dB peok,O.I-Hz cutoff (-3dB)

B 8 K 2209 linear
128 dB peak, 2-Hz cutoff (-3dB)
-----~

GR 1933 linear
~~~~~~---~-~~~--=-125dB peok,_?-Hz cutoff (-~dB)

B 8 K 2209 C-weighted
85-13c_ _ _-;Jl 111AMI.I>A!\M-.N""V1IV.kJ Mw-v..._.~-~~--~-------12Q~_I?~~k,_~5j:l3__~utof!__{-3d8)
C-slow meter reading
0 0.5 109 dB C
I""'P""""""""''"""liiiiiiil"!!"""liiiiiil,..........,~
TIME, sec
FIGURE 20. · Type I airblast measured four different ways.
 23

-10

-20 0.1-Hz, linear-peak

-30

-40

-50

-10
CD
"0
~ -20 2-Hz, linear-peak
w
a
::J
1- -30
_J
0..
:! -40
<{

w
> -50
1-
<{
_J -20
w 5-Hz, linear-peak
0::

-30

-40

-50

-20
C-weighted-peak
-30

-40

-50

-60
0 10 20 30 40 50
FREQUENCY, Hz

FIGURE 21. - Spectra of Type I airblast measured four different

ways.
24

DP-7 No.8
48-13 I 21 dB peak,O. 1-Hz cutoff (-3 dB)

B a K 2209 linear
I 19 dB peak, 2-Hz cutoff (-3d B)
~

GR I933 linear
Ill dB peok,5·Hz cutoff (-3dB)
~~~

GR 1933 C-weighted
48-13c 99 dB peok,31.5·Hz cutoff (-3dB)
--~~~NW~~~~~~~--------------~---~~~-
0 0.5 C-SIOW meter reading
~ 85 lt2 dBC
TIME, sec

FIGURE 22.- Type II airblast measured four different ways.

A typical Type I coal mine highwall shot is shown in figure 20. The 2-Hz
instrument reads 5 dB low and shows a shift in peak value to the right and
below the center line because of "ringing." This ringing causes a reduction
in amplitude and a phase shift due to the instrument's inability to follow the
lower frequency excursions of the airblast. The 5-Hz and C-weighted instru-
ments show further reductions in amplitude of 8 and 13 dB, respectively, with
a C-slow meter reading 24 dB below the most linear instrument. The frequency
spectra for figure 20 are shown in figure 21. This further illustrates the
loss of amplitude at lower frequencies.

A typical Type II airblast is shown in figure 22, with its respective

frequency spectrum in figure 23. The 2-Hz instrument shows a shift in peak
value to below the center line and a 2-dB decrease in total amplitude. The
respective reductions in amplitude for 5-Hz, C-peak, and C-slow are 10 dB, 22
dB, and 35.5 dB.

The 2-Hz instrument has its response least affected for both types of
airblast, since the 1-Hz energy is so close to the -3 dB point. The response
of the 5-Hz and C-weighted instruments sharply decreases (a lower reading for
the same airblast) for the Type II blasts because of the absence of higher
frequency energy. The 2-Hz instrument gives a more accurate measure of the
total energy in an airblast than the 5-Hz or C-weighted instruments.

Shake table tests were run on the Validyne DP-7 to determine the effects
of vibration at a constant peak acceleration. A comparison with the DP-7 can
be made to a standard B&K 2209, with a l-inch microphone (fig. 24). The level
generated by external vibration ranges from 68 to 73 dB at a constant sinusoi-
dal acceleration of 1 g. An additional test was made to determine the effect
of air being driven into the open port of the transducer. A relatively low
level of noise (58 to 63 dB) was generated at a high level of acceleration
(16 g). Vibration effects were minimal.
 25

-10
0.1-H z, I in eo r-peak
-20

-30

-40

-50

-10
CD
-o
20
w
Cl
:::>
1- -30
..J
a..
:::iE -40
<t
w -50
>
1-
<t
..J
w
c:: -10

-20

-30

-40

-50

C-weig hted-peok
-30

-40

-50
0 10 20 30 40 50
FREQUENCY, Hz
FIGURE 23.- Spectra of Type II airblast measured four different
ways.
26

100

95
B & K 2209
90
~--a. .... ....
85
KEY
o Vertical motion perpendicular to diaphragm (a lg)
80 o Horizontal motion perpendicular to diaphragm (a= lg)
r:D e:. Air port pointed in, in direction of motion
-o
r {a=l6g) diaphragm parallel to motion
:::£ 75
<!
w
(L

70

65

60

55 ~

50
2 4 7 10 20 40 70 100
FREQUENCY, Hz
FIGURE 24. - Vibration response of DP-7 sound measurement transducer.

The change in sound level readings by type of shot and instrument or

measurement technique are shown in figures 25 to 30, with their statistics in
table B-2. The horizontal axis is the airblast as measured with the most lin-
ear instrument and vertical axis is the scale for the 2-Hz peak, 5-Hz peak,
C-slow, and PL dB readings. For example, in figure 25, a 130-dB airblast
would produce a 127-dB peak reading on a 2-Hz instrument, a 126-dB peak on a
5-Hz instrument, a 106-dB C-slow on an integrating type sound level meter, and
a reading of 93 dB using perceived level (PL) techniques.
 27

135

130

125

120
CD
"
-'
w 115
>
lU
-'
0
z
::;)
110
0
(/)

105

100

95

90 95 100 105 110 115 120 125 130 135 140 145 150
SOUND LEVEL, decibels measured with 0.1-Hz low-frequency cutoff

FIGURE 25. - Sound levels for cool mine highwoll shots.

135

130

125

..,
Ol

..J
w 115
>
w
_)

0
:z II 0
:::>
0
(/)

105

100

95

90 95 100 105 110 115 120 125 130 135 140 145 150
SOUND LEVEL, decibels measured with Q.I-Hz low-frequency cutoff

FIGURE 26. - Sound levels for cool mine porting shots.

28

140

135

130

125

120
.,m
_j
w 115
>
w
-'
Cl
3110
0
(/)

105

100

95

90 95 100 105 110 115 120 125 130 135 140 145 150
SOUND LEVEL, decibels measured with 0.1-Hz low-frequency cutoff

FIGURE 27. - Sound levels for coal mine assorted shots.

135

130

125

120

d> 115
w
-'
Cl
z 110
::>
0
(/)

105

,.
100

95

90 95 100 105 110 115 120 125 130 135 140 145 150
SOUND LEVEL, decibels measured with 0.1-Hz low-frequency cutoff

FIGURE 28.- Sound levels for quarry shots.

 29

135

130

125

120

-'
~ 115
w
-'
0
2 110
::J
Sl
105

100

95

90 95 100 105 110 115 120 125 130 135 140 145 150
SOUND LEVEL, decibels measured with 0.1-Hz low-frequency cutoff

FIGURE 29. - Sound levels for metal mine shots.

135

130

125

120
II)
'0
_;
~ 115
w
-'
0
3 II 0
Sl
105

100

95

90 95 100 105 110 115 120 125 130 135 140 145 150
SOUND LEVEL, decibels measured with 0.1-Hz low-frequency cutoff

FIGURE 30. - Sound levels for all shots.

30

The various shots are classified by the kind of mine; the coal mine shots
are broken down into three types. Highwall coal shots have a full or partial
free face but are well confined, since the overburden is not extensively dis-
placed. Parting shots are in the thin, hard material separating two coal
seams and are difficult to confine well. The assorted shots included ditch
and "sweetner" shots, which employed relatively shallow blast holes. "Sweet-
ner" shots were used to break up the overburden sufficiently to level the sur-
face for a walking dragline. All quarry blasts were in limestone. Metal mine
blasts were recorded at iron mines on Minnesota's Mesabi Iron Range; these
were very large and were recorded at greater distances than the coal or quarry
blasts.

Parting shots are very strong Type I blasts and show consistently higher
readings for 2-Hz, 5-Hz, C-slow, and PL dB methods. Metal mine shots at large
distances are strongly Type II, hence the 2-Hz, 5-Hz, C-slow, and PL dB meth-
ods are consistently lower. The selection of an instrument to measure Type I
blasts does not appear to be so critical as for Type II blasts, since 2-Hz and
5-Hz instruments read very nearly the same as 0.1-Hz instruments when measur-
ing parting shots. Generally, the standard deviations are smallest for the
2-Hz and 5-Hz instruments with two exceptions: coal mine parting and quarry
shots. These two exceptions are characterized by poorer confinement and the
heaving of material. Coal mines and quarries are often close to residences
and, therefore, reduce the chance of dispersing their high-frequency energy.

Readings obtained with different forms of instrumentation or techniques

(0.1-Hz peak, 2-Hz peak, 5-Hz peak, C-slow, and PL dB) may be compared from
these graphs.

CONCLUSIONS AND RECOMMENDATIONS

The type of instrument recommended is influenced by the frequencies gen-

erated by the airblast. Since 0.1-Hz equipment is expensive and difficult to
maintain and use routinely, a standard sound level meter or seismograph with
an airblast channel may be preferable and used effectively with the equiva-
lence graphs presented in this report.

Any sound level meter used should be a Type I, impulse, precision instru-
ment. This type of meter is preferred because of its higher crest factor. A
peak "hold" feature is recommended when one is monitoring a short, unexpected
event. A C-slow reading may be obtained with such meters that have "true" or
"quasi" rms detectors with equal accuracy for blast measurements. The C-
weighting should meet ANSI 81.4-1971 specifications for Type I meter.

The airblast channel on a blasting seismograph should be able to record a

complete time history from which a peak measurement can be obtained. The fre-
quencies generated can be calculated from a time history for a blast design
analysis.

When obtaining monitoring equipment, documentation of the linearity of

the frequency band and rolloff rates should be requested, since small varia-
tions in frequency response can change output levels considerably. The
 31

rolloff should be standardized to minimize the deviations in readings where

frequencies are present below the -3 dB point. Appendix D contains a list of
instruments and their rolloffs. When a sound level meter or seismograph is
serviced, its low-frequency response and dynamic calibration need to be
verified.

Production blasts are confined and delayed such that they generate a
lower band of frequencies than do open-air, single charges. Even when a hole
craters or blows out, confinement is enough to prevent the fast pressure rise
times seen in open airblasts. The distances at which production blasts are
monitored also reduce high frequencies through attenuation and dispersion. An
upper frequency of 200 Hz is sufficient for regulatory monitoring. For air-
blasts generated by mine production blasting, the frequency response should
meet or exceed the following specifications:

0.1-Hz peak instrumentation •••••••• 0.1-200 Hz + 3 dB

2-Hz peak instrumentation •••••••• 2-200 Hz + 3 dB
5-Hz peak instrumentation •••••••• 5-200 Hz + 3 dB
C-slow . •.....•.....••....•..••••. ANSI 81.4::-1971 (Type I meter)

The 5-Hz and 6-Hz rolloff instruments will read essentially alike, so for
practical purposes they are interchangeable. For unconfined surface blasts at
short distances, an upper limit of 450 Hz or higher is recommended. For exam-
ple, uncovered detonating cord requires an extended high-frequency response.
For research tests, the lower limit should meet or exceed 0.1 Hz. A sound
level calibrator or pistonphone should be used to verify dynamic calibration
before each use. This calibrator should be checked anually against a source
traceable to the National Bureau of Standards.

A microphone windscreen will prevent false readings from wind gusts and
protect the microphone from shock damage or adverse weather. Foreign matter
could shift the frequency response by blocking the pressure equalization hole.

Microphones with a very low frequency response are more sensitive to wind
than those used for voice communication. The microphones should be on a tri-
pod or held motionless during a measurement, because variations in altitude
register as air pressure changes. The microphone should be at least 3 feet
aboveground and to the side of a structure to minimize reflections. The ori-
entation of the microphone is of minor importance, since it is directional
only at high frequencies (above 1,000 Hz) and essentially omnidirectional at
blast-generated frequencies.

A diaphragm of l-in diameter or greater will have a slight advantage for

low-frequency response and sensitivity, but this advantage can be compensated
for by electronic circuits for smaller microphones.
32

REFERENCES

1. Allen, D. S. An Integrating Real Time Analyzer. Sound and Vibration,

March 1978, PP• 4-6.

2. American National Standards Institute. American National Standard

Specification for Sound Level Meters. ANSI S1.4-1971, 1971, 22 pp.

3. Brach, J. T. Acoustic Noise Measurements. January 1971, 203 pp.

4. Bruel, P. V. Measuring Microphones. Bruel and Kjaer Instruments, Inc.,

November 1971, 138 pp.

5. Bruel and Kjaer Instruments, Inc. Impulse Precision Sound Level Meter,
Type 2209. June 1972, 114 pp.

6. Crocker, J. J., and L. C. Sutherland. Instrumentation Requirements for

Measurement of Sonic Boom and Blast Waves--A Theoretical Study.
J. Sound Vib., v. 7, No. 3, 1968, pp. 351-370.

7. E. I. duPont de Nemours & Co., Inc. Dupont Blasters Handbook.

Wilmington, Del., 1977, 494 pp.

8. Fredericksen, E. Low Impedance Microphone Calibrator and Its Advantages.

Bruel and Kjaer Instruments, Inc., Tech. Rev. 4, 1977, pp. 18-25.

9. General Radio Co. Instruction Manual Type 1925, Multifilter. Concord,

Mass., 1969.

10. Instruction Manual Type 1926, Multichannel rms Detector.

Concord, Mass., 1970.

11. Higgins, T. H., and L. K. Carpenter. A Potential Design Window for

Supersonic Overflight Based on the Perceived Level (PL dB) and Glass
Damage Probability of Sonic Booms. U.S. Dept. Transportation, Fed.
Aviation admin., Rept. FAA-RD-73-116, August 1973, 25 pp.

12. Hunt, A., and P. D. Schomer. High-Amplitude/Low-Frequency Impulse

Calibration of Microphones: A New Method. J. Acoust. Soc. Am., v. 65,
No. 2, February 1979, pp. 518-523.

13. Ireland, A. T. Design of Air-Blast Meter and Calibrating Equipment.

BuMines Tech. Paper 635, 1942, 20 pp.

14. International Electrotechnical Commission. Precision Sound Level Meters.

Pub. 179, 1973, 26 PP•

15. First Supplement to Publication 179 (1973), Precision Sound Level

Meters. Pub. 179A, 1973, 21 pp.
 33

16. Kamperman, G. (Kamperman Associates, Inc., Chicago, Ill.). Quarry Blast

Noise Study for the Illinois Institute for Environmental Quality.
Final rept. to IIEQ, December 1975, 37 PP•

17. Kamperman, G., and M. A. Nicholson (Kamperman Associates, Inc., Chicago,

Ill.). The Transfer Function of Quarry Blast Noise and Vibration Into
Typical Residential Structures. Final rept. to U.S. Environmental
Protection Agency, Washington, D.C., Epa 550/9-77-351, February 1977,
43 PP•

18. Kryter, K. D. (Stanford Research Institute). Definition Study of the

Effects of Booms From the SST on Structures, People, and Animals.
Final rept. to Nat. Sonic Boom Evaluation Office, U.S. Air Force,
Washington, D.C., June 1966, 76 pp.; contract AF 49(638)-1696.

19. Kryter, K. D., P. J. Johnson, and J. R. Young (Stanford Research

Institute). Psychological Experiments on Sonic Booms Conducted at
Edwards Air Forch Base. Final rept. to Nat. Sonic Booms Evaluation
Office, U.S. Air Force, Washington, D.C., August 1968, 84 pp.;
contract AF 49(638)-1758.

20. Kundert, W. R. Everything You've Wanted To Know About Measurement r1icro-

phones. Sound and Vibration, March 1978, pp. 10-23.

21. Leventhal!, H. G., and K. Kyriakides. Environmental Infrasound: Its

Occurrence and Measurement. Ch. 1 in Infrasound and Low Frequency
Vibration, by w. Tempest. Academic Press, New York, 1976, pp. 1-18.

22. Olson, J. J., and L. R. Fletcher. Airblast-Overpressure Levels From

Confined Underground Production Blasts. BuMines RI 7574, 1971, 24 pp.

23. Peterson, A. P. G., and E. E. Gross, Jr. Handbook of Noise Measurement.

General Radio Co., Concord, Mass., 1972, 322 pp.

24. Reed, J. w. Atmospheric Attenuation of Expolsion Wave. J. Acoust. Soc.

Am., v. 61, No. 1, January 1977, pp. 39-47.

25. Schomer, P. D. Evaluation of C-Weighted Ldn for Assessment of Impulse

Noise. J. Acoust. Soc Am., v. 62, 1977, pp. 396-399.

26. Human and Community Response to Impulse Noise. Final rept. to

Ill. Inst. for Environmental Quality, IIEQ Doc. 78/07, Chicago, Ill.,
March 1978, 92 pp.; contract 80.109.

27. Predicting Community Response to Blast Noise. U.S. Army Corps of

Engineers, Civil Eng. Res. Lab., Tech. Rept. E-17, December 1973,
96 PP•

28. Siskind, D. E. Structure Vibrations From Blast Produced Noise: Energy

Resources and Excavation Technology. Proc. 18th U.S. Symp. on Rock
Mechanics, Keystone, Colo., June 22, 1977, pp. 1A3-l to 1A3-4.
34

29. Siskind, D. E., and V. J. Stachura. Recording System for Blast Noise
Measurement. Sound and Vibration, June 1977, pp. 20-23.

30. Siskind, D. E., V. J. Stachura, and K. S. Radcliffe. Noise and

Vibrations in Residential Structures From Quarry Production Blasting:
Measurements at Six Sites in Illinois. BuMines RI 8168, 1976, 17 pp.

31. Siskind, D. E., V. J. Stachura, M. S. Stagg, and J. W. Kopp. Structure

Response and Damage Produced by Airblast From Surface Mining. BuMines
RI 8485, 1980.

32. Siskind, D. E., and C. R. Summers. Blast Noise Standards and Instrumen-
tation. BuMines TPR 78, 1974, 18 pp.

33. Snell, C. M., and D. L. Oltmans. A Revised Impersonal Approach to Air-

blast Prediction. U.S. Army Waterways Experiment Sta., Explosive
Excavation Res. Off., Tech. Rept. 39, November 1971, 108 pp.

34. Sound and Vibration. S&V News. V. 12, No. 20, October 1978.

35. Stachura, V. J., and D. E. Siskind. Measurement of Airblast Produced by

Surface Mining. Proc. 51st Ann. Meeting, Minn. Sec. AI~lli, and
39th Ann. Min Symp., Duluth, Minn., Jan. 11-13, 1978, pp. 26-1 to
26-12.

36. Von Gierke, H. E. Guidelines for Preparing Environmental Impact

Statements on Noise. Rept. on Evaluation of Environmental Impact of
Noise, Comm. on Hearing, Bioacoustics, and Biomechanics, Assembly of
Behavioral and Social See., Working Group 69, Nat. Res. Council, Nat
Acad. Sci., Washington, D.C., 1977, 144 pp.

37. Windes, S. L. Damage From Airblast: Progress Report 1. BuMines RI

3622, 1942, 18 pp.

38. Damage From Airblast: Progress Report 2. B~1ines RI 3708, 1943,

50 PP·
39. Wiss, J. F., and P. W. Linehan (Wiss, Janney, Elstner, and Associates,
Inc.). Control of Vinration and Blast Noise From Surface Coal Mining.
Final rept. to the Bureau of Mines, Rept. WJE 75191, v. I, II, and III,
May 1978; contract J0255022.

40. Young, J. R. (Stanford Research Institute). l1easurement of the Psycho-

logical Annoyance of Simulated Explosion Sequences. Rept. to U.S. Army
Corps fo Engineers, Civil Eng. Res. Lab., Champaign, Ill., SRI Proj.
3160, February 1976, 36 pp.; contract Daca 23-74-C-0008.
 35

APPENDIX A.--CHARACTERISTICS OF 14 TYPICAL AIRBLASTS

A set of 14 typical airblasts and their frequency spectra were assembled

for illustrative purposes. A description is supplied to aid in the analysis
of the characteristics of different kinds of airblasts generated at coal
mines, quarries, and metal mines. Table A-1 lists the levels recorded at the
field sites and processed values obtained in laboratory tests.

TABLE A-1. -Test airblasts: field and laboratory measurements

Field measurements Laboratory measurements 1

r peak B&K 2209 GR 1933
Shot B&K C-slow, Linear 10-sec Flat True CSEL,
DP-7 2209 GR 1933 C-slow peak 1 C-slow time peak GR 1925/1926
constant
23/10 160 L1J1 159 L128 129 155.5 132
27/9 129 126 103 102 126 101.5 102.5 121 102
31/1 135 110 131 111 110 127 110
31/2 130 2107 127 107 108 124 108
31/3 132 2107 128 107 108 126 108
31/4 128 110 127 110 110.5 126 110
35/9 122 96 116 96 100 115 100
84/14 134 130 97 2107 131 108 107 126 108
101/1 121 120 )100 101 120 101 102.5 118 102.5
105/7 132 130 103 103 127 103.5 105 126 105
126/7 136 115 111.5 135 112 111.5 133 112
147/7 131 123 93 2g3 125 95 96 117 97
C5/12 139 2109 138 111.5 109.5 135 112.6
163/1 155 128 154 129 ~ 127
! 152 129
1Linear peak and flat peak are equivalent.
20verload light indicating on input overload.

Shot 23/10 (fig. A-1).--A very strong type I airblast that illustrates
the presence of energy of up to 35 Hz at which point it is only 20 dB below
the peak value. The blast has a very fast pressure rise time, with one peak
predominating because of a clay seam that had been loaded through. This is a
quarry blast on a 95-ft face with four holes each containing a long and short
deck.

(fig. A-1).--A type II airblast that illustrates the drop-off

of energy the 1-Hz peak. The energy at 6 Hz is about 25 dB below the
peak and continues to drop as the frequency increases. An interesting phenom-
enon is the reflection present in the latter third of the time history, at ~
0.95 sec after the highest peak. This represents the travel time across the
quarry to an opposing face and back to the gage. This phenomenon appears more
clearly in shot 105/7, shown later. This second airblast, or echo, is fil-
tered by the physical shape of the reflecting face; it may contain higher
frequencies (5 to 20 Hz) and cause a more severe structure response though
lower in peak value than the first arrival.
36

-----
0 0,5

TIME, sec
23/10

0
----- 0.5

60 ~~--~--~~-.~
Shot 23/10 50 Shot 2 7/7
40
30
20
10
~-L~QWWU~~~
20 40 60 80 100 0 20 40 60 80 100
FREQUENCY, Hz FREQUENCY, Hz
FIGURE A-1. - Quarry shots, 95-ft highwall: Type I airblast (shot 23/10) and reflected Type
II airblast (shot 27 /7).

Shots 31/1, 31/2, 31/3, 31/4 (fig. A-2).--In these time histories, a
quarry blast of two rows of 10 holes each was instrumented at four locations.
The direction of initiation was down the free face and toward the gage loca-
tion 31/1, with gage 31/3 located in the opposite direction. Gage 31/2 was
behind the shot on the top of the bench, and gage 31/4 was in front on the pit
bottom.

Shots 31/1 and 31/3 are similar in character except that the former
appears to be a time-compressed version of the other. This happens because
each successive borehole is closer to the gage station, shortening the travel
time and compressing the delay intervals. This also appears in the spectra as
a corresponding change in frequency content. Shot 31/1 has more energy above
20 Hz, while 31/3 continues to drop off gradually. Directing the initiation
away from a structure could change the frequency content sufficiently to
reduce problems.

In Shot 31/4, the individual pulses from the 10 front holes occur in 30-
to 45-msec intervals. This corresponds to the energy present in the spectra
 37

0 0.5

Tl ME, sec

0 0.5

TIME, sec
~~

~~~~ 0.5

~3-J~~~\~~~~ TIME, se_,-_c ~O-.S

31/4
----- Tl ME, sec

Q) 60 ··~---. ·.--.- ,-,- ~

60
w "0~ 50 Shol 31/1 50 Shol 31/2
> 0w 40 40
I- :J 30 30
<J:
_j ~ 20 20
w _.J 10
n:: CL 10
~
<J: 0 20 40 60 80 100
0 20 40 60 80 100
~ 60 60
w .. so Shot 31/3 50 Shot 3!/4
>~40 40
f- 30
:J 30
'Sf- 20 20
~0: 10 10
~
<(
0 20 40 60 80 100 0 20 40 60 80 100
FREQUENCY, Hz FREQUENCY, Hz
FIGURE A-2.- Quarry shots: airblast, initiation towards gage station (shot 31/1); Type II
airblast (shot 31!2); airblast initiation away from gage station (shot 31/3);
and Type I airblast (shot 31/4).
38

at 22 to 35 Hz, which is only about 12 dB down from the peak. There is more
mixing of pressure pulses in shot 31/2 than in shot 31/4 because of the prox-
imity of the second row of holes and the absence of a direct free face. More
energy is directed upward rather than back toward the gage. The energy in the
22- to 35-Hz range for shot 31/2 is 20 dB or more below the peak value, in
contrast to that in shot 31/4.

Shot 35/9 (fig. A-3).--This is a very long iron mine blast of 507,000
pounds of explosives, with a total of 147 delays in 148 holes. The distance
to the gage station was about 3,400 ft, and the shot duration was about 3.1
sec.The spectra shows a 22-Hz peak that is only 20 dB below the low-frequency
peak.

Shot 84/14 (fig. A-3).--This is a highwall coal mining shot with large
blast holes (15.5 in) monitored at about 750 ft. The first arrival is the
vertical motion of the ground near the microphone, since ground vibration
travels faster than airblast. The airblast arrived about 650 msec later and
showed the much sharper gas and stemming release pulses. Even though the time
0 0.5
TIME, sec

0 0.5
TIME, sec

0
----- TIME. sec
0.5

101-1

~ 60 r--1..........,.--,--,--,--.---~.,--,---,
l
~.~
;::: 50 Shot 35/9 60
::::; 40
Q
::;; 30
<l: 20 .

~ 10 ~~~~~~~~~_j 10 20 30 40 50
5
w
0 20 40 60 80 100 FREQUENCY, Hz

a:: FREQUENCY, Hz
FIGURE A-3. - Metal mine shot, very long duration oirblast (shot 35/9); coal mine shot, Type
II airblast, Iorge holes (shot 84/14); and coal mine shot, long-duration airblast,
six decks (shot 101/1).
 39

history shows many sharp spikes, the frequency spectra indicate a continual
drop in energy to about 5 Hz, owing to the randomness of the gas and stemming
release pulses that prevented a buildup of energy at any particular frequency.

Shot 101/1 (fig. A-3).--In this shot the top decks were probably 60 msec
apart, which generated a large amount of energy at 17Hz, about 13 dB below
the peak value. The relatively uniform series of pulses caused the energy to
build at ·this frequency and shook the house at this site strongly.

Shot 105/7 (fig. A-4).--A reflection is obvious about 730 msec after the
first airblast arrival in this quarry shot. This "echo"contains energy in the
18-to 22-Hz region and shows up strongly in the spectra about 12 dB below the
peak. The house at this location responded more to the second airblast
because of its frequency content.

0 0.5
TIME, sec

0 0.5

Tl ME, sec
126/7

0 0.5

TIME, sec
147/7

60 ~~~~~~~~~
60 ~~~~~~~~~

50 50 Shot 147/7
Shot 105/7
40 40
30 30
20 20
10 10
20 40 60 80 I 00 0 40 80 120 160 200 0 20 40 60 80 100
FREQUENCY, dB FREQUENCY, Hz FREQUENCY, Hz
FIGURE A-4. - Quarry shot, Type II airblast initial arrival, Type I airblast when reflected
(shot 105/7); coal mine shot, parting, Type I airblast (shot 126/7); and metal
mine shot, Type II airblast (shot 147/7).
40

Shot 126/7 (fig. A-4).--This is a coal mine "parting" shot in a thin (6-
ft) layer of limestone. The peak in the spectra is at 20 Hz. A lack of con-
finement caused the generation of considerable energy in the 20-to 50-Hz
range. The home responded vigorously.

Shot 147/7 (fig. A-4).--At greater distances most high-frequency energy

is despersed, as shown in this iron mine blast, recorded at a distance of
7,000 ft. The structure at this location responded weakly.

Shot CS/12 (fig. A-5).--This is a coal mine highwall blast in which a

hole blew out. The single sharp pulse has much energy at 10 Hz and above, as
shown in the frequency spectra. The 10 Hz point is only 10 dB below the high-
est point.

Shot 163/1 (fig. A-5).--A partial misfire with uncovered detonating cord
is illustrated on this large iron mine shot monitoring at about 700 ft. A lot
of high-frequency energy was generated by the detonating cord. The gage,
located in the near field, caused the source to appear to move rather than to
act as a single point.

C5/12
-----
0
TIME, sec
0.5

0 0.5

TIME, sec

(D 60 60 .----r---.--.--~--,
w w
.
"C
50 Shot C5/12 50 Shot 163/1
> 0 40 40
1- :::> 30 30
<t 1- 20 20,..
_J
w _J 10 10
c:: 0..
::!!
<t 0 20 40 60 80 100 0 20 40 60 80 100
FREQUENCY, Hz FREQUENCY, Hz

FIGURE A-5. • Coal mine shot, airblast from a blowout (shot CS/12), and metal mine shot,
airblast from a partial misfire, exposed detonating cord (shot 163/l).
 41

APPENDIX B.--DATA FOR MEASUREMENT TECHNIQUE COHPARISONS

TABLE B-1. - C-slow to CSEL comparison data

Shot C-slow CSEL

X, dB X, psi Y, dB Y, psi
C-5 112.0 o. 001155 112.0 0. 001155
C-4 109.0 .000817 108.0 .000728
C-10 106.0 .000579 108.0 .000728
351 97.0 .000205 97.0 .000205
36N 88.0 .000073 88.0 • 000073
148KK 97.0 .000205 95.0 .000163
1460 93.0 .000130 87.0 .000065
150 93.0 .000130 91.0 .000103
45T 94.0 .000145 93.0 .000130
62 99.0 .000258 99.0 .000258
64 100.0 .000290 100.0 .000290
67 100.0 • 000290 99.0 .000258
70 97.0 .000205 99.0 .000258
71 99.0 .000258 100.0 .000290
78 101.0 .000325 101.0 .000325
79 106.0 .000579 105.0 .000516
80 103.0 .000410 104.0 .000460
81 102.0 .000365 103.0 .000410
84 116.0 .001830 116.0 .001830
85 106.0 .000579 107.0 .000649
86 107.0 .000649 107.0 .000649
90 102.0 .000365 102.0 .000365
92 102.0 .000365 103.0 .000410
94 103.0 .000410 103.0 .000410
95 101.0 .000325 100.0 .000290
99 94.0 .000145 97.0 .000205
100 99.0 .000258 97.0 .000205
101 103.0 .000410 102.0 .000365
103REFL1 98.0 .000230 99.5 .000274
103W 107.0 • 000649 108.0 .000728
105WDTR 2 101.0 .000325 101.0 .000325
126 113.0 .001295 113.0 .001295
128 102.0 .000265 102.0 .000365
130 107.0 .000649 108.0 • 000728
141 100.0 .000290 100.0 .000290
154 93.0 .000130 97.0 .000205
161 112.0 • 001155 110.0 .000917
Reflected airblast or echo.
ll

2Direct airblast or first airblast arrival.

42

TABLE B-2. -Comparison of measurement techniques to 0.1-Hz linear peak

air blasts

Heasurement technique Correlation Standard Equation

coefficient deviation
Coal mine shots:
High wall shots:
2 Hz peak •• .•••••••••• 0.953 31.9% +2.4dB, -3.3dB Y=O. 740X
5 Hz peak ••• •••••••••• .951 32.0% +2.5dB, -3.5dB Y= .662X
C-slow . ..••........... • 911 45.6% +3.3dB, -5.3dB Y= .064X
PLdB •••••••••••••••••• .785 81.2% +5.2dB,-14.5dB Y= • 014X
Parting shots:
2 Hz peak •••• ••••••••• • 986 17.5% +1.4dB, -1. 7dB Y=l. 014X
5 Hz peak . ...•......•• .985 18.0% +1.4dB, -1. 7dB Y= .885X
C-slow . ............... .940 37.3% +2.8dB, -4.1dB Y= .095X
PLdB • ••.••••. • · • • • • • • · .978 24.7% +1.9dB, -2.5dB Y= .016X
Assorted shots:
2 Hz peak .. .......••.. .981 20.5% +1.6dB, -2.0dB Y= .856X
5 Hz peak .....•... •... .991 15.5% +1.3dB, -1. 5dB Y= .637X
c- slow . ............... .936 39.9% +2.9dB, -4. 4dB Y= .052X
PLdB • ••••••••••• • • • • • • .698 114.7% +6.6dB,-16.7dB Y= .015X
Quarry shots:
2 Hz peak • •••••••••••••• .893 52.2% +3.6dB, -6.4dB Y= .601X
5 Hz peak . •••••••••••••• • 937 38.3% +2.8dB, -4.3dB Y= • 535X
C-slow . .........•.•..... • 931 41 • 1% +3 • 0 dB , -4.6dB Y= .044X
PLdB • .••••••••.•.••••••• .959 32.4% +2.4dB, -3.4dB Y= .013X
Metal mine shots:
2 Hz peak . ...........•.. • 972 25.9% +2.0dB, -2. 6dB Y== .571X
5Hz peak • •••••••••••••• • 949 36.2% +2.7dB, -3.9dB Y= .438X
C-slow .. ..•....•........ .887 54.4% +3.8dB, -6.8dB Y= .030X
PLdB • •..•...••••••••.•.• .872 62.8% +4.2dB, -8.6dB Y= .004X
All shots:
2Hz peak.• •••••••••••••• .947 34.2% +2.6dB, -3. 6dB Y= .750X
5 Hz peak • •••••••••••••••• • 946 34.5% +2.6dB, -3.7 dB Y= .666X
C-slow . .........•....•.... .890 51.3% +3.6dB, -6. 2dB Y= .063X
PLdB • ••••••••••••••••••• • • .783 80.6% +5.1dB,-14.2dB Y= .013X
 43

APPENDIX C.--EQUIPMENT USED IN TESTS

TABLE C-1. - Precision sound level meters tested by the Bureau of Mines

Bruel and Kjaer 2209 General Radio

impulse precision 1933
Microphone type •••••••••• 1- and 1/2-in. condenser •••••••••• 1- and 1/2-in.
Frequency response, electret-
±3 dB, Hz: condenser.
l-in. microphones •••••• 2-30k, selectable to 6-20k •••••••• 5-15
1/2-in. microphones •••• 3-l/2-40k, selectable to 5-l/2-40k 5-24
Maximum dynamic range,
dB:
l-in. microphones •••••• 140 130
1/2-in. microphones •••• 140 140
Weightings:
l-in. microphones •••••• A, B, c, D, Linear •••••••.•.•••..• A, B, c, Flat.
1/2-in. microphones •••• A, B, c, D, Linear ••••.•.••...•••• A, B, c, Flat.
Dimensions, em (in.):
Length ••••••••••••••••• 33 (14.8) 22.9 (9)
Width •••••••••••••••••• 12 (4.75) 15.8 (6.19)
Height................. 9 (3.5) 7.6 (3)
Notes •••••••••••••••••••• "Hold" feature on peak, Battery check,
battery check. available with
nicad batteries
and charger.
10utput method for all: Meter reading or analog waveform; all with fast,
slow, impulse, and peak response time.
 TABLE C-2. - Seismographs with airblast channels tested by the Bureau of Mines -1:-
-l:-

1
Dallas Instruments, Inc. Phillip R. Berger VME-Nitro Consult, Vibra-Tech Engineers,
and Associates, Inc., Evanston, Ill. Inc., Hazelton, Pa.
Bradfordwood~ Pa.
Model • •...••.....•••...• St-4 (Vi-Sel Monitors, Safeguard Seismic Model F sound vel- Vibra-tape 1,000 and
Inc., Dallas, Tex.; Unit II. ocity recorder. 2000 series.
EverLert Vibra-Tape
(Vibra-Tech); White Seismo
Sentinel (White Engi-
neering Inc., Joplin, Mo.).
Microphone type ••••••••• 1-1/8-in. ceramic ••••••••••• l-in. ceramic ••••••• l-in. ceramic ••••••• 1-1/8-in. ceramic.
Output method ••••••••.•• Records FM analog on mag- Light beams on Light beams on Records FM analog on
etic tape; separate play direct-write photo- direct-write photo- magnetic tape; anal-
back system for recorded graphic paper. graphic paper. ysis of playback
digital level and analog available from
waveform; analysis avail- Vi bra-Tech.
able from distributors.
Frequency response, Hz •• 5-200 on analog waveform, 3-1/2-200 2-2,000 5-5,000 on 1000 series,
±3 dB nominal. 5-500 on digital. 4-1,000 on 2000 series.

Maximum dynamic dB •• I 137 on tape, 140 on digital 128-148 on paper 140 137 on 1000 series,
range. readout. tape, switchable. 137 or 147, switchable
on 2000 series.

Power source •••••••••••• IRechargeable 12-volt Rechargeable 12- Rechargeable 12- 3- to 6-volt lantern
battery. volt battery. volt battery. battery (1000 series).
2- to 6-volt lantern
battery (2000 series).
Weight ......... kg (lb) .. ll3.6 (30) 12.3 (27) 18.1 (40) 11.6 (26), 1000 series,
12.3 (27), 2000 series.
Dimensions, em (in.):
Length •••••••••••••••• l 52.1 (20.5) 45.8 (18) 49.2 (19.4) 38.1 (15), 1000
43.2 (17), 2000.
Width ••••••••••••••••• 25.4 (10) 35.6 (14) 24.8 (9.8) 33.0 (13)
34.9 (7.3).
Height ................ 117.8 (7) 15.3 (6) 22.9 (9) 19.0 (7.5)
18.4 (7 .3).

Notes ••••••••••••••••••• IOperates up to 1 month Self-contained unit Self-contained unit Other dynamic ranges
unattended; in a standby in aluminum fin- in carry case. optional; electro-
mode, records 4-channel ished briefcase. magnetically shielded
event when vabration case.
exceeds a preset threshold.
1Multiple distributers, see model.
 45

TABLE C-3. - Long-term monitors tested by the Bureau of Mines 1

VME-Nitro consult, Inc., Dallas Instruments,

Sound-tector model 1800 model AR-2
Microphone type ••••••••• l-in. ceramic •••••••••.•••• 1-1/8-in. ceramic.
Frequency response, Hz •• 2-2,000 5-8,000
±3 dB nominal.
\faximum dymanic dB •• 140 plus 8 over range Up to 150 optional.
range. capability.

Weightings •••••••••••••• A, C, Flat ••••••••••••••••• A, B, C, Flat.

Response time ••••••••••• Fast rms and impulse peak. Slow rms and peak impulse.
Power source •••••.••••.• 110 or 220 volts ac; 12 3- to 6-volt lantern battery
volts de. or 12-volt car battery.
Weight ••••••••• kg (lb) •• 10 (22) 10.4 (23)
Dimensions, em (in.):
Length .••....•.....••. 34.3 (13.5) 36.8 (14.5)
Width .•••••• .•.•.••• •. 27.9 (11) 25.4 (10)
Height • ......•..••...• 26.7 (10.5) 17.8 (7)
Notes •••••• • • • • • • • • • ·, • • Operates for 30 days con- Can operate for 1 month con-
tinuously on ac power, 15 tinuously; type II tolerance.
days on batteries; type
S-1 tolerance; A- and C-
weighting conform to
type I ANSI specifications
below 2,000 Hz.
1common features: Output is a bar graph on pressure-sensitive paper, dot printed.

TABLE C-4. -Wide-band, research-type instrumentation tested by the Bureau of M1nes 1

Bruel and Kjaer, model 2631 . Validyne Engineering Corp.,

model DP-7/cD~l6
Transducer type •••••••••• l-in. and 1/2-in. condenser Variable reluctance differ-
microphones. ential pressure transducer.
Frequency response, Hz •• 0.1-8,000 (for 1-in.), 0.1->380. 1
±3 nominaL 0.02-16,000 (for 1/2-in.).
0.02-16,000 (for 1/2-in.).
Maximum dynamic dB •• 162 177.
range.
Power source ••.••••.••••• 100-240 volts ac ••••••••••• 22-35 volts de.
Weight •••••••••• kg (lb) •• 2 (4.3) 1.5 (3.3).
Control unit dimen-
sions, em (in.):
Length ................. 20.0 (7.9) 25.4 (10).
Width .................. 6.1 (2.4) 10.2 (4).
Height................. 13.3 (5.2) 6.4 (2.5).
Notes •••••••••••••••••••• "Sonic boom"microphone car- Adapted from gas line
ier system. pressure systems.
1Adjustable, requires additioi of needle valve for acoustic use.
46

APPENDIX D--LOW-FREQUENCY ROLLOFF OF TESTED EQUIPMENT

The following table D-1 lists the approximate low-frequency rolloff of

typical airblast measurement instruments. Since airblast frequencies often
fall in the region of the rolloff for 2-Hz and 6-Hz instruments, a measurement
difference between instruments can occur. The rolloff is listed in dB/decade
and dB/octave and can be used to estimate possible reading errors between
instruments. It is hoped that manufacturers will adopt a standard low-
frequency rolloff in the interest of consistency.

TABLE D-1. -Low-frequency rolloff of tested equipment

Instrument dB/decade dB/octave -3dB frequency, Hz

6-Hz instruments:
B&K 2209 1 in (10Hz lower limit) ••• 33 10 6
B&K 2209 0.5 in (10Hz lower limit). 45 13.5 5.5
B&K 2209 direct (10 Hz lower limit). 18 5.5 7
B&K 2209 0.5 in (2Hz lower limit) •• 26.7 8 3.5
Berger seismograph •••••••••••••••••• 30 9 4.5
ST-4 . •.••.••.•....••.••.•.•..•.•••.. 21 6.3 4
Vibra-Tape 1000 ••••••••••••••••••••• 22.5 6.8 5
GR 1933 0.5 in electret ••••••••••••• 35 '9. 5 5
GR 1933 1 in electret ••••••••••••••• 22 6.5 5
GR 1933 2 in ceramic •••••••••••••••• 39.5 12 5
AR-2 acoustic monitor ••••••••••••••• 23 7 4
2-Hz instruments:
B&K 2209 1 in (2Hz lower limit) •••• 42.5 12.5 1.8
VME "F" seismograph ••••••••••••••••• 55 16.5 1.7
0.1-Hz instruments:
B&K 2631 1 in ••••••••••••••••••••••• 24.5 7.5 •1
B&K 2631 0. 5 in .................... . 21.3 6.5 • 02
Validyne DP-7 • ..••...•.•.•.••.....•• 30 9 .09
 47

APPENDIX E.--}1ICROPHONE TYPES SUITABLE FOR AIRBLAST MEASUREMENT

There are three basic types of microphone construction; air condenser

(condenser), electret-condenser (electret), and ceramic (piezo-electric).

Air-condenser microphones are built so the diaphragm is one plate of a

capacitor. The motion of the diaphragm varies the capacitance at a rate pro-
portional to the change in air pressure. A potential (polarization voltage)
across the capacitor maintains a constant charge in this system and produces a
voltage inversely proportional to the change in capacitance for the duration
of a time constant (20-~).

The electret-condenser microphone is similar to the air-condenser except

for the method of maintaining a constant charge. An electret material is con-
structed with a permanent charge and is used along with an air gap as part of
the dielectric to form a capacitor. No polarization voltage is necessary.
The capacitance is measured across this dielectric between the backplate and a
conductive diaphragm (20-~).

Ceramic microphones have a connecting rod between diaphragm and piezo-

electric material. A voltage is generated across this material when it is
deformed by a pressure change (20-21 •

"Random" incidence microphones are for use in a diffuse field; that is,
one made up of many reflected waves from all directions that combine to form a
uniform energy-density level. This feature makes such microphones suitable
for indoor use or in other areas with many reflections. At frequencies below
1 kHz, its response is the same as a grazing or perpendicular incidence micro-
phones. "Grazing" and "perpendicular" incidence re.fer to the preferred angle
at which the pressure wave strikes the diaphragm, 90 degrees and 0 degrees,
respectively. The angle is the orientation needed to obtain the best
response. Microphones may also be said to have "free field" response, which
means all sound of interest arrives from one direction with no reflections
such as those found in an enclosed space. Grazing incidence response is also
called pressure response. Any of these three microphones (condenser, elec-
tret, and ceramic) is appropriate since their directional sensitivity is
indistinguishable at airblast frequencies (1, 20, 23).
48

APPENDIX F.--RMS PROCESSING SYSTEM

The rms detector.used to determine weighted-energy levels was the GR

1926, shown in figure F-1. For both airblast and ground vibration signals,
equipment was assembled to record and process them quickly and efficiently
(fig. F-2). The original blast time histories were recorded on F~1 tape.

FIGURE F-1. - RMS detection system.

 49

FM tape
recorder

... Channels 5, 6, 7
' I 26 dB amp! ifier
1;3 octave
multifi Iter --
Dual Hr/LO
fi Iter
'
rms
detector

\ \ I

-
Amplifiers

-c -cc
c
·-C'l
1/)
-cc ·-C'l
1/)
"'C C'l
Q)
·- "'C
,_
Q)

-
1/)
,_
"'0
0 Q)

-tJ...
1/)
(.)
Q)
0::
E
,_ ·-
\~ 'I \I

Oscillograph

FIGURE F-2. • RMS detection system block di.agram.

50

From the tape recorder, they were first fed into a General Radio Type 1925
Multifilter, which is a spectrum shaper and equalizer and which has ten 1/3-
octave-band, six-pole, active Butterworth filters. These filters range in
center frequency from 3.15 Hz to 2.5 kHz. There is a set of calibrated 1-
dB/step attenuators, one for each of the 30 channels can be output separately
or summed, and internal filters are also available to obtain A-, B-, and C-
weighting for airblast noise analysis (1)·

For the required application, two modifications were made to the filter.
A special low-frequency band was added with a center frequency of 1.6 Hz.
This modification lowered the upper limit of the frequency range to 1.25 kHz.
Then, the filter that was in channels 5, 6, and 7 was replaced by a 26-dB gain
amplifier (fig. F-3). Cables were run from this printed circuit board card to
the housing to allow the introduction of an external filter. This filter
could then be adjusted to a frequency range not available on the Multifilter
itself, or to bypass the 1/3-octave filtering characteristics of the Multi-
filter. This externally filtered signal could then be accessed by using the
channels available on the dummy card. An externally filtered version of the
input signal could then be introduced without providing a separate interface
to the rest of the system. The 26-dB gain was necessary because that gain is
present in the other 1/3-octave-band filters.

The external filter was Rockland 1022F Dual Hi/Lo filter consisting of
two identical, independent filters mounted in the same case, with separate
input and output terminals. This can be set to act as either a high-or a low-
pass filter, with either RC or 4th-order (24-dB/octave) Butterworth character-
istics. The cutoff frequency was digitally selected by a dial on the front of
the instrument, and the dynamic range was 0.1 Hz to 111 kHz. The filter was
used in a band pass configuration accomplished by setting the input in the
high-pass mode and cascading it with the second in the low-pass mode, and then
selecting the frequency range of interest; the 4th-order Butterworth charac-
teristics were used on both. The bandpass-conditioned signal from the Rock-
land was fed into the Multifilter, and the signal was available on its
channels 5-7.

10 KU

470 ,Q

21167
Signal
input
o~----jf---1~--.---i
180 fLf

-II VDC +7 VDC

FIGURE F-3.- Amplifier modification for Multifilter.

52

Once this range has been determined, a reference signal is needed. In

all cases, a calibration signal was recorded on the FM tape, just before or
after the data were recorded. This was done to take into account electronic
variances owing to temperature, humidity, etc. This calibration was then fed
into the detector, and the input level was adjusted with the GR 1925 Multi-
filter input signal attenuator until the rms value of the calibration signal
equaled the maximum band level setting, as shown by a Nixie-tube digital dis-
play of the band level energies. The airblast was then run through the system
and the output recorded on the oscillograph. Figure F-4 shows a processed
airblast. The top trace is of the rms levels of each integration period. The
data are displayed in the manner of the expanded view shown in figure F-5.
The output scale is linear over the 60 dB range; therefore, the rms level of
the signal can be measured by dividing the height of the signal output by the
full-scale output and taking this ratio times 60 dB. The final level is

Full-scale value

154/4b Section processed RMS output

for the RMS value

0 2

TIME, sec

154/4 Unfiltered a i rblast

154/4a Filtered airblast

FIGURE F-4. - Typica I RMS output from an a irblast.

 51

The GR 1926 rms detector was then directly interfaced to the GR 1925 Mul-
tifilter. The detector measured the rms value of each band of the Multifilter
selected. Any range of bands can be scanned, and the rms value of each
computed (..!:.2_).

The GR 1926 rms detector has a 70-dB dynamic input range and a 60-dB
dynamic output range. The input signals from the multifilter are sampled with
integration times of 1/8, 1/4, 1/2, 1, 2, 4, 8, 16, or 32 sec, depending on
the nature of the signal being processed. The number of samples and sampling
rate depend on the integration time chosen, as shown in table F-1. For most
frequencies the sampling rate was below the Nyquist rate. Since reconstruc-
tion of the input signal was not required, this is of no consequence. The
measurements of the rms energy of each band can be made accurately at a low
sampling rate.

TABLE F-1. - Integration-time parameters for GR 1926 rms detector

80-pct confidence
Integration Number of Average sampling Average sample limits for indepen-
time, sec samples rate, samples/sec spacing, msec dent samples
(noise input, ±dB)
1/8 128 1,024 0.781 o.s
1/4 256 1,024 .813 .34
1/2 512 1, 024 .877 .24
1 1, 024 1,024 1.00 .17
2 1, 024 512 2.01 .17
4 1, 024 256 4.02 .17
8 1,024 128 8.04 .17
16 1, 024 64 16.08 .17
32 1,024 32 32.2 .17

The GR 1926 rms detector has both analog and digital outputs, and can be
interfaced to a PDP 11 computer, storage oscilloscope, printer, or X-Y
plotter. The output was fed into a Bell & Howell 5-135 oscillograph.

To find the best correlation between rms energy levels of structure

response and excitation, a sample was chosen that was considered representa-
tive of the many channels of recorded data. Samples of airblast, ground
vibrations, corner, and midspan structure responses were chosen and run
through the rms detector system with integration times of 1/8, 1/4, 1, 2, and
4 sec.

To process an airblast signal through the detector, it is necessary to

determine its peak level in decibels and then adjust the maximum band level
control. This control sets the limits for 60-dB dynamic output range. For
example, a maximum band level setting at 130 dB will measure the part of the
signal between 70 and 130 dB. In other words, the level specifies the.high
end of the 60-dB dynamic output range.
 53

found by adding the floor level; that is, the maximum band level minus 60 dB,
which could be done for each integration period. For our purposes, however,
only the peak energy level was of interest; it alone was measured.

:--

'
T
60 dB
'
22.1 dB
w

I"'
I

tRMS output channel 7

RMS output channel 6
-RMS output channel 5
Minimum value channel 4
Full scale, channel 3

0 5
p-r---
~- ---
--
TIME, msec
~
10

FIGURE F-5. - Expanded output of rms detector.

>'IU.S. GOVERNMENT PAINTING OFFICE: 1981-703.002/17 INT.·BU.OF MINES,PGH.,PA. 25171