background image

783 

Federal Aviation Administration, DOT 

Pt. 36, App. A 

the output of the analyzer without slow 
time-weighting, slow time-weighting must 
be simulated in the subsequent processing. 
Simulated slow time-weighted sound pres-
sure levels can be obtained using a contin-
uous exponential averaging process by the 
following equation: 

L

s

(i,k) = 10 log [(0.60653) 10

0.1

Ls[i,

(

k

¥

1

)

]

(0.39347) 10

0.1 L

(

i, k

)

where L

s

(i,k) is the simulated slow time- 

weighted sound pressure level and L(i,k) 
is the as-measured 0.5s time average 
sound pressure level determined from the 
output of the analyzer for the k-th in-
stant of time and i-th one-third octave 
band. For k = 1, the slow time-weighted 
sound pressure L

s

[i, (k 

¥ 

1 = 0)] on the 

right hand side should be set to 0 dB. An 
approximation of the continuous expo-
nential averaging is represented by the 
following equation for a four sample 
averaging process for k 

4: 

L

s

(i,k) = 10 log [(0.13) 10

0.1 L[i,

(

k

¥

3

)

]

+ (0.21) 10

0.1

 

L[i,

(

k

¥

2

)

]

+ (0.27) 10

0.1 L[i,

(

k

¥

1

)

]

+ (0.39) 10

0.1 L[i,

 

k]

where L

s

(i, k) is the simulated slow time- 

weighted sound pressure level and L (i, k) 
is the as measured 0.5s time average 
sound pressure level determined from the 
output of the analyzer for the k-th in-
stant of time and the i-th one-third oc-
tave band. 

The sum of the weighting factors is 1.0 in 

the two equations. Sound pressure levels cal-
culated by means of either equation are valid 
for the sixth and subsequent 0.5s data sam-
ples, or for times greater than 2.5s after ini-
tiation of data analysis. 

N

OTE

: The coefficients in the two equa-

tions were calculated for use in determining 
equivalent slow time-weighted sound pres-
sure levels from samples of 0.5s time average 
sound pressure levels. The equations do not 
work with data samples where the averaging 
time differs from 0.5s. 

A36.3.7.6 The instant in time by which a 

slow time-weighted sound pressure level is 
characterized must be 0.75s earlier than the 
actual readout time. 

N

OTE

: The definition of this instant in 

time is needed to correlate the recorded 
noise with the aircraft position when the 
noise was emitted and takes into account 
the averaging period of the slow time- 
weighting. For each 0.5 second data record 
this instant in time may also be identified as 
1.25 seconds after the start of the associated 
2 second averaging period. 

A36.3.7.7 The resolution of the sound pres-

sure levels, both displayed and stored, must 
be 0.1 dB or finer. 

A36.3.8 

Calibration systems. 

A36.3.8.1 The acoustical sensitivity of the 

measurement system must be determined 
using a sound calibrator generating a known 

sound pressure level at a known frequency. 
The minimum standard for the sound cali-
brator is the class 1L requirements of IEC 
60942 as amended (incorporated by reference, 
see § 36.6). 

A36.3.9 

Calibration and checking of system. 

A36.3.9.1 Calibration and checking of the 

measurement system and its constituent 
components must be carried out to the satis-
faction of the FAA by the methods specified 
in sections A36.3.9.2 through A36.3.9.10. The 
calibration adjustments, including those for 
environmental effects on sound calibrator 
output level, must be reported to the FAA 
and applied to the measured one-third-oc-
tave sound pressure levels determined from 
the output of the analyzer. Data collected 
during an overload indication are invalid and 
may not be used. If the overload condition 
occurred during recording, the associated 
test data are invalid, whereas if the overload 
occurred during analysis, the analysis must 
be repeated with reduced sensitivity to 
eliminate the overload. 

A36.3.9.2 The free-field frequency response 

of the microphone system may be deter-
mined by use of an electrostatic actuator in 
combination with manufacturer’s data or by 
tests in an anechoic free-field facility. The 
correction for frequency response must be 
determined within 90 days of each test series. 
The correction for non-uniform frequency re-
sponse of the microphone system must be re-
ported to the FAA and applied to the meas-
ured one-third octave band sound pressure 
levels determined from the output of the an-
alyzer. 

A36.3.9.3 When the angles of incidence of 

sound emitted from the aircraft are within 

±

30

° 

of grazing incidence at the microphone 

(see Figure A36–1), a single set of free-field 
corrections based on grazing incidence is 
considered sufficient for correction of direc-
tional response effects. For other cases, the 
angle of incidence for each 0.5 second sample 
must be determined and applied for the cor-
rection of incidence effects. 

A36.3.9.4 For analog magnetic tape record-

ers, each reel of magnetic tape must carry at 
least 30 seconds of pink random or pseudo- 
random noise at its beginning and end. Data 
obtained from analog tape-recorded signals 
will be accepted as reliable only if level dif-
ferences in the 10 kHz one-third-octave-band 
are not more than 0.75 dB for the signals re-
corded at the beginning and end. 

A36.3.9.5 The frequency response of the en-

tire measurement system while deployed in 
the field during the test series, exclusive of 
the microphone, must be determined at a 
level within 5 dB of the level corresponding 
to the calibration sound pressure level on 
the level range used during the tests for each 
one-third octave nominal midband frequency 
from 50 Hz to 10 kHz inclusive, utilizing pink 
random or pseudo-random noise. Within six 
months of each test series the output of the 

VerDate Sep<11>2014 

12:50 Apr 30, 2019

Jkt 247046

PO 00000

Frm 00793

Fmt 8010

Sfmt 8002

Y:\SGML\247046.XXX

247046

spaschal on DSK3GDR082PROD with CFR