EN-21
The human perception of a sound depends not only on the sound pressure level, but
also on the frequency. At high or low frequencies, a sound is felt to be less loud than a
sound of equal level in the midrange. The frequency weighting A compensates for this
effect and produces measurement results which are close to the perceived sound level.
For this reason, this type of frequency weighting is widely used for purposes such as
sound level evaluation.
The frequency weighting C curve produces almost flat response, but with a roll off below
31.5Hz and above 8000Hz. This is suitable for sound pressure level measurements in
situations with unwanted low-frequency or high-frequency components.
18. APPENDIX B RMS DETECTION CIRCUIT AND TIME WEIGHTING
The sound level meter uses rms detection. The effective value E (rms) is defined by the
following equation.
E rms
T
e dt
T
(
)
=
z
1
2
0
The voltage e which changes over time is raised to the power of 2, and integration for
the time interval T is performed. The result is divided by T and the square root is
extracted. The circuit configuration for performing the above mathematical operation
looks as follows.
ei
Power of 2
ei
2
1
0
T
T
z
ei
2
ei
2
E(rms)
Input Voltage Output Voltage
During sound level measurements, the level often fluctuates drastically, which would
make it difficult to evaluate readings if some kind of averaging were not applied. Sound
level meters therefore provide the capability for index weighting (index averaging)
using the rms circuit. The parameters of this weighting process are called the time
weightings, determined by the time constant (see next page).
Sound level meters usually have a F(Fast) and S(Slow) setting for the time weighting.
The time range that is considered for averaging is narrow in the F(Fast) setting and
wide in the S(Slow) setting. In the F(Fast) setting, the instantaneous level has a larger
bearing on the displayed value than in the S(Slow) setting. From the point of view of the
measurement objective, the F(Fast) setting is more suitable to situations with swiftly
changing sound level, whereas the S(Slow) setting yields a more broadly averaged
picture. The F(Fast) setting is more commonly used, and sound pressure level values
given without other indication are usually made with F(Fast) characteristics.
Time weightings and time constant
Time constant
Time
Weightings
Rise time
Decay time
F(Fast)
125ms
125ms
S(Slow)
1s
1s
