Civil Engineering Reference
In-Depth Information
A major problem performing these comparisons is the availability of a complete set
of specifications for the measured object. Material data as well as dimensions may not be
completely described. Presumably, the loss factor is the most critical parameter. A
prediction of the sound reduction index in the frequency range around the critical
frequency and above will be quite uncertain without this information.
80
Concrete - 120 mm
Steel - 1 mm
70
60
50
40
f
c
30
20
10
0
63
125
250 500 1000 2000 4000
Frequency (Hz)
Figure 6.23
Sound reduction index of a 1 mm steel panel and a 120 mm concrete wall. Dashed lines: calculated
data from Equations
(6.103).
The critical frequency for the concrete wall is indicated. Measured data from
Homb et al. (1983).
Two examples on such a comparison are shown in
Figure 6.23
, measured and
predicted sound reduction index of a 1 mm thick steel panel and a 120 mm thick concrete
wall. In the first case we find that the critical frequency is approximately 12 kHz, making
the panel mass controlled in the whole measuring range. The fit between measured and
calculated data is very good. As for the 120 mm concrete, the fit around the critical
frequency between these data is rather poor. However, no measured data for the loss
factor were available making it necessary to use an estimate from Equation
(6.75)
.
Results from reproducibility tests, comparing results from different laboratories
measured on the same specimen, have shown that the reproducibility standard deviation
becomes very large if the loss factor is not properly controlled. The prediction accuracy
is also not very good around the critical frequency. Measured data on thick and massive
walls normally exhibit a more or less constant “plateau” in the reduction index curve, as
opposed to thin panels where there is normally a distinct “dip” in the curve.
