Overall Sound Pressure Level between two Frequencies

Click For Summary
SUMMARY

The discussion focuses on calculating the overall sound pressure level (SPL) between 31.5 Hz and 8000 Hz, with an initial band level of 70 dB at 31.5 Hz decreasing by 3 dB per octave. The correct overall SPL is determined to be 73.0 dB, while the A-weighted sound pressure level is 60.8 dBA. The error in the original calculation stemmed from averaging the dB values instead of summing the intensities, which are proportional to the square of the pressure values. The final calculations utilized the formula SPL=20log(P_rms/P_ref) and confirmed the correct values through proper intensity summation.

PREREQUISITES
  • Understanding of sound pressure level (SPL) calculations
  • Familiarity with decibel (dB) and A-weighting concepts
  • Knowledge of logarithmic functions and their application in acoustics
  • Proficiency in using Excel for data calculations
NEXT STEPS
  • Study the principles of sound intensity and its relationship to pressure levels
  • Learn about A-weighting and its significance in sound measurement
  • Explore advanced SPL calculation techniques using logarithmic properties
  • Investigate tools for acoustic analysis, such as MATLAB or specialized sound measurement software
USEFUL FOR

Acoustic engineers, sound technicians, and students in audio engineering or environmental science who are involved in sound level measurements and analysis.

roldy
Messages
206
Reaction score
2

Homework Statement


The band levels for a noise are 70 dB reference 20\muPa at 31.5 Hz and decrease 3 dB for each octave. Find the overall sound pressure level between 31.5 and 8000 Hz and the A-weighted sound pressure level for this frequency range.


Homework Equations


SPL=20\frac{P_{rms}}{P_{ref}}



The Attempt at a Solution


I think my misunderstanding of the problem is prohibiting me from calculating the correct answer.

I used excel to do the calculations

The band level noise for each octave are as follows

31.5 Hz: 70 dB
63 Hz: 67 dB
125 Hz: 64 dB
250 Hz: 61 dB
500 Hz: 58 dB
1000 Hz: 55 dB
2000 Hz: 52 dB
4000 Hz: 49 dB
8000 Hz: 46 dB

The reason I did this was because the problem stated that the band level noise decreased 3 dB for each octave.

Then I added in the corrections to the dB values above. The correction values for each octave band level are as follows.

31.5 Hz: -39.4 dB
63 Hz: -26.2 dB
125 Hz: -16.1 dB
250 Hz: -8.6 dB
500 Hz: -3.2 dB
1000 Hz: 0 dB
2000 Hz: 1.2 dB
4000 Hz: 1.0 dB
8000 Hz: -1.1 dB

The final dB values:
31.5 Hz: 30.6 dB
63 Hz: 40.8 dB
125 Hz: 47.9 dB
250 Hz: 52.4 dB
500 Hz: 54.8 dB
1000 Hz: 55 dB
2000 Hz: 53.2 dB
4000 Hz: 50 dB
8000 Hz: 44.9 dB

I then square them
31.5Hz: 936.36
63 Hz: 1664.64
125 Hz: 2294.41
250 Hz: 2745.76
500 Hz: 3003.04
1000 Hz: 3025
2000 Hz: 2830.24
4000 Hz:2500
8000 Hz: 2016.01

Then the average
2335.051111
and then square root
48.32236657

This gives me the rms pressure value for 31.5 to 8000 Hz. I then use the equation
SPL=20log\frac{P_{rms}}{P_{ref}} to find the overall sound pressure level. The answer I get is 127.66 dB. The correct answer is 73.0 dB and the A-weighted value is 60.8 dBA. What did I do wrong?
 
Physics news on Phys.org
This is an old (12+ years at time of answering) question. Since communications with the OP is unlikely I’m posting a fairly detailed answer.

The OP appears to have ssumed that the overall dB value depends on the average of the bands’ intensities. This is wrong. The overall dB values depends on the sum of the bands’ intensities.

It’s worth noting that intensity is proportional to pressure squared. And squaring (in a later step) can be avoided by remembering $$20log\frac{P_{rms}}{P_{ref}} = 10log\frac{P_{rms}^2}{P_{ref}^2}$$We can first find the unweighted (no A-weighting) dB values as follows:

As noted by the OP, the band levels are
31.5 Hz: 70 dB
63 Hz: 67 dB
125 Hz: 64 dB
250 Hz: 61 dB
500 Hz: 58 dB
1000 Hz: 55 dB
2000 Hz: 52 dB
4000 Hz: 49 dB
8000 Hz: 46 dB

We can calculate the value of ##\frac{P_{rms}^2}{P_{ref}^2}## for each band using ##10^{\frac {dB}{10}}##. This gives the following values:
31.5Hz: 10000000
63 Hz: 5011872.33627272
125Hz: 2511886.43150958
250Hz: 1258925.41179417
500Hz: 630957.344480193
1kHz: 316227.766016838
2kHz: 158489.319246111
4kHz:79432.8234724282
8kHz: 39810.7170553497

The total of these is 20007602.1498474. Converting back to dB gives
10log(20007602.1498474) = 73.0dB matching the official answer.

To find the dB(A) values, the procedure is the same, except that we start from the A weighted values correctly given by the OP in Post #1:
31.5 Hz: 30.6 dB
63 Hz: 40.8 dB
125 Hz: 47.9 dB
250 Hz: 52.4 dB
500 Hz: 54.8 dB
1000 Hz: 55 dB
2000 Hz: 53.2 dB
4000 Hz: 50 dB
8000 Hz: 44.9 dB
 

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 9 ·
Replies
9
Views
7K
Replies
16
Views
3K
Replies
3
Views
1K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 11 ·
Replies
11
Views
6K
  • · Replies 1 ·
Replies
1
Views
945
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K