Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations TugboatEng on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Combining noises
- Thread starter EcoMan
- Start date
Jeez people.
This has nothing to do with pure tones. It is to do with whether the sources are correlated or not.
If the sources are uncorrelated then sum the rms sound pressures. If they are correlated then you need to sum the instantaneous sound pressures, then take the rms.
Take 10 random noise generators and connect each one to its own amp and speaker.
Now take 1 random noise generator and feed the signal to 10 amps and speakers.
Assume everything is in a free field, the speakers are small, omnidirectional and far enough apart that no speaker is in any other's nearfield. i.e. each speaker produces an independent pressure field.
For the 10 noise generators the time averaged sound pressure level will be 10 dB more than for just one noise generator.
For the single noise generator connected to 10 speakers this is not true and there will be anything between a 20 dB increase and nothing at all depending on the degree of correlation of the sources (in this case, 100% correlated) and the location of the measurement point with respect to the sources.
The confusion has arisen in this thread because of the original post which says "identical sources". As the above example shows, it very much depends on what you mean by "identical sources"
M
--
Dr Michael F Platten
This has nothing to do with pure tones. It is to do with whether the sources are correlated or not.
If the sources are uncorrelated then sum the rms sound pressures. If they are correlated then you need to sum the instantaneous sound pressures, then take the rms.
Take 10 random noise generators and connect each one to its own amp and speaker.
Now take 1 random noise generator and feed the signal to 10 amps and speakers.
Assume everything is in a free field, the speakers are small, omnidirectional and far enough apart that no speaker is in any other's nearfield. i.e. each speaker produces an independent pressure field.
For the 10 noise generators the time averaged sound pressure level will be 10 dB more than for just one noise generator.
For the single noise generator connected to 10 speakers this is not true and there will be anything between a 20 dB increase and nothing at all depending on the degree of correlation of the sources (in this case, 100% correlated) and the location of the measurement point with respect to the sources.
The confusion has arisen in this thread because of the original post which says "identical sources". As the above example shows, it very much depends on what you mean by "identical sources"
M
--
Dr Michael F Platten
SomptingGuy
Automotive
I have inadvertently done this experiment in a simulation before. Adding flow noise to an exhaust pulsation noise, I re-seeded the random number generator for each tailpipe and got (of course) identical random noise from each (two identical sources gave +6dB). I soon realised and rearranged the re-seeding so that there was no correlation between the sources.
- Steve
- Steve
- Status