CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Sound pressure level (SPL) spectrum (

cfd novice October 12, 2008 11:15

Sound pressure level (SPL) spectrum
Hi guys

I want to compute SPL spectrum of my fluctuating pressure that I have monitored at a point in my domain. Does anyone know how to do that. Thanks is advance for your help.


Harish October 12, 2008 19:53

Re: Sound pressure level (SPL) spectrum
You need to calculate the RMS of the fluctuating pressure and compute log(p_rms/p_ref). You might want to monitor the fluctuating pressure over the entire domain( say at finite intervals) as at the end you might be interested in computing the directivity.

franckb October 13, 2008 04:54

Re: Sound pressure level (SPL) spectrum
SPL[dB]=20*log_10(p_rms/pref) with pref=2x10^5 [Pa]

You can have a look for example at: for more details.

Hope this helps...

cfd novice October 13, 2008 05:33

Re: Sound pressure level (SPL) spectrum
Hi Guys

Thank you for answers. Do I have to do fft? If yes, do I have to do fft on pressure first and then use SPL formula. Or do I have to find SPL pressure first and then do fft on it.


Louis October 13, 2008 08:20

Re: Sound pressure level (SPL) spectrum
I think what you want to do is a power spectral analysis, which involves performing an FFT. I think programs like Matlab or Mathmatica do these for you. It is not as simple as doing an FFT because you need to be wary of the errors introduced by random signals. If you need to write your own program, I suggest you look at Bendat and Piersol.

cfd novice October 13, 2008 09:59

Re: Sound pressure level (SPL) spectrum
Hi Louis

I want to do SPL spectrum, not the power spectrum. It is quite easy to do FFT in matlab, they have loads of tutorials which explain how to perform FFT. So that is not a problem.

I need to know the steps for generating SPL spectrum? Thanks for your answer.


Harish October 13, 2008 12:58

Re: Sound pressure level (SPL) spectrum
I would suggest you to take a look at the Aeroacoustic benchmarks. Its available on the NASA technical servers. It would give you an idea of how people do it.

Steve October 14, 2008 05:41

Re: Sound pressure level (SPL) spectrum
You will almost certainly get the amplitude wrong unless you really know what you are doing.

Some advice. Calculate an RMS in the time domain. Do not trust your frequency domain data until you can get the same RMS value using it. Or understand & justify the difference.

cfd novice October 14, 2008 06:02

Re: Sound pressure level (SPL) spectrum
Thank you for your suggestion guys. You have been very helpful.


All times are GMT -4. The time now is 22:26.