Learning Forum

[Gorilla Koala]

Hello,

I'm calculating acoustic pressure inside pipe using LES+Ffowcs Willams and Hawkings (FW-H) model. 

During the calcualtion, I noticed that the results extremly depend on the time stpe size. I tried that time step size 5E-5, 1.1E-6 and 5E-7 sec. And the result is the figure below. The time step size is smaller, the peaks shift to right. I calculated this result by the steps;

1) ANSYS fluent calcuates acoustics pressure with LES+FW-H model

2)Power spectral density (psd) is calculated from the acoustic pressure

3)spl = dB(sqrt(psd*df))

Do you have any idea why the peak of the sound pressure level shift to large when the time step size becomes small?

And how much should I decrease the time step size?

Thanks!

 

[Federico Alzamora Previtali]

Hello,

I suggest you look into aliasing errors, which occurs in a FFT due to insufficient sampling of a high frequency signal.