CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Near field noise prediction (https://www.cfd-online.com/Forums/cfx/25085-near-field-noise-prediction.html)

 Usman January 22, 2008 19:50

Near field noise prediction

I want to calculate near field noise source (i.e. pressure fluctuations). I have following questions with regards to this.

1. I understand that compressibility of fluid can cause acoustic waves in the flow. However, what is the cause of these acoustic waves? Is it fluctuations in velocity field or density field. Wikipedia states that it is velocity fluctuations, if so, then why we need to set up compressible simulation to calculate pressure fluctuations.

2. How do we set up near-field noise prediction problem in CFX-Pre. Do we make the denisty function of pressure or we set up user points to take pressure readings at each time step. In either case how do we do it in CFX-Pre.

3. How small my time step should be or how would you calculate it.

Thank you very much. Usman

 Andreas January 23, 2008 12:15

Re: Near field noise prediction

To 1: Since velocity and pressure are coupled, velocity fluctuations lead to pressure fluctuations, i.e. acoustic waves.

To 2: You should think whether compressibilty is important in your case. It is e.g. for Helmholtz resonator, Rijke tube and a recorder, but not for turbulent flow around an obstacle (cylinder, airfoil) or oscillations due to instabilities (e.g. edge tone). In some cases it may be important (flow over cavity). If it is, you can specify a fluid as "ideal gas" or, if you change material group e.g. to user, set any other pressure dependent density for your material.

To 3: Time step shall be so small that (i) CFD simulation converges and (ii) sampling frequency (i.e. 1 over time step) is at least twice of interesting acoustic frequencies.

 All times are GMT -4. The time now is 09:10.