The Acoustics Region group box is used to set up the geometry of the source mask and sponge regions.

There are two methods to define geometry data:

The two methods may be combined together. The source mask and the sponge layer UDFs are created using the following macros:

DEFINE_SOURCE_MASK(src_mask_define, c, t)

DEFINE_SPONGE_LAYER(sponge_define, c, t)

where src_mask_define and sponge_define are example function names, c is a cell number, and t is a pointer to a cell thread. There can only be one UDF of each of these two types per simulation. Each UDF returns a single real value, which fills a corresponding storage variable (SV_SOUND_MODEL_SRC_MASK or SV_SOUND_SPONGE) in the cell c of the thread t.

Simple geometries can be built out of the basic shapes (rectangular hexahedron, cylinder, sphere), which must be preliminarily defined as cell registers of the type Region.

 Solution → Cell Registers  New → Region...

After you create one or more cell registers, you can use them to specify the desired geometry by clicking Basic Shapes... in the Acoustics Model dialog box (Figure 23.16: The Acoustics Model Dialog Box), which opens the Basic Shapes dialog box (Figure 23.17: The Basic Shapes Dialog Box). In this window you can add cell registers to either the source mask or sponge layer geometry definitions, and specify the individual transition thickness for each shape for each purpose.

Figure 23.17: The Basic Shapes Dialog Box

The TUI commands for creating and editing source mask and sponge layer geometry can be found in the menu:

/define/models/acoustics/wave-equation-options/basic-shapes/

To visualize the source mask and the sponge layer geometries, use the field variables Sound WaveEq Model Source Mask and Sound Sponge Layer Marker from the Acoustics… variable group. These output fields directly show the storage variables SV_SOUND_MODEL_SRC_MASK and SV_SOUND_SPONGE without any transformations.

When the wave equation model is active, the Solution Controls task page includes an additional frame titled Acoustics Wave Equation Solver Controls (see Figure 23.18: The Acoustics Wave Equation Solver Controls Task Page).

Figure 23.18: The Acoustics Wave Equation Solver Controls Task Page

The parameters Relative Convergence Criterion and Max Iterations/Time Step control the sub-iterations of the implicit wave equation solver. The need for sub-iterations comes from explicit skewness corrections to the discretization of the basic Laplacian term in Equation 11–11, as well as the artificial viscosity term. At each sub-iteration, the system of linear equations is processed by the AMG linear solver to reduce the residual norm by a factor of 0.1 (this can be controlled by changing the rp-variable acoustics-waveeq/amg-alpha). The optimal values of these three algorithm parameters may depend on the mesh quality. Their default values may change in future releases after accumulating more experience with the wave equation model.

The TUI commands for the acoustics wave equation solution controls are found in the menu:

/solve/set/acoustics-wave-equation-controls/

The sub-menu /expert allows you to specify under-relaxation factors to enforce convergence on low quality meshes. Under normal circumstances, under-relaxation is not recommended.

The wave equation model is to be activated after the development of a time-dependent flow solution. Therefore the acoustics field requires a separate initialization process, that includes ramping in time the sound source term found in Equation 11–11. With the wave equation model enabled, a Sound Potential field and a new button Initialize Acoustics… appear on the Solution Initialization task page. The initial Sound Potential value should always be kept at zero. The Initialize Acoustics… button opens the Acoustics Initialization dialog box (Figure 23.19: The Acoustics Initialization Dialog Box), where you can specify the number of timesteps for the ramping process.

Figure 23.19: The Acoustics Initialization Dialog Box

You can re-initialize the acoustics field separately at any time, since it does not influence the transient flow solution which is simply continued further.

You can specify the number of timesteps for the ramping process using the TUI command:

/solve/initialize/init-acoustics-options

The following field variables are available in the Acoustics… group for post-processing:

You can also postprocess results using the Ansys Sound Analysis dialog box. Refer to Postprocessing the FW-H Acoustics Model Data for additional information.

The Kirchhoff integral model is available for a three-dimensional acoustics simulation with the wave equation model. It is activated by selecting the checkbox, Remote Receivers, in the Acoustics Model dialog box. When this checkbox is selected, a subframe, Kirchhoff Integral Options, with additional buttons appear in the Wave Equation Options frame, as shown in Figure 23.20: Acoustics Model Dialog Box with Kirchhoff Integral Options.

Figure 23.20: Acoustics Model Dialog Box with Kirchhoff Integral Options

Clicking Integration Surface... opens a list of all currently available surfaces, see an example in Figure 23.21: Integration Surface Dialog Box. The two most useful surface types for the Kirchhoff model are quadric surfaces (ribbon menu Domain→Surface→Create→Quadric) and iso-surfaces. With the current implementation only one surface can be selected in the list and used for the integration.

Figure 23.21: Integration Surface Dialog Box

Clicking Define Receivers... opens the same Acoustic Receivers dialog box that is used to set up the Ffowcs Williams and Hawkings (FW-H) model, see Specifying Acoustic Receivers. For convenience, to simplify the simultaneous use of the FW-H model and the wave equation-based Kirchhoff integral model, receivers are shared between the two models and do not need to be set up twice. The signal file name extension for the Kirchhoff model is fixed to .ark.

Clicking Write Signals writes the computed sound signals at the remote receiver locations to ASCII files. The written Kirchhoff signals are always auto-pruned, see Pruning the Signal Data Automatically for the details. If a sound signal is too short and does not yet contain the complete sound pressure data, then it is not written, and a warning message is printed in the Fluent console.

The Kirchhoff model is currently available only in “on-the-fly” mode, which means that the next set of signal samples is computed during a transient simulation after each time step. The “read-and-compute” mode for the FW-H model makes it possible to use the same transient flow simulation for both the FW-H and the wave equation model (with the Kirchhoff integral). To perform such a simulation:

The text commands for the Kirchhoff model, which can be used in the Fluent console and in journal files, are shown here in bold.

/define/models/acoustics/wave-equation-options> 
basic-shapes/                                   sponge-layer-base-level
remote-receivers-options/                       sponge-layer-factor
remote-receivers?                               sponge-layer-udf
source-mask-udf                                 time-filter-source?

/define/models/acoustics/wave-equation-options/remote-receivers-options>
integration-surface         write-signals