Passive Scalar Sources¶
Passive scalar sources can be used to simulate the propagation of spiecies like smoke from a burning car in a garage or the diffusion of dust or pollutants in a tunnel.
Passive scalar sources are implemented in the Passive Scalar Transport analysis. Please, see the documentation page for this type of analysis Passive Scalar Transport. The passive scalar source has to be assigned to a volume, defined through a geometry primitive (cartesian box, sphere and cylinder) or a cell zone (see the Surface Refinement section of Mesh settings for further details on how to create a cell zone). It is generally good practice to refine the mesh in the vicinity of the source, in order to better capture the dynamics of the flow and avoid convergence problems.
Creation of a passive scalar source¶
In the simulation tree navigate to “Advanced Concepts” and add a “Passive Scalar Source”. Two types of sources are supported:
Passive scalar source¶
This type of source is defined by a flux, which may be interpreted as a concentration per unit time. The user needs to specify the flux value and select the region previously defined.
Volumetric passive scalar source¶
This type of source is defined by a flux per unit volume. Consequently, the actual flux is implicitly computed using the volume of the source region. As in the first case, the user needs to specify the specific flux value and select the region previously defined.
In the Figure above, a passive scalar source is defined using a sphere as geometry primitive (highlighted in blue).
It is important to understand the correct interpretation of units used for passive scalar sources. Since passive scalars do not affect the dynamics of fluid flow, units for passive scalars are independent to the unit system of simulation. Thus one can define a passive scalar flux as 100 1/s and interpret it as a flux of 100 g/s or 100 kg/s as per one’s convenience. However, these interpreted units must be kept consistent throughout the simulation setup (for values specified in initial conditions, boundary conditions etc). Scale of the results will directly correspond to absolute value of input variables.
Below you can see comparison between cases for two different passive scalar source values, one defined as 500 1/s (interpreted in grams) while other as 0.5 1/s (interpreted in kilograms). We see that the final results are only scaled by the equivalent factor (1000 here, see legends), whereas the actual dynamics remain unaltered. Thus the relative distribution of passive scalar remains the same. This is evident from following figure
The full setup is contained in this project: