FlowSight is a new advanced visualization tool for FLOW-3D based on the world-leading EnSight® post-processor. FlowSight offers users new and powerful ways to analyze, visualize, and communicate their simulation data. FlowSight’s capabilities include:
- Analyze and compare multiple simulation results simultaneously
- Volume rendering – allows users to see inside their results to reveal details not available in iso-surface displays
- CFD calculator – define new variables such as time averages, integrated values, dimensionless parameters, multiply/divide/scale variables
- Show fluid-structure interaction/thermal stress evolution results with fluid solution results simultaneously
- Flipbooks – flipbooks are essentially interactive animations that allow large results to be examined without the delays associated with stepping through results
frame by frame
- Animated time-dependent plots – multiple, time-varying plots to help convey complex analyses to customers of varying technical backgrounds
- Save and restore settings to make post-processing standard images and animations faster and easier
- Interactive 2D slices through flow and structural domains
- Animated streamlines – streamlines can be animated with moving arrows or cones
To see more examples of the power of FlowSight for data visualization and advanced post-processing capabilities, watch our YouTube playlist.
The FLOW-3D Solver Version 11.0
Advances in Meshing
The Unstructured Memory Allocation approach to addressing solution arrays on structured grids has been extended to have separate memory space (sub-domains) for the following types of numerical solutions: fluid flow, heat transfer in solid and core-gas flow. A solution variable is now stored only in cells where it is needed. The result is a much more compact and efficient representation of the solution in memory (e.g., no fluid variables are stored in fully solid cells and vice versa), shorter simulation times (due to addressing a smaller memory footprint) and smaller flsgrf files (because of writing out smaller datasets). There is no input required from the user to take advantage of this development; the creation of the sub-domains happens automatically based on the geometry and model selection. There is also no loss of solution accuracy.
Conforming mesh blocks
Instead of being rectangular, as all standard mesh blocks are, conforming mesh blocks are shaped according to the geometry. There may be two types of conforming mesh blocks – cavity-conforming and solid-component-conforming. A cavity-conforming mesh block is commonly used in casting modeling, where the cavity typically requires a finer mesh than the mold. The simplest setup is to have one rectangular mesh block with a relatively coarse mesh covering the whole domain and a nested cavity-conforming mesh block around the cavity.
A component-conforming mesh block is useful when resolving thin structures or thin boundary layers around solid structures. The user can select which component(s) a mesh block should conform to. By default, it conforms to all solid components within its rectangular volume.
The extent of a conforming mesh block beyond the boundaries of the shape it is conforming to is defined by the input parameter overlap. By default, it is equal to five times the average cell size in the conforming mesh block. Conforming mesh blocks can be nested, linked, partially-overlapping or simply standalone mesh blocks.
Partially overlapping mesh blocks
The limitation of mesh blocks to be purely linked or nested has been lifted. Mesh blocks can overlap each other in an arbitrary fashion. In areas common to several blocks, the flow equations are solved in the one with the finest mesh (based on the average cell size in the block); the solution is interpolated in all the other blocks. The user can override this order by setting the mesh block ranking – for now by editing the prepin file and setting the variable MESH_RANK in each MESH namelist (with 1 being the highest rank). Mesh block ranking falls back to the defaults behavior based on the average cell size if any of the MESH namelists misses the setting of MESH_RANK. Partially overlapping mesh blocks make mesh generation less cumbersome and should generally result in fewer mesh blocks in a simulation.
External boundary conditions in nested blocks
When a nested block’s boundary coincides with an external boundary of the containing block, the user can define the standard boundary conditions at that nested block’s boundary, independently of those of the containing block. Previously, a nested block always assumed the boundary conditions of the containing block.
Sediment transport and erosion model
The sediment transport (scour) model has been largely rewritten to use FAVOR™ functions to describe packed bed, improving the accuracy of the turbulent shear stress at packed sediment surface and providing better visualization of the results. The accuracy of the approximations for sediment erosion and deposition has also been enhanced with the inclusion of surface roughness based on d50 diameter. The sediment transport model has also been extended to shallow water flows, allowing users to model large scale sedimentary problems within the framework of the hybrid 3D/shallow water approach.
Surface tension model
A new surface tension model replaces the old one. The new model has demonstrated better accuracy and stability of the solution. In particular, the accuracy of the calculation of wall adhesion forces is greatly improved. The old surface tension model is still available in an unlikely case of a failure of the new model. The old model is activated by adding IFSFT_OLD=1 to the namelist XPUT.
k-omega turbulence model
The k-omega turbulence model compliments the existing two-equation k-omega and RNG k-epsilon turbulence models. The new model has superior properties near walls and will benefit the accuracy of the sediment erosion and transport model.
The granular flow model has been extended from describing granular/gas mixtures to granular/liquid, or slurry, flows. This may be useful when modeling, for example, the flow of mud. The model could be viewed as a simple version of the sediment transport model without the complications of erosion and multiple sediment species. It uses the average grain properties to describe the granular material in the mixture.
The forces acting on the granular media in both granular flow models, with gas and with liquid, now include dispersive pressure due to grain collisions in shearing flow. The additional force is a function of the so-called friction angle which is an input parameter to the models.
Shallow water model
The reservoir surface roughness is now included in the calculation of turbulent shear stress in shallow water flows. For standard solid components the roughness is defined by the user in the same way in the 3D flow model, and for a packed-sediment type component the roughness is proportional to the value of d50 diameter calculated from the sediment content in the packed bed.
Linear damping has been added to the springs & ropes model, including the torsion springs, significantly extending its area of applicability. In particular, springs can now be used to model shock absorbers in cars and wave energy conversion devices.
The choice of constitutive stress-strain relationships for visco-elastic fluids has been extended to include Oldroyd-B and Giesekus models. The original model did not accurately capture the dynamics of true viscoelastic materials since it treats materials as being either purely viscous or purely elastic depending on the local state of stress. Conversely, the new viscoelastic models allow for simulation of simultaneous viscosity and elasticity as a continuous function of the local state of stress and time. As such, it more accurately predicts the instabilities that occur during coating processing, application of adhesives, bottle filling of viscoelastic materials, and injection molding of polymers.
Constant-pressure bubbles with vaporization
The vaporization model to constant-pressure bubbles has been brought in line with the other liquid/vapor phase change options. The required vapor physical properties are now consistent across all the phase change models. In addition, the vapor partial pressure in the voids can be defined allowing users to describe bubbles that contain non-condensable gases, such as air.
FSI components can now dynamically interact with each other as well as with solidifying fluid regions. There are two options for the coupling between any two deforming objects – complete coupling, as when the objects are glued to each other, and partial coupling, when the two objects can slide and separate at the contact surface. Among other applications, this is especially useful in modeling stresses developing in metal and mold during solidification in die casting. Stresses and deformations can be calculated for different parts of the die and slides as they press together to form the cavity.
Tetrahedral FE mesh
Both the FSI and TSE solvers can now use tetrahedral mesh, instead of the hexahedral ones. A tet mesh can be generated either internally in the preprocessor or imported from a file generated by a third-part mesher. The choice of the mesh type – tetrahedral or hexahedral – can be made by the user for each FSI component and the TSE region individually.
Tetrahedral and hexahedral finite element meshes can be imported from third-party mesh generation tools via the Exodus-II file format. This provides the means to use high-quality meshes in FLOW-3D’s FEA simulations. The user has to be careful that the domains covered by the imported meshes coincide with the definition of the geometry in FLOW-3D. The FE mesh can also be exported, along with the results at restart data edit intervals, if so desired, using the same Exodus-II format. These files can then be used for post-processing or importing of the results into other FEA simulation and analysis tools.
Advection of FEA solution
Advection terms have been added to the TSE equations, allowing users to model stresses in a moving solidified phase as in the continuous casting process, where minimization of residual stresses is a very important part of the design process.
Multiple initial particle blocks
Users can now define multiple initial blocks of particles, so that particles can be introduced at the beginning of a simulation at different locations and with different properties.
Particle origin tags
Particles are now tagged according to their origin indicating which initial block or particle source they came from. A tag is an integer corresponding to the initial block or source number. Two independent sets of tags are used for the initial particle blocks and the particle sources.
Metal Casting models
Core gas model
The accuracy of the core gas model has been enhanced through the introduction of the core gas solution sub-domain, which includes not only the fully blocked cells of a core gas component as in previous versions, but also the partially blocked ones. Also, the core gas model has been SMP parallelized.
The versatility of the cooling channel model has been enhanced by making it completely separate of the void model which was previously used to define cooling channels. Cells occupied by a cooling channel are still empty of solid and fluid material, but they are not included in the calculation of void regions. As a result, the reported void regions represent the actual physical voids in the problem. Also, a cooling channel can be connected to a void region and still retain its properties and functionality as a cooling channel. Finally, when cooling channels are defined with STL files, those STL objects can be loaded in the 3D viewer and manipulated (e.g., hidden or made transparent) independently of the rest of the geometry.
Cores and liners in die casting
Core and liners are inserted into a die assembly and are removed after each cycle. This can now be modeled by defining special types of components. Unlike other parts of the die that are reused after each cycle and thus retain their temperature history, the temperature of cores and liners is reset at the beginning of each cycle.
Tabular time-dependent input
Input tables for any time-dependent property such as mesh boundary conditions, GMO motion or mass source rate, can now be defined using external files which are then read during the simulation without copying the data into the prepin file. Unlike the tables defined directly in the prepin file where the number of points is currently limited to 500, there is no limit on the number of time points in the external tables. This allows users to define very detailed behavior of transient objects in a simulation. The impact of using the external tables on the memory and CPU usage is very small.
Solution output at a fluid or FSI probe can be used to determine simulation termination conditions, in addition to the existing conditions based on time, fill fraction or steady-state conditions. For example, a simulation can be automatically stopped when fluid pressure at a probe location reaches a predefined value.
When using a reduced heat transfer solver inside solid components, i.e., the thermal penetration depth model or the constant non-uniform temperature option, the solid temperature in solid region outside the active domain are now retained and can be used for post-processing and restarts. This is especially useful when running, for example, a thermal die cycling simulation with the full heat transfer solver, then a filling restart with a reduced solver for speed and then a restart simulation for solidification, again with the full heat transfer solver. In this case, the die temperatures obtained during the first simulation are available in the last one as the initial conditions in the newly activated die regions.
Component permeability input
Permeability can now be directly input to define porous properties of the core gas, permeable mold and porous components. Non-Darcian permeability can also be defined for core gas and porous components (not for permeable mold components). The old options to use sand grain size for the core gas and permeable mold, or drag coefficients for porous components are also retained.
Hydrostatic mesh boundary conditions
An integer flag has been added to the definition of the existing hydrostatic pressure mesh boundary conditions to distinguish it from the uniform pressure BC, IHPBCT(n), where n=1÷6. A hydrostatic pressure BC is primarily characterized by the fluid elevation. The hydrostatic pressure boundary condition has also been extended from the x- and y-direction mesh boundaries to the z-direction boundaries. When fluid elevation is defined at a Zmin or Zmax boundary, the uniform boundary values of fluid fraction and pressure are computed there accordingly.
Time-dependent void pointers
Void pointers can now be used to define not only an initial state of a void region, but also its pressure and temperature during the simulation by using a tabular definition of the pressure (or temperature) vs. time. This feature effectively acts as a pressure and/or temperature time-dependent boundary condition for the fluid surrounding the void containing the pointer. A time-dependent pointer becomes inactive when covered by fluid or a GMO component, and is reactivated as soon as it becomes surrounded by void again. Time-dependent void pointers could be used to easier model heat treatment of a casting by defining the time-dependent temperature in the void surrounding the part. A prescribed pressure vs. time at a void pointer could also drive the growth of a vapor bubble in a printer nozzle simulation.
Mass source and steady-state
Mass and mass/momentum sources are now compatible with the constant-velocity flow solver, IFVELP=1. This is useful when a fully transient simulation with mass sources/sinks is run first to achieve a steady-state at which point a restart simulation is carried out with the constant velocity field.
New input flags
Integer flags have been added to some existing modeling options to reduce user confusion and to simplify the GUI.
- A new flag IOEPOTM(N) determines if the component #N has a prescribed or dynamically calculated electric potential.
- The same was done for the thermal properties of component to distinguish between adiabatic component, components with prescribed lumped temperature, dynamically varying lumped temperature and conducting ones. The input variable is IHTOBS(N).
- An input flag IFSLD_DRG has been added to namelist XPUT to distinguish between the drag-based and viscosity-based solidification model.
Transformation center for fluid and baffle regions
The transformation center – a point around which scaling and rotation of geometric regions is done, – has been extended from geometry subcomponents to fluid and baffle regions.
Flow rate at valves/vents
Gas flow rate at valves is output to the General history data catalog. This is useful for the evaluation of the effectiveness of the gas evacuation at valves.
Users can now define the initial uniform scalar concentration at droplet sources of fluid #1 or #2.
The V11.0 Graphical User Interface (GUI)
A new mesh quality checking feature has been added that identifies and display FAVOR™-related mesh resolution issues for STL-defined geometries. The FAVORizer is now equipped with a diagnostic tool for detecting poor resolution of geometric objects defined using STL files. After running the FAVORizer to preview geometry, the results can be loaded into the graphical display window in Meshing & Geometry for interactive analysis. This feature does not work for geometry defined with primitives.
The Materials Databases have been upgraded to include more materials, more temperature-dependent properties, as well as better navigation and organizational abilities, including:
- New example simulations
- External CSV file support – external files can be easily used to specify time and temperature tabular data via CSV files.
- Runtime changes to maximum time step – the maximum allowable time step can now be modified throughout the simulation.
- Multiple baffle regions – baffles containing multiple regions can now be created and edited in the GUI
- Gravity vector display – inertial gravity direction is now displayed in the Meshing and Geometry window.
- Simulation status in Portfolio – the status of all simulations in the Portfolio are immediately indicated when the GUI is opened. Previously a simulation would have to be loaded to determine to determine a simulation’s status.
- Delayed simulation loading – single-clicking on simulations in the Portfolio allows simulation diagnostics and runtime plots to be viewed without loading the geometry.
- Simulation unloading – simulations can be unloaded from the File menu. Memory will be uncommitted from the GUI. Note that the size of the executable will appear to be the same after a simulation is unloaded due to the way modern operating systems report memory usage.
- New hydraulic data output – maximum flow depth (flow inundation), specific and total hydraulic head are now available.
- Simulation project layout – the layout of listboxes (e.g., Meshing) on the Meshing and Geometry tab is stored in the .Flow3dProj project files for each simulation. The .Flow3dProj can be copied with simulation copies and loaded so that the listbox layout of the simulation being copied is retained.
- Simulation conversion logging between versions – when obsolete input variables are converted automatically by the GUI, the changes are now displayed in a conversion log when the simulation is opened. The obsolete variables are listed as well as what they have been converted to. The converted simulation file is appended with the version number to which they were converted.