MESH-PARTICLE MULTIPHYSICS​ ​

meeting highest demands for complexity, accuracy and validity in the simplest possible way.​
Talk To An Expert
Sign up for Knowledge Base

Combining the Best of Both Worlds​

To Create Something Even More Powerful​
A Unified Interface Capable of Handling
Mass Flux​
​e.g. to let water seep into a sponge
Momentum Flux​
e.g. to accurately depict forces on an elastic body​
Energy Flux​
e.g. to heat up a solid body​
Consistent Boundary Handling​
e.g. for open and wall boundary conditions​

Explore a Large Variety of Materials​

To Tailor Your Model
Oil
Water
Grease
Air
Steel
Chocolate
Oil

Oil is a viscous liquid used in lubrication, cooling and hydraulic systems. It's modelled in CFD studies involving fluid dynamics in engines, pumps, gearboxes and bearings.

Key Properties
  • Density: ~850-900 kg/m3
  • Kinematic Viscosity: ~10-1000 cSt
  • Thermal Conductivity: ~0.15 W/mK
  • Specific Heat Capacity: ~2000 J/kgK
Water

Water is a low-viscosity, incompressible fluid with high thermal capacity. It's used in heat exchange, hydraulics and washing systems applications. It's frequently modeled in CFD free surface and multiphase simulations in cleaning nozzle applications, drainage and cooling systems.

Key Properties
  • Density: ~997 kg/m3
  • Kinematic Viscosity: ~0.9 cSt
  • Thermal Conductivity: ~0.6 W/mK
  • Specific Heat Capacity: ~4184 J/kgK

‍

Grease

Grease is a semi-solid lubricant with complex rheological properties. It exhibits non-Newtonian flow characteristics, meaning its viscosity changes with shear rate. CFD simulations often analyze its behavior in lubrication systems, bearings, and mechanical contacts.

Key Properties
  • Density: ~900 kg/m3
  • Shear-thinning viscosity

‍

Air

Air is a compressible gas with varying density under different pressure and temperature conditions. CFD simulations of airflows are crucial to assess aerodynamics, windage effects and air entrapments.

Key Properties
  • Density: 1.29 kg/m3
  • Thermal Conductivity: 0.025 W/mK
  • Specific Heat Capacity: 1004 J/kgK
  • Kinematic Viscosity: ~15 cSt
Steel

Steel is a solid material with high density and thermal conductivity. It's widely used in mechanical components such as engine blocks, gearbox housings and axles. In CHT simulations, steel is coupled with fluid domains to analyze temperature gradients and heat dissipation.

Key Properties
  • Density: ~8200 kg/m3
  • Thermal Conductivity ~15 W/mK
  • Specific Heat Capacity: ~500 J/kgK
Chocolate

Chocolate is a viscous, non-Newtonian fluid widespread in the food industry. CFD models its shear-thinning properties to assess filling and coating applications.

Key Properties
  • Density: 870 kg/m3
  • Shear-thinning viscosity

Explore A Vast Amount of Combinations​

To Combine More Physics in One Single Software
Conjugate Heat Transfer
Our solver captures heat transport in both fluids and solids within the same simulation. By coupling SPH for fluids with FEM for solids, we simulate thermal interactions across material boundaries with high physical accuracy. A consistent surface treatment ensures realistic heat transfer between domains – whether it’s jet cooling on a metal surface or evolving temperature distributions in complex machinery.​
Porous Flow
Our method bridges detailed free-surface flow with the complex behavior of flow through porous materials within a single, unified framework. While flow in open regions is fully resolved, porous regions are modeled efficiently at a coarser level, capturing key transport effects without the need to represent every tiny pore. This allows accurate and scalable analysis of washing processes, filtration and membrane systems, wastewater treatment, and many other applications where flow spans a wide range of spatial scales.​
Multiphase
When fluids of different types interact, our method preserves sharp and well-defined interfaces, even under high density ratios of more than 1000:1. Up to five phases can be modeled simultaneously, with consistent treatment of boundaries across all materials. This enables the simulation of true multiphase phenomena, including entrapped air pockets or fast-moving gas flows that affect fluid behavior, which is difficult to capture with single-phase models alone.​
Phase Transition
The field of phase transitions is vast. Understanding industrial applications such as adhesive dispensing and refrigeration necessitate robust solvers that provide accurate approximations to the flows encountered in the field. We offer essential modelling approaches for the solidification and melting of phases, be it due to changing thermal conditions or complex non-Newtonian material behavior. Both our conjugate heat transfer solver and our multiphase model thus enable the analysis of sophisticated multiphysics problems such as bubble entrapment in molding flows.
Higher-Order Operators
Traditional SPH methods can suffer from excessive numerical dissipation, where kinetic energy is artificially lost and important flow patterns and dynamics are suppressed. In some cases, this can lead to unrealistic flow stagnation or pressure increases, especially in narrow gaps between components or across sharp pressure jumps at phase interfaces. By introducing consistent higher-order operators, we were able to significantly reduce these artifacts, preserving the flow patterns and improving physical realism. This enables our solver to capture dynamic fluid behavior far more clearly, especially in complex scenarios like pump flows or rotating machinery involving multiple phases.​

Run On World‘s Fastest Chips​

Solvers Architectured For Today‘s Cutting Edge​
8 CPU
32 CPU
120 CPU
Today, engineers work with a wide spread of low and mid quality hardware.
GPU A100
All Dive Users
GPU H100
Beta Phase
GPU B200
Test Phase
All Dive users access only the latest and fastest HPC resources.
... and they update right after the next generation is available.
Simulation Speedup
Always Faster Than Your Competition
Legacy software needs to run on multiple platforms, old and new. Thus, never really optimized for any of those.​
In the cloud-native CAE, we choose the best suitable hardware for you.
32
18
11
1
Discover how we, together with NVIDIA are using cloud and cutting-edge computing to transform mechanical engineering.
Read the paper

Development Principles​

To Model Reality
Precision
Speed​
Physicality
Adjustability​
Conservation
Consistency​
Unification
Fragmentation
Speed is essential, but enterprise-grade simulation demands accuracy above all, at scale and in the cloud
We prioritize physical accuracy, backed by a continuously growing validation suite of industrial and academic benchmarks. Complex designs impose high demands on simulation fidelity, which we meet by explicitly including detailed geometries through semi-analytical boundary conditions into our models. This level of detail comes at a computational cost, which we handle efficiently in parallel with modern cloud compute hardware accelerators.
Precision
Speed​
Fitting past results is easy. Predicting the future requires physically grounded simulations.
We don’t rely on data calibration – we model reality. While solver parameter tuning may match individual measurements, it rarely generalizes across varying operating conditions. Our solver operates solely with physical input parameters, ensuring consistency with the fundamental laws of physics. This foundation allows us to simulate complex systems reliably for all design spaces.
Physicality
Adjustability​
We build methods that preserve what nature preserves – mass, momentum and energy.
Many commercial solvers aim for higher-order schemes that neglect basic conservation principles . These may perform well in low-complexity academic benchmarks but often break down in industrial-grade applications. We develop numerical methods that strongly conserve mass, momentum and energy to ensure physically meaningful results, even in the most complex simulations.
Conservation
Consistency​
Our one-solver architecture combines mesh and particle methods into a unified, end-to-end simulation environment.
We customize every part of the workflow to perfectly fit our solver. Tailored preprocessing handles complex geometries with high accuracy, and postprocessing reflects the solver’s internal logic to ensure consistent and reliable results. This close integration avoids fragile combinations of separate tools and methods, making the entire simulation setup easier, more automated, and less error-prone.
Unification
Fragmentation

Built-In Features

To Focus On Just Engineering
Preprocessing
Complex
Movements
Local
Refinement
Automation via
Python Scripts
Restart From
Previous Results
Non-Constant
Initial Conditions
Non-Constant
Open Boundaries
.STEP files Importer
Mesh Check Mode
Time-Dependent
External Force
Moving
Reference Frame
Solver
Surface
Tension
Multiphase
Model
Heat
Transfer
Air
Drag
Non-Newtonian
Materials
Temperature-Dependent
Materials
Free
Flow
Pause & Resume 
Run
Postprocessing
Point
Sensor
Surface
Sensor
Surface
Reconstruction
Colormap
Visualization
Download
Data
Plotting
Over Time
Statistical
Calculations