Scale simulation to even the most complex geometries With Multi-Grid Modeling

Tradition FEA simulation approaches result in exploding runtimes and memory consumption as models increase in size

Multi-Grid Modeling solves this problem by allowing for nearly linearly scaling of performance as model size increases

All while maintaining the accuracy of the finest mesh!

Efficiently simulate large DED builds
Distortion simulation of a 1.25m tall DED build. Runtime = 75sec on an 8 core laptop. Peak memory consumption = 1.5 GB.

PanX is not limited to LPBF simulations, but is also applicable to DED, welding, and cladding processes

Our novel part-scale simulation approach can efficiently simulate even the largest DED parts

Dramatically reduce simulation hardware requirements

Fast runtimes, low memory consumption, and small output data size makes it possible to run relatively complex parts even on a laptop

Geometries, such as the one shown, are notoriously computationally expensive due to the combination of thin (down to the resolution of the machine) and thick sections

Simulation of a LPBF rocket component on an 8-core laptop. Runtime = 36 mins. Peak memory use = 56 GB. Number of nodes = 1.00M.
Optimize Support design
Optimized support volume converges after 8 iterations.
Total runtime: 31 seconds on an 8 core laptop.

Minimize support volume while limiting recoater jams or part temperature for Laser Powder Bed Fusion builds

Efficient gradient-based MMA optimizer allows results to converge in just a few iterations

Solver computes sensitivities (adjoint problem for manufacturing)

Optimize laser scanning strategy

Optimize laser scan power point-by-point for the full scan path of the build to maintain a constant melt pool volume

Results into uniform melting, eliminates key-holing, and improves overall part quality

Calculated optimal power distribution to maintain a constant melt-pool size.