Free Courses Fluids Structures Electronics 3D Design Optics Photonics Materials Installation Python Optimization Safety & Cybersecurity Embedded Software MBSE Digital Mission Engineering STEM NVH Student Teams Semiconductors Indo Kursus Partner Courses My Courses

[FAQ]

The evolution of the temperature field in large number of connected CHT bodies, i.e. PCBs, can be slow in some cases. In order to faster achieve the final temperature level the following hints can be useful. 1. Partitioning Method Try an alternative partitioning method like ORCB or user specified direction. Ensure that the coupled partitioning method is applied: Multidomain Option = Coupled Partitioning 2. Expert Parameters solver target reduction scalar = 0.01 (Default) The default value is set not to spend too much time on solving temporary equations, usually due to the advection term still in progress. A tighter solution can improve convergence, i.e. setting a value 0.005. mg small grid = 20 (default) The AMG solver does not agglomerate across parallel partition boundaries, nor domain interfaces of any kind. Therefore, the number of equations at the coarsest grid level could still be too many that the “direct solver” is never used and they are only smoothed, which results in a slow target reduction rate within the iteration. 3. Smaller number of partitions In case a different partitioning method does not help, then a smaller number of parallel partitions might be useful.