Meshing Procedure Development for Component Models in Electronics Thermal Simulation
Arola, Janne (2015)
2015
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2023121938445
https://urn.fi/URN:NBN:fi:amk-2023121938445
Tiivistelmä
The aim of this Master’s study was to find out the critical point from a meshing point of view where simplificated model temperatures will start deviating com-pared to a detailed model with a dense mesh.
The CFD simulation software is a good tool to reduce design cycles and prototypes. It can provide very accurate results when the inputs are reliable. The software can provide very poor results if the model, boundary conditions, or solution parameters are not represented properly. To adequately use the soft-ware, the user must become familiar with the built in functions and their limits of applicability. This also suits for a currently used software solving method where the user has to know the right ways of meshing for particular areas of interest. The user must also have a complete understanding of a thermal theory so he/she can judge when a calculated result is error based on an error in your input.
The work included finding a suitable test component model for the study, building used simulation models and environment. A simple mechanical structure was modeled for emulating a hand-held device environment for a component. The original detailed thermal model of the component was simplified for the meshing study purpose.
To get results, several steady-state thermal simulations were run. The results show a significant drop in solution time when using a simplified component with a coarse mesh. Component surface temperatures were fluctuating quite heavily between different mesh counts, but an ideal ratio was found as a result of this work.
It needs to be noted that the simplified models are not always applicable; when component junction temperatures are of interest, the highly simplified nature of the simple component model cannot predict the said temperatures and the us-age of detailed component models is advised.
The CFD simulation software is a good tool to reduce design cycles and prototypes. It can provide very accurate results when the inputs are reliable. The software can provide very poor results if the model, boundary conditions, or solution parameters are not represented properly. To adequately use the soft-ware, the user must become familiar with the built in functions and their limits of applicability. This also suits for a currently used software solving method where the user has to know the right ways of meshing for particular areas of interest. The user must also have a complete understanding of a thermal theory so he/she can judge when a calculated result is error based on an error in your input.
The work included finding a suitable test component model for the study, building used simulation models and environment. A simple mechanical structure was modeled for emulating a hand-held device environment for a component. The original detailed thermal model of the component was simplified for the meshing study purpose.
To get results, several steady-state thermal simulations were run. The results show a significant drop in solution time when using a simplified component with a coarse mesh. Component surface temperatures were fluctuating quite heavily between different mesh counts, but an ideal ratio was found as a result of this work.
It needs to be noted that the simplified models are not always applicable; when component junction temperatures are of interest, the highly simplified nature of the simple component model cannot predict the said temperatures and the us-age of detailed component models is advised.