Thermal Resistance Modeling And Structural Optimization Techniques For Microfluidic Cascade Of LTCC Components

The package density of electronic devices is increasing,and at the same time,the existing electronic devices are developing in the direction of integration and miniaturization,which inevitably leads to a large accumulation of heat.Currently,the junction temperature of devices under different conditions is generally obtained by means of simulation analysis,and the accuracy of the simulation analysis results is based on the complexity of the model,and an overly detailed model will greatly increase the complexity of the problem and the development cycle of the thermal design.For the purpose of shortening the thermal design cycle and improving the reliability,it is necessary to establish a thermal resistance network model of the system to achieve temperature prediction under specific conditions,which is necessary to improve the working reliability of heat generating components and shorten the thermal design cycle.In this topic,a new multi-stage thermal structure is proposed to address the urgent thermal problem of LTCC components in communication electronics,and how to establish its accurate thermal resistance network model is the focus of this research.The main research contains the following parts:(1)Structural design of multi-stage heat sink: A multi-stage heat sink structure is proposed according to the heat dissipation requirements,which is composed of heat source,Low Temperature Co-fired Ceramics(LTCC)substrate,thermal interface material and carrier plate in order of connection.Micro runners are set up inside the substrate and metal inserts are arranged to improve the overall thermal conductivity of the substrate.(2)Establishment of micro-runner thermal resistance model and optimization of crosssectional shape: Establishing an accurate thermal resistance network model in the "microrunner-embeded" region is the key to establishing a thermal resistance network model for multi-stage heat sinks.The flow channel is divided into auxiliary flow channel and main flow channel by taking the position of the embedded body as the dividing line.Three different thermal resistance network models are proposed based on the heat transfer path for a single auxiliary runner micro-element,and the thermal resistance network models are compared with the simulated values from numerical simulations by selecting certain structural parameters,and it is found that the average absolute deviation of the thermal resistance network models established by the transverse three-stage calculation method(HTCM)from the simulated thermal resistance obtained from numerical simulations does not exceed 4%.On the basis of the auxiliary flow channel thermal resistance model,two thermal resistance network models for the main flow channel microelements are proposed,and it is found that the average absolute deviation of the main flow channel thermal resistance model obtained by the longitudinal center calculation method(LCCM)from the simulation results does not exceed 5%.The cross-sectional shapes of the main flow channel and auxiliary flow channel are optimized with thermal resistance and pressure drop as the objective functions,respectively.The foundation for the subsequent study is laid.(3)Establishment and validation of single heat source cascade thermal resistance network model: Firstly,the thermal conductivity of the thermal interface material is measured experimentally,and the thermal resistance network model of the single heat source system level is established on the basis of the thermal resistance models of the main and auxiliary flow channels.The error rate of the cascaded thermal resistance model is calculated for different inlet flow rates and heat source powers.When the inlet flow rate increases from0.5m/s to 2m/s,the error rate is 4.076% at the maximum and 0.205% at the minimum,with an average absolute error of 2.35%.When the heat source power increases from 54 W to216W,the error rate is 2.89% at maximum,1.27% at minimum,and the average absolute error is 2.20%.(4)Establishment and accuracy verification of the thermal resistance network model of multiple heat sources: Based on the idea of thermal resistance matrix,the thermal resistance network model of multiple heat sources with an inlet flow rate of 1.25 m/s is established,and the effect of different powers on the error rate of the thermal resistance network model of multiple heat sources is studied with four heat sources as an example,and the error rate is 0.029% at maximum when different powers are applied to the heat sources respectively,which has a good temperature prediction effect.