Thermal Evaluation of Printed Circuit Board Design Options and Voids in Solder Interface by a Simulation Tool
Quad Flat No-Lead (QFN) packages have become very popular for turners, converters and audio amplifiers, among others applications, needing efficient power dissipation in small footprints. Since semiconductor junction temperature (TJ) is a critical parameter in the product quality. And to ensure that die temperature does not exceed the maximum allowable TJ, a thermal analysis conducted in an earlier development phase is essential to avoid repeated re-designs process with huge losses in cost and time. A simulation tool capable to estimate die temperature of components with QFN package was developed. Allow establish a non-empirical way to define an acceptance criterion for amount of voids in solder interface between its exposed pad and Printed Circuit Board (PCB) to be applied during industrialization process, and evaluate the impact of PCB designs parameters. Targeting PCB layout designer as an end user for the application, a user-friendly interface (GUI) was implemented allowing user to introduce design parameters in a convenient and secure way and hiding all the complexity of finite element simulation process. This cost effective tool turns transparent a simulating process and provides useful outputs after acceptable time, which can be adopted by PCB designers, preventing potential risks during the design stage and make product economically efficient by not oversizing it. This article gathers relevant information related to the design and implementation of the developed tool, presenting a parametric study conducted with it. The simulation tool was experimentally validated using a Thermal-Test-Chip (TTC) in a QFN open-cavity, in order to measure junction temperature (TJ) directly on the die under controlled and knowing conditions. Providing a short overview about standard thermal solutions and impacts in exposed pad packages (i.e. QFN), accurately describe the methods and techniques that the system designer should use to achieve optimum thermal performance, and demonstrate the effect of system-level constraints on the thermal performance of the design.