Characterizing the Thermal Behavior of Pyrolytic Graphite Sheets (PGS) at Low Interface Pressures

Abstract

As the United States Navy continues to pursue its goal of developing fully electric ships the cooling of the critical electronic components on board must be solved. One of these critical components is the integrated Power Electronics Building Block (iPEBB); a universal converter that is programmed for its specific application when installed. The iPEBB is a modular unit that can be easily swapped by a single person. This unique modularity has led the Navy to pursue the design of a dry interface liquid cooling system to cool the iPEBB. This means that no liquid can cross the boundary of the iPEBB and thus the cooling system must be separate.

In this thesis, an integral portion of the dry interface cooling solution, the thermal interface material (TIM) between the cold plate and iPEBB, was explored in a multitude of ways. First, commercially available TIMs were investigated for their thermal behavior at pressures less than 10 PSI as well as their structural qualities and usability metrics. Pyrolytic Graphite Sheets (PGS) were chosen to be investigated further. Second, a fourth order thermal conductivity model for PGS as a function of interface pressure was derived in the 0 – 10 PSI range. This model is important as it allows engineers to have conductivity inputs for the PGS in any thermal modeling done for future iterations of the iPEBB or in other systems where PGS is used as a TIM. Third, the design and testing of an experimental rig (PPR) for testing thermal interface materials under various average pressures and pressure profiles was presented. An empirical model was developed that demonstrates the effect that interface pressure profile has on component temperatures with PGS as the acting TIM between the cooling solution and the heated system. Finally, using the conductivity model, CFD simulations were run of PPR experiments. These simulation results were then compared to the results of the PPR experiments and it was discovered that using the conductivity model for PGS as an input in a CFD simulation is an effective way of modeling the contact resistance of PGS as a function of pressure. The effectiveness of the conductivity model – CFD simulation setup has a mean error of 1.4C ± 1.3C between the simulation’s outputted average resistor temperature and the actual average temperatures measured.

The experiments and simulations conducted in this thesis provide a blueprint for the necessary steps required to thermally model not only the iPEBB dry interface cooling system, but also other systems that might use PGS as a TIM, using CFD. The information in this thesis will also help researchers model the thermal behavior of the iPEBB cooling system once a clamping mechanism for the iPEBB structure is designed.