Thermosyphon simulation code: transient thermal performance and experimental validation

Gravity-driven, and thus passive thermosyphons represent a step-change in technology towards the goal of greatly increasing efficiency of datacenters by replacing energy hungry fans of heavy and expensive air cooled heat spreaders whilst also providing a highly-reliable solution able to dissipate the rising heat loads demanded in a reliable and cost-effective manner. In fact, the European Union has launched a zero carbon-footprint target for datacenters by the timeline of 2030. In the present paper, a newly updated version of the general simulation software previously presented at ITHERM 2017 and ITHERM 2020 is considered. For the industrial transition to the passive thermosyphon cooling solution, with its more "complex" flow phenomena, the availability of a general-use, widely validated simulation code that handles both air-cooled and water-cooled types of thermosyphons is of paramount importance for the successful transition to gravity-driven micro-scale two-phase cooling systems. The simulation code must be able to design and predict the thermosyphon-based cooling systems with high accuracy and handle the thermal-hydraulic behaviour of the different working fluid's flow regimes, besides that of the secondary side coolant. Additionally, the simulation code must be able to determine the steady-state and transient thermal-hydraulic performance for optimal designs. Reason why the simulation code is continuously validated versus new thermosyphon designs and new working fluids. A transient validation of the thermosyphon simulation code is performed here versus experimental data gathered for a 7-cm high mini-thermosyphon design with water as the coolant in the condenser, which is being considered for the cooling of high-performance servers. The comparison shows that the simulation code can predict transient thermal performance under heat load values from 50 W to 200 W for a single heat source, including cold start-up operations with an accuracy of +/- 2 degrees C in terms of junction temperature. Validations analyzing junction temperature and working fluid pressure over time are presented in this study. The simulation code is also able to determine the working fluid mass flow rate generated by the gravity-driven thermosyphon, which is the key to correctly predicting the critical heat flux and ensure safe upper operating limit of the LTS cooling system.