Speaker
Description
The thermal-hydraulic design and analysis of the PLATO focal plane assembly (FPA) and normal-front end electronics (N-FEE) thermal-vacuum calibration experiment is presented. Optical calibration of the instrument, which consists of a 300mm diameter Ti-6Al-4V FPA with 4xCCDs, connecting via flex-cables to the 15W N-FEE, requires two separate thermal control loops with temperature ranges of 150K – 210K and 220K – 308K, respectively. The FPA connects to the PLATO telescope at four different points, and operates with time-varying power dissipations at each point. The N-FEE conductive interface consists of three mounting points that are thermally and geometrically asymmetrical. Both instruments expect a temperature variation of less than 1K across their respective interfaces. A network of gold electro-plated copper straps connected to a static liquid nitrogen vessel, is used to deliver thermal control for the FPA. This temperature control system has been custom designed using ESATAN. A bespoke phase change subroutine has also been developed to capture the physics of time-dependent pool-boiling and evaporation of liquid nitrogen during the experiment, which allowed the investigation of several normal control and potential failure scenarios. This subroutine interacts with the geometric model via a non-geometric node. Heat absorption, heat transfer coefficients and the development of temperature gradients across the vessel and wider system are predicted. Cooling of the N-FEE is delivered using two custom designed closed-loop pipe networks connected to a heat pump circulating Galden heat transfer fluid across the system. This has been developed using a purpose-written thermal-hydraulic program developed on Matlab. The program predicts flow rates, pressure drops, heat flows and temperature gradients of the working fluid throughout the system, and has been used to specify compressor flowrates and pipe sizes for the full working range. The design objective is to achieve a temperature variation of less than 1K across both pipe loops without the use of supplementary PID controllers. In this presentation, we discuss the thermal development and design of both purpose-made systems and their supplementary mathematical tools and we provide comparisons to currently available experimental data.
