As the design phase of MOVE-II advances further and CDR level is in sight, thermal design of the entire satellite gets more and more important. During the last weeks, a thermal prototype was built, and experiments in vacuum environment were carried out. This was done to validate our numerical models, and to get meaningful predictions of temperatures on orbit.
Our prototype reflects all major hardware components, built from similar materials to the real satellite: The interior board stack is made from test PCBs, hold together through PC104 connectors and a standoff system. All interior parts are mounted in the aluminum structure, designed in the official 1U CubeSat dimensions. In order to reflect final thermo-optical properties as well as possible, the external boards are equipped with spare solar cells on their respective position. Built together, the model already looks convincingly CubeSat-like…
The model does not include any electronically functional subsystem boards. To simulate expected heat loads, major sources are represented through Kapton heaters and resistors. When operated with loads corresponding to our expected power budget, the model can provide a good estimation of expected on-orbit temperatures.
During the summer break, we performed a series of thermal vacuum experiments with the satellite. Within a week of experimental work at the institute’s thermal vacuum chamber, we were able to try different stacking schemes and get thermal balance data for different modes of operation (like Hot and Cold Cases). The experiments already revealed that major heat sources in the battery’s vicinity could be an issue – especially long operation of the power-demanding communication systems could cause dangerous overheating of the sensitive part. In satellite thermal design, overheating has to be avoided, as no active cooling mechanism is available on a CubeSat. A battery getting too cold would be less critical, a built-in heater would prevent freezing damage in this case. To avoid excessive temperatures, the battery will probably be kept away from the S-Band COM board, and its operation will be limited to a few minutes per orbit.
Right now, the experimental data are being used for correlating numerical simulations. We’re making good progress in this direction, and are able to predict satellite temperatures within an accuracy of 10 K.