Current forecasts suggest that, by 2030, at least 50% of telecommunication satellites will use electric propulsion (EP) as the only propulsion system on board. The ever-increasing operations time has led to a substantial increment of the amount of propellant stored on-board, from the initial 200-350 kg to present-day masses of 800-1500 kg. Despite the need to know the available propellant mass during operations, the retrieval is still challenging, and inaccurate, as no technological alternatives have been proven to satisfy with the needed requirements for long duration missions.

Recently we have proposed a new gauging method that uses TLR-9 hardware components. The method, called Improved-PVT method, is based on a better understanding of the thermal properties of the stored xenon. Laboratory experiments and theoretical work demonstrated an accuracy improved by a factor of 8 compared to classical Pressure-Volume-Temperature retrievals [A. Soria-Salinas et al., 2017]. In fact, it gives an error of mass gauging of 0.1% with respect to the initial mass, at a pressure of about 70 bar.

This method has been implemented in the PVT-GAMERS experiment, selected to fly at the ESA Fly Your Thesis! parabolic flight campaign, between October 22 nd to November 2 nd . As a technology demonstrator, it consists of a suit of 6 small-scaled and pressurized CO2 tanks, sensed with pressure (P) and temperature (T) sensors, with a heating duty cycle and real-time data processing. The PVT-GAMERS experiment will fly on three Airbus A310 Zero-G flights, where micro/hyper gravity-loads will allow to demonstrate the robustness of the method against thermal gradients, and i) simulated thruster ignitions, ii) external accelerations and iii) propellant management operations scenarios. These flights will also increase the TRL of the full system from 4 to 6.

In this work, we shall present the development, testing and ground calibration of the PVT-GAMERS experiment, including: 1) ground tests, assembly validation, expected results and operational procedures estimation; and 2) calibration of mass retrieval algorithm applied over a CO2 PVT-GAMERS engineering model (EM). Ground tests results indicates a relative error expected for the mass retrieval on flight model ground tests < 1.4% for CO2 gas at EOL conditions. Furthermore, a compromised solution shall be reached between retrieval expected accuracy, cooling rate for the retrieval application, and sensitivity to ambient condition changes. The EM tests suggest on this particular configuration to approximate the cooling ratio dP/dT using a least-squared-error procedure and over an interval of at least 900 seconds.