At the chair of Space Systems Engineering, students and staff work on the development of small propulsion systems in a wide thrust capability. Especially in recent years low thrust propulsion systems have gained an increased amount of interest in the entire department of Space Engineering. To support the development of these propulsion systems, there is a need for the testing of micropropulsion systems that provide a thrust in the 1 ?N to 5 mN range. This capability complements the existing measurement equipment of the Delft Aerospace Rocket Thrust Stand facility of 0.5 mN to 1 N. To provide a solution to this need, the research question for this research is phrased as: "How can the DARTS facility be upgraded to measure thrust levels in the 1 ?N to 5 mN range and impulse bits in the 1 ?Ns to 1 mNs range?". Using numerical modelling the answer to that question is found in an upgraded design of the TB-2m thrust stand that was developed in 2010 by Perez-Grande. By introducing several improvements, amongst which are the replacement of the sensor system by a capacitive displacement sensor and the use of a segmented counter mass, the range of the TB-2m has been extended and the accuracy has been improved to meet the new requirements. A complementary calibration system is designed that allows an in situ calibration for the measurement of both the thrust and impulse bits. Where typical calibration actuators supply a force that is non-linear with engagement distance and have to rely on displacement measurements to linearize the calibration force, the new actuator is able to provide a force that is independent on the distance to the target. Using a specially designed solenoid that has a linearly varying turn-density along its length, the magnetic field is shaped to provide a constant magnetic gradient. The thrust stand and calibration system are manufactured and tested. Using a previously tested cold gas thruster that is provided by Bradford Engineering, the complete thrust measurement system is validated with hardware-in-the-loop. This process has shown that the pendulum is in a state constant of oscillation. It is expected that this oscillation is removed by the introduction of a Foucault damper in the next design iteration. The validation process has shown that the thrust range is on par with the requirements. Also impulse bits can be measured, but the constant state of oscillation prohibited the detection of the smallest impulse bits of 1 ?Ns. Future experiments have to show whether the required and predicted accuracies are achieved.