A semiempirical methodology to characterise a family of microvibration sources

It is well documented that reaction wheels are among the most significant microvibration sources in space applications. These components, despite being nominally identical, can show differences in the generated signals due to manufacturing imperfections in their internal elements, such as ball bearing, internal and external race. In this article a methodology to account for those variations in microvibration predictions is proposed, aiming at generating a disturbance input matrix that encompasses the effects of a family of reaction wheels. With such a tool, it is possible to provide a more accurate microvibration budget at an early stage of the mission, reducing the uncertainty margin usually applied to quantify reaction wheel effects on the structure. As a consequence better designs are produced faster and cheaper. This allows for more flexibility in the mission design and reduces the degree of uncertainties in the predictions. Furthermore, it is shown that the proposed approach is able to characterise the effects of the entire family of wheels by considering only a limited number. The methodology is validated by assessing the microvibration excitation on different structures, including a real space structure with various reaction wheel mounting configurations.