In industrial environments the operating time of equipment is a critical factor. Through predicitve maintenance the necessary inspection tasks that require machine downtimes are minimized. For industrial fans that evacuate very hot mediums or air with a high content of abrasive dust or particulates there is currently no possibility to reliably sense the wear and tear of the fan blades during operation. This work investigates the use of a radar sensor for estimating the wear of such fans by sensing vibrations and rotational imbalances. For this task a frequency modulated continuous wave (FMCW) radar sensor is evaluated and suitable signal processing algorithms are developed to extract the desired information. Furthermore, the optimal sensing position, modulation waveform and fan optimizations are considered. This paper focuses on the basic signal processing techniques necessary as well as the measurement setup, simulations and first measurement evaluations.