Rotordynamics Experimental Validation of a Sliding-Blade Architecture for an Inside-Out Ceramic Turbine

Abstract

The integration of monolithic ceramic blades into sub-megawatt microturbines is a low-cost option for increasing Turbine Inlet Temperature and efficiency. The Inside-Out Ceramic Turbine (ICT) is a promising concept for the integration of ceramic blades by loading each blades in compression using a carbon-polymer composite rim to convert the blade radial loads to tangential hoop stress. High tangential velocities lead to elevated radial displacement of the rim and, therefore, the rotor hub needs to be able to maintain the contact with the blades for a large range of radial displacements. This displacements comes with hub structural challenges and rotordynamics considerations. For these reasons, blade tip speed have been previously limited to about 360 m/s. This paper presents a hub design that allows high radial displacement using the combination of inclined blade roots, inclined hub grooves and an axial spring. The contact between the blade root and the hub is maintained through the inclined planes by the axial forces from the spring creating internal friction in the rotor that can induce sub-synchronous rotordynamics instabilities. The onset of instabilities is investigated experimentally with cold spin tests of a simplified ICT prototype. The results first show that the concept remains stable up to the maximum speed tested of 127 kRPM (tip speed of 387 m/s) if the spring is designed such that it remains in contact with the blade roots at all time. On the other hand, when reducing the preload sufficiently to test the limits of the concept, the rotor first mode became unstable at 120 kRPM resulting in failure of the prototype. These results suggest that, provided a sufficient spring preload to prevent excessive relative motion, the blades can reach the desired radial displacements, removing the main constraint on ICT tip speed.