Mechanics of ice detachment applied to turbomachinery

Flying in icing conditions is a real hazard for aircraft as they can undergo potential disastrous increase in drag, reduction in maximum lift which lead to an increase in fuel consumption. Additionally pitot tubes and other sensors can become blocked or their operation compromised. Ice shed from other parts of the aircraft can enter the engine and lead to blade damage. Whilst ice protection systems are comonly used on propellers, the potential benefit of applying them to a fan have not, as yet, been considered suffi cient to o set the cost and energy penalties of such system. As engines become larger, it is more diffi cult to contain ice and self-shedding becomes an increasing hazard for the nacelle and other parts downstream of the fan. The main objectives of the project were to determine the mechanical proper- ties of ice such as might form on an engine fan, in order to help Rolls-Royce in building a finite elements model able to simulate ice shedding from fan blades. Lots were written about ice however only little information about the mechanical properties of impact ice was available in the literature and the values which were, were generally not applicable in the case of aeroengine in icing conditions. According to the literature and from Rolls-Royce photos and films of ice shedding from fan blades, self-shedding mechanisms were ruled by adhesive shear strength and tensile strength. Therefore, the experimental part of the project consisted of measuring these two mechanical properties as well as the density, the sti ness and the grain size of ice grown on titanium substrate. Two test rigs were used to measure the mechanical properties: the \mode I" and the \shear" test rig. The mode I test rig was already available and was only modified in order to test more specimens during each run. This test al- lowed to measure the pressure needed to remove the ice from the substrate in a running icing tunnel. Using equations from the literature, fracture energy, fracture toughness and tensile strength were calculated. The influence of ambient total temperature, cloud liquid water content and tunnel wind speed were investigated. Tensile strength was found to be increasing as the total temperature is decreasing, decreasing as the LWC is increasing and going trough a maximum as the tunnel wind speed is increasing. Values obtained lied in the range from 0.6 to 1.5 J.m-2 (corresponding to between 2 and 10 MPa) which is, in general, higher than the ones reported by other authors. This difference can be explained by the fact that the mode I test was conducted in a running icing tunnel while the previous authors have conducted the mechanical test after the tunnel has been stopped. In parallel, finite element models have been developed and results similar to the experiments were obtained ...[cont.].