Water droplet erosion of aeroengine fan blades

Abstract

A droplet of water may seem fairly innocuous, but accelerate it to a sufficient velocity and it can fracture diamond. When aircraft fly through the atmosphere, droplets of water are ingested into the aeroengines. These impinge the fan blades at high-speed, resulting in Water Droplet Erosion (WDE) of the leading-edge. Despite decades of reduced aeroengine efficiency and increased risk of unstable blade vibrations, there had been little previous investigation of WDE in this context. Increasingly demanding efficiency targets has transformed WDE from a form of wear that can be tolerated by aeroengine manufacturers, to one that needs to be minimised or prevented. To enable the design of in-service solutions, this PhD project sought to establish the foundations of a theoretical and modelling framework for the WDE of aeroengine fan blades. Two major achievements were made. The first was to understand the effect of solid surface curvature on a high-speed droplet impingement. The radius of a typical water droplet (encountered in-service) and the radius of curvature of a typical fan blade leading-edge are approximately the same; this is unlike other contexts, where the typical droplet radius is at least an order of magnitude smaller. This meant the solid surface could not be assumed planar (flat), as in previous work. Both theory and experiment were used to investigate: analytical modelling was extended to consider a curved solid surface and, by developing a new test rig, planar and curved test specimens were subjected to high-speed impingements of water. The results showed the curvature of the solid surface significantly affects the impingement in aeroengine fan blade WDE— an asymmetric (instead of the usual axisymmetric) impingement occurs. The second major achievement was making (for the first time) full-field measurements of a solid’s response to high-speed impingement. Within this achievement is also the interpretation of the measured fields by a new implementation of an analytical/numerical modelling approach. This work enabled the assumptions regarding WDE to be probed in a new and insightful way. Tens-of-thousands of individual displacement measurements were taken at a rate of 5 MHz. Grid Method was used to make the measurements, which were subsequently interrogated with the analytical/numerical iv modelling. The results revealed that the widespread assumption of a rigid solid surface is incorrect; only a model that included the effect of solid surface compliance accurately predicted the fields measured. This accurate model also demonstrated that a very low proportion of a droplet’s kinetic energy is transferred to the solid, contrary to the previous assumptions of some. In addition, the macroscale stress-state predicted could not explain the micro-plasticity observed in the early-stages of the WDE of metals, raising important questions for further work. Finally, the results of both sets of experiments supported a new direct-characterisation methodology that improved the ease and level of characterisation for the high-speed impingements generated. This work makes a small but significant step towards establishing the foundations of theory and modelling to support combating this form of erosive wear on aeroengines.

More information

Identifiers

ORCID for Robert Wood: orcid.org/0000-0003-0681-9239

Catalogue record

Export record