Reduced-order modelling of mineral dust deposition in turboshaft engine hot sections

Abstract

Operating in brownout conditions is a hazardous yet often unavoidable consequence of using rotorcraft in dry, arid environments. Significant quantities of the lofted dust cloud can be ingested into the helicopters engines where they cause damage in both the main gas path and secondary air systems. Molten particle accumulation on nozzle guide vanes has the potential to restrict the core mass flow and cause the engine to surge. Predicting the conditions under which this happens is complicated by the many separation and sorting processes that occur in the preceding sections of the engine. This contribution demonstrates a reduced-order, probabilistic methodology that can be used in conjunction with in-field particle sampling to predict the extent of nozzle guide vane damage and assess the risk of engine surge when operating in a particular environment. The processing of the raw ingested dust in the compressor and particle separator system is shown to be significant in accurately predicting the extent of damage and suggests that taking particle samples from the bulk dust cloud is likely to provide under-estimations of the actual damage seen in the engine.