Environmental particle rebound/deposition modeling in engine hot sections

Abstract

The aircraft engine hot section is most vulnerable and failure prone to environmental particle ingestion, which, particularly for helicopters, can cause detrimental effects ranging from reduced performance to complete engine failure. The objective of this work is to develop an analytical tool to assess environmental particle impact on engine hot sections. The current state of the art in experimental and analytical research on environmental particle ingestion related to engine hot sections was reviewed, with emphasis on sand particles. From these efforts, the available experimental data for model calibration were identified, and an innovative particle rebound/deposition model has been developed. A semi-empirical approach is selected to model particles bouncing off metal surfaces, where the coefficients of restitution measured in a temperature range of 297–1323 K are used to calculate particle bounce-back velocity components. The developed deposition model is based on non-dimensional parameter analysis over more than seventy experiments related to particle deposition in engine hot sections. The metal surface temperature, one of two critical parameters in particle deposition, is also included in the model. The model was successfully implemented into commercial software and checked step by step. It was calibrated by two cases: sand [Arizona road dust (ARD)] particle impingement on a circular plate and Mt. St. Helens volcanic ash impinging on a first-stage air-cooled nozzle guide vane (NGV). For the former case, the calibrated model predicts fairly well the variation of particle deposition rate with flow/particle temperature. The latter case indicates that the particle deposition rate at engine operating conditions can be assessed by the developed model. Due to the lack of experimental data that would permit a full calibration/validation, for the time being the model can be only used under limited conditions. As additional relevant experimental data appears, the model will be continuously improved.