This thesis describes validated improvements in the modelling of micron-sized particle deposition within gas turbine engine secondary air systems. The initial aim of the research was to employ appropriate models of instantaneous turbulent flow behaviour to RANS CFD simulations, allowing the trajectory of solid particulates in the flow to be accurately predicted. Following critical assessment of turbophoretic models, the continuous random walk (CRW) model was chosen to predict instantaneous fluid fluctuating velocities. Particle flow, characterised by non-dimensional deposition velocity and particle relaxation time, was observed to match published experimental vertical pipe flow data. This was possible due to redefining the integration time step in terms of Kolmagorov and Lagrangian time scales, reducing the disparity between simulations and experimental data by an order of magnitude. As no high temperature validation data for the CRW model were available, an experimental rig was developed to conduct horizontal pipe flow experiments under engine realistic conditions. Both the experimental rig, and a new particulate concentration measurement technique, based on post test aqueous solution electrical conductivity, were qualified at ambient conditions. These new experimental data compare well to published data at non-dimensional particle relaxation times below 7. Above, a tail off in the deposition rate is observed, potentially caused by a bounce or shear removal mechanism at higher particle kinetic energy. At elevated temperatures and isothermal conditions, similar behaviour is observed to the ambient data. Under engine representative thermophoretic conditions, a negative gas to wall temperature gradient is seen to increase deposition by up to 4.8 times, the reverse decreasing deposition by a factor of up to 560 relative to the isothermal data. Numerical simulations using the CRW model under-predict isothermal deposition, though capturing relative thermophoretic effects well. By applying an anisotropic Lagrangian time scale, and cross trajectory effects of the external gravitational force, good agreement was observed, the first inclusion of the effect within the CRW model. A dynamic mesh morphing method was then developed, enabling the effect of large scale particle deposition to be included in simulations, without continual remeshing of the fluid domain. Simulation of an impingement jet array showed deposition of characteristic mounds up to 30% of the hole diameter in height. Simulation of a passage with film-cooling hole off-takes generated hole blockage of up to 40%. These cases confirmed that the use of the CRW generated deposition locations in line with scant available experimental data, but widespread airline fleet experience. Changing rates of deposition were observed with the evolution of the deposits in both cases, highlighting the importance of capturing changing passage geometry through dynamic mesh morphing. The level of deposition observed, was however, greater than expected in a real engine environment and identifies a need to further refine bounce-stick and erosion modelling to complement the improved prediction of impact location identified in this thesis.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:748784 |
| Date | January 2017 |
| Creators | Forsyth, Peter |
| Contributors | Gillespie, David ; McGilvray, Matthew |
| Publisher | University of Oxford |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://ora.ox.ac.uk/objects/uuid:61556237-feed-43cb-9f4a-d0aed00ca3f8 |
Page generated in 0.0013 seconds