Infrared Thermography Techniques for Flow Visualisation and Heat Transfer Coefficient Measurement

Accurate knowledge of the boundary layer state on an aerothermal component is critical for predicting loss and heat transfer in design. Despite this importance, practical computational methods that predict boundary layer state remain unreliable thanks to the sparse, scattered data on which they are tuned. There is a need for experimental methods that can diagnose the boundary layer state on a component and measure the resulting heat transfer characteristics. This thesis develops experimental methods that use infrared (IR) thermography to visualise the boundary layer state on components and measure full-field heat transfer coefficients. IR cameras are a mature technology; however, the methods for their use in experiments are immature and leave opportunities for improvements. The aim of this thesis is to use a low-order modelling approach to analyse IR methods and identify routes for optimisations. Flow visualisation with IR uses the change in surface temperature that occurs at a change in boundary layer state due to changes in heat transfer coefficient and recovery temperature. A one-dimensional conduction model is developed to analyse the magnitude of this temperature difference. Distinct behaviours are shown between low- and high-speed flows due to the interaction of recovery temperature and heat transfer coefficient changes. The 1D model is validated experimentally in both regimes, and a pair of case studies are presented that demonstrate the use of IR flow visualisation for diagnosing an unknown boundary layer state in aerothermal testing. Existing methods for using IR to measure heat transfer coefficients rely on measurements of temperature amplitude in steady-state or transient experiments. They are, as a result, susceptible to noise, as well as to conduction errors that arise at the Biot numbers of 5 or less that are accessible in a rotating turbomachinery environment. A new method is developed that instead uses temperature phase-shift to calculate heat transfer coefficients. A Monte Carlo uncertainty analysis demonstrates that the phase-based approach is less sensitive to noise by a factor of 10-50 compared to a traditional transient method. This robustness allows the phase-based method to be used in rotating experiments, where short IR integration times increase measurement noise by a factor of 10. Phase-based measurements are applied to medium- and high-speed turbine rigs, producing first-of-their-kind, full-field measurements of heat transfer coefficient on rotor blades.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2025

Author dc:creator

• Davis, Will

Advisor dc:contributor.advisor

• Atkins, Nicholas

• heat transfer
• infrared
• thermography
• flow visualisation
• turbomachinery
• engineering

Licence

• https://www.repository.cam.ac.uk/bitstreams/80a1f13f-2699-42ba-b09e-249956833a65/download
• http://purl.org/NET/rdflicense/allrightsreserved

Language dc:language

eng

Author Identifier

0000-0002-4521-9351

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/398868