New wind tunnel facility for icing experiments on models of turbofan compressor surfaces

New wind tunnel facility for icing experiments on models of turbofan compressor surfaces

Compressors of modern turbofan engines are often sensitive to ice crystal accretion which can occur if aircraft fly through cloud regions associated with storm complexes. Such flight paths are occasionally necessary, and if ice accretion within the compressor does occur, the ensuing engine performance degradation can range from a mild reduction of effciency through to a complete loss of power. Many parameters influence the initiation and rate of ice accretion but the physical processes and parameters governing the sensitivity of compressor surfaces to ice accretion are not fully understood. To develop reliable engineering models that can be used to aid the design and operation of compressors under icing conditions, further experimental data is needed. Existing icing wind tunnels around the world are very capable, but the operating costs for these wind tunnels are typically very high. The objective of this work is to establish a new icing wind tunnel that has modest operating costs and yet can also facilitate hardware testing, instrumentation development, and fundamental studies of ice crystal icing at compressor-relevant flow conditions.

A wind tunnel arrangement was proposed involving water droplet freeze-out using liquid nitrogen evaporation followed by natural particle melting through dilution with warm air. The viability of the arrangement was demonstrated theoretically using a conservation of energy analysis. Thermodynamic performance of the facility is dictated largely by the availability of the liquid nitrogen and the proposed operating concept specified using a maximum of 20 litre of liquid nitrogen per run in the facility within 2 minutes to achieve the target operating conditions for the facility: flow speed around 50 m/s, temperatures around 0 degrees C, and total water content up to 10 g/m3 with melting ratio up to 0.2.

The hardware developed for the facility includes an icing jet generator with nozzle exit diameter of 170 mm, and an open circuit wind tunnel. Ice particles are generated by injecting water from atomiser nozzles into a mixture of recently-evaporated liquid nitrogen and air which provides a low-temperature medium for the freezing process. A liquid nitrogen receiver and valve system was designed to supply the liquid nitrogen into the evaporator at a metered and controllable rate. The suspended ice particle mixture is then delivered to a diffuser with perforated walls through which further air is injected for the purpose of raising the temperature of the mixture, and generating some natural melting of the ice particles. The icing jet nozzle contraction, which is attached to the downstream end of the diffuser chamber increased the flow velocity and decreased the non-uniformity of the flow velocity at the exit of the jet.

The performance targets for the facility have mostly been achieved, and this has been confirmed through experimentation with individual components and with the facility working as a combined unit. Experimental results have demonstrated a generally favourable agreement with the energy equation analysis, and with results from Computational Fluid Dynamics (CFD) simulations. The probe traversing system developed for the icing jet nozzle exit flow enabled quantification of the velocity uniformity at the exit of the icing jet generator. Within a core flow diameter of 140 mm, the flow speed was 28.1±1.1 m/s. This speed is somewhat lower than the target figure of 50 m/s, but it is expected that this can be readily rectified through installation of a higher power blower. The jet exit temperature uniformity was also reasonable: over the same jet core flow region at one particular operating condition, the temperature was -9:1±1.9 degrees C. However, results from the isokinetic total water content probe developed for this work indicate that improvements in the uniformity of the water distribution are needed.

Initial experiments with a 12.7mm diameter cylindrical test article have demonstrated some ice accretion at glaciated conditions, and more significant accretion was registered with a non-zero melting ratio operating condition. However, additional improvements are needed in the facility and in the instrumentation used to quantify the facility performance. The introduction of humidity control, melting ratio control, temperature control, and more extensive instrumentation having a faster-response time is achievable in the near term and is expected to have significant impact on the quality of data derived from the new icing wind tunnel in the near future.