An investigation of scaling parameters governing film-cooling

Thesis

The thesis describes an investigation into the fundamental scaling parameters which govern the process of film-cooling. This method of cooling has wide applicability in the hot portion of gas turbine engines. Well-designed cooling configurations and accurate prediction of heat transfer loads are essential in this extremely hostile environment. However, a great deal of the design data used has been acquired in constant properties experiments which are unrepresentative of engine conditions. At present, there is no consensus on how best to scale this data to the turbine environment.

In particular, the effect of varying the ratio of coolant-to-freestream density has been studied. Experimentally, this variation has been produced, as in the engine, by varying the the injection-tofreestream temperature ratio. For the three geometries tested (a single and double row of 3O holes and a 30 slot), the effect of this parameter was found to be very significant. The widely used mass flux ratio does not take this dependence into account, so that other scaling parameters need to be determined.

For injection rates where the coolant flow remains close to the surface, good scaling is achieved with the momentum flux ratio. All injection rates for the slot and double row occurred in this largely unseparated, "weak" injection regime. However, for the single row, the injection jets penetrated into the freestream at a critical value of the momentum flux ratio. At higher injection rates, in the "strong" injection regime, fluid mechanic scaling was best with velocity ratio.

A simple model is proposed, based on theoretical and experimental considerations, suggesting scaling parameters which successfully collapse all the data in each regime for different injection rates and distances downstream from injection. Alternative scaling parameters based on pressure and temperature measurements are also found to be highly successful.

The effect on heat transfer of the variation of properties through the injection layer at the wall-to-freestream temperature ratios encountered in gas turbines was also found to be significant. This has been quantified.

Experiments were performed using an Isentropic Light Piston Tunnel, a transient facility which enables conditions representative of those in engines to be attained. The results were interpreted using a superposition model, which is shown to be a valuable and concise method of characterising the effects of injection.

Authors: Forth, C; Forth, C. J. Patrick

• Publication date: 1985
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Subjects: Heat; Transmission; Gas-turbines; Cooling