Unsteady Performance Predictions for Low Pressure Turbines

Project Description

Modern jet engines, in a push for greater efficiency, use increasingly larger diameter fans, rotating at lower RPM. The fan is driven by the low pressure turbine (LPT), and running a low pressure turbine at lower speeds presents a severe technical challenge. The lower relative velocity between moving and stationary blades means that less work can be extracted per stage, which in turn requires more stages and results in a heavier, less efficient turbine. The efficiency of the LPT is critical: a shortfall of one percentage point in the efficiency of the LPT results directly in the loss of almost one point in overall engine efficiency. The resulting increase in fuel burn would have a major impact on payload, range and operating cost, and significantly lessen the economic and environmental advantages offered.
The solution to the problems of weight gain and efficiency loss is (a) to increase the flow turning in each blade row (permitting fewer stages, and reducing weight), and (b) to accomplish the turning in each blade row with fewer blades (reducing weight and reducing skin friction losses). Both options (a) and (b) push the blades closer to flow separations and an abrupt loss of efficiency. Unless we have an accurate flow prediction capability, we cannot determine whether our attempts to improve efficiency and save weight will result instead in a significant efficiency loss.
The prediction challenge is made worse by the fact that in an LP turbine, at high altitude cruise conditions (where most of the fuel is burned), the flow on much of the blade surfaces may be at the point of transition from laminar to turbulent flow. It is known experimentally that unsteady effects on transitional flow can be significant, with transition from laminar to turbulent flow on one blade row being triggered by the wakes from upstream blades sweeping over the blade surfaces. Mathematical models that are designed to take account of unsteady transitional flow have been developed, but are at best in their infancy. The LPT is thus an ideal candidate for a thorough exploration of the difference between steady and unsteady flows in turbomachinery – as revealed by our current CFD techniques.