Success Stories: NASA Glenn Turbine Cooling

NASA Glenn Research Center uses GridPro to analyze turbine heat transfer.

Program Development Company would like to acknowledge the late Vijay K. Garg for his groundbreaking work in turbine cooling CFD. Work is continued today in this field by his colleagues at NASA Glenn Research Center including James D. Heidmann.

Gas turbine engines today are ubiquitous in large scale power generation. Ships use gas turbine engines because of their high efficiency and power-to-weight ratio, gas turbines are used in industrial electrical power production because of their high efficiency, and modern military and large commercial aircraft all use turbofan jet engines because of their increased performance and efficiency at higher speeds and altitudes. Today though, with increasing fuel costs, and fuel as already the single largest operating expense for any airline, there is pressure to maximize the efficiency of these engines, which produces many challenges.

A basic gas turbine engine consists of 3 stages: Stage 1 is a compressor which compresses air into stage 2, the combustion chamber. Fuel mixed with the air is burned at high temperature and pressure in the combustion chamber, which is then expelled through stage 3, the turbine. The turbine collects energy from the escaping gas, which in turn powers the compressor as well as external components in most cases. In turbofan engines for example, some energy collected by the turbine from the escaping gas powers a fan via shaft, and some escapes through the rear to provide thrust.

It is known from thermodynamic analysis that increased turbine inlet temperature leads to increased overall efficiency. Modern engines are designed to operate at turbine inlet temperatures around 1800-2000 K, but these temperatures greatly exceed the maximum allowable temperatures for metals. Even the most advanced metal alloys and ceramics available have their limits. In order to provide acceptable turbine life and safety, an effective means of cooling these components must be utilized.

One method of providing cooling to turbine blades is the insertion of cooling holes on the leading edge of hollow turbine blades, through which cooler air is injected into the boundary layer on the surface of the blade. This is known as film cooling. Since air is diverted from the compressor stage of the engine though, it must be carefully balanced in order to minimize power loss from the compressor while providing the necessary cooling to protect the turbine. In this situation, finding this balance can be difficult. Design variations can be too difficult or expensive to test thoroughly. This is where computational fluid dynamics simulation can be applied.

Computational fluid dynamics (CFD) can be used to simulate flow and heat transfer in this type of environment, however there are several challenges to overcome in this case as well. The internals of a gas turbine present both complex geometries (multiple periodic turbine blades, blade tip clearances, cooling holes and passages) and complex physics (high speed, mixed temperatures, turbulence). Complex physics like this demand robust CFD codes, as well as robust grid generation. In such an environment, high-quality conformal hexahedral computational grids are preferred because they are verified to produce more accurate CFD results, but complex geometry like this can make generating grids a time consuming task.

Leading gas turbine engineers and researchers like those at NASA Glenn Research Center and the Whittle Labs at Cambridge University are using GridPro to generate grids on complex geometries like this. Vijay Garg at NASA Glenn Research Center performed numerical simulation of an AlliedSignal film-cooled rotor blade with 172 cylindrical cooling holes using Glenn-HT CFD and heat transfer code with a GridPro generated multi-block hexahedral grid. No other hexahedral grid generation software had the flexibility needed to grid on such a complicated and intricate geometry, while maintaining grid quality and the ability to easily adapt to geometry changes.

GridPro's topology based grid generation allows for adaptation to geometric changes for the same structure. Blade geometry and cooling hole placement, angle, and shape could all easily be modified and the grid re-generated with minimal user effort due to this system. In addition, GridPro offered a flexible component based approach to assemble the hole topologies. The user is able to generate the grid for one hole, and automatically link multiple holes using GridPro's component feature. In the case of Vijay Garg, he was able to link up 172 holes just by generating the grid for a single hole. With the help of GridPro grid generation, He calculated variable heat transfer coefficients for this complex turbine geometry. This was one of the earlier applications of thorough CFD analysis of a complete film cooled turbine blade.

Today, James Heidmann and many others at NASA Glenn Research Center continue to use GridPro to help with turbine research and development, and GridPro has even more powerful tools for this type of case. A special hole topology feature will automatically generate topology for multiple holes given only the surfaces through which the holes pass. Generating high-quality hexahedral grids on intricate geometries requires a flexible approach to grid generation such as GridPro's topology system. Coupled with its unique dynamic boundary conforming technology, GridPro can help solve the most complex grid generation problems with high quality grids, leading to accurate results.

For more information about the turbine research conducted at NASA Glenn Research Center, see the list of references provided or visit their website. For more information about GridPro, click here.

Program Development Company