CFD Press Room

STAY UP TO DATE

RSS Feed . Get our RSS News Feed.

Bookmark our RSS feed today and make sure you never miss any of the latest news from CD-adapco.
What is RSS?

Press Releases
Press Releases - 2008
Press Releases - 2007
Press Releases - 2006
Press Releases - 2005

dynamics Magazine
dynamics Issue 2.01
dynamics Issue 27
dynamics Issue 26
dynamics Issue 25

e-dynamics
Current Issue
Archive

Start Sending Me
dynamics
e-dynamics

Thermofluids in Turbomachinery
Conjugate Heat Transfer; Dynamic thermal response

Fred Mendonça, CD-adapco

 

Fig:01 The traditional process in blade cooling analysis – the external flow, internal flow and metal thermal analyses are performed separately.

Fig:01
The traditional process in blade cooling analysis – the external flow, internal flow and metal thermal analyses are performed separately.

The recent advances brought to you in STAR-CCM+ are changing the standard in Thermal analysis for turbomachines. Those pains, particularly caused by limitations in complex geometry handling and fluid-solid contiguous meshing, are set to become just a memory.

CD-adapco’s geometry and automated meshing
technologies have been successfully applied to the
combined fluid and thermodynamics in a wide range of
turbomachinery flows. In this article, we expound two
key areas; turbine-blade cooling and turbocharger turbines. From
geometry to solution, full conjugate-heat-transfer analyses reduce
from weeks to days.

Turbine-blade Cooling:
Making the analysis process more efficient The high-temperature of operation in aero and industrial gas turbines, require active cooling to prolong turbine blade life, with cooler air bled from the compressor to flow paths cut inside the blade to provide direct cooling of the metal. The analysis of these geometries presents a significant challenge, including extreme physical conditions and highly complex geometries with many small features relative to the overall size of the domain, stretching computational and software resources to the limit. The traditional way of modeling turbine blade cooling is to split the system into separate functions, external flow, internal cooling and solid thermal analysis, all coupled together through common boundary conditions (Figure 1). The flow analysis is completed first, using a “guessed” surface temperature or heat flux distribution (which is then modified later in the “loop”), before the other elements of the analysis are carried out in turn. The workflow continues until all separate parts of the system have interacted often enough for the combined system to ‘converge’. Furthermore, the internal cooling path is often geometrically so complex that 1-D analyses are performed, and supplemented by experimentally correlated heat-transfer coefficients.

Fig:02 - Rate of heating of a turbocharger turbine
Fig:02 - Rate of heating of a turbocharger turbine

Many project hours continue to be spent in this area, amongst most manufacturers and maintainers of small and large gas turbines, and in collaborative efforts [1]. There is always need to improve the design and optimize the analysis with the ability to mesh continuously between the flow paths, including the solid, and resolve the internal passages with sufficient fidelity brings some significant benefits. With this ability, the ‘process’ may be simplified to that shown in Figure 2, where all ‘parts’ are
performed in one simulation, and requires no ‘iteration’ because all the parts are implicitly connected.

 

Fig:01 Full CHT analysis of a turbocharger including compressor, turbine and lubricating channels

Fig:03
Full CHT analysis of a turbocharger including
compressor, turbine and lubricating channels

Thermal transience response to load changes
Now take the fact that you can easily mesh continuously through the fluid and solid domains, and suddenly a whole new world of turbine analysis opens up. The combined system can be run dynamically so as to assess the thermal effects which changes in flow condition, or operational load conditions, make to the system. Previously such analyses would be prohibitively long and impractical so as well having to use 1D assumptions and mapped results, only a “snapshot” of the flow field at a set range of operating conditions could be studied without being able to analyze how the effects of transitioning between one operating point and another.

In a recent project performed on a dual turbocharger assembly, the rate of heating in the metal was assessed during a change in condition from low-load to high-load. The change in load alters the flow path into a bypass channel and wastegate, the metal heats up, subsequently cools down, and results in a low-cycle thermal loading which can lead to problems locally.

The CD-adapco solution
The advanced polyhedral meshing technology implemented in STAR-CCM+ allows the simultaneous and conformal meshing of all three “domains” considered in a turbine blade cooling analysis. Shared boundaries, or interfaces, are recognized and meshed so that the one-to-one connectivity is maintained; ensuring that simultaneous solution of both fluid and solid fields is carried out without the need for mapping or interpolation of boundary condition. Tools also exist to automatically ensure that the mesh within the cooling passages and around the blade tip, to the required level of detail to capture flow features and heat transfer correctly.

With a continuous mesh in place, CD-adapco tools have the ability to simultaneously solve for all the fields required in the blade analysis:

  • Flow, both primary gas path and internal cooling
  • Thermal, through both fluid and solid fields and any corresponding heat transfer
  • Stress, both mechanically and thermally induced in the blade, shroud and hub

The solutions provided by CD-adapco’s meshing and solution technologies help provide significant benefit, both analytical and financial, to the analysis engineer and the wider company. By reducing the number of pieces of analysis software from three or more down to one ensures that only one mesh needs to be built per geometry, only one software package needs to be learnt and no errors can occur in mapping from one solution to another. These benefits in turn lead to the freeing up of computer and man power resources helping to provide, more accurate, more numerous and more cost effective solutions.

REFERENCES:
1] AITEB-2 project www.aiteb-2.eu

OpenCube Drop Down Menu (www.opencube.com)