|May 16, 2010, 19:16||
How to model intersecting blades in turbogrid
Join Date: May 2010
Posts: 2Rep Power: 0
I am trying to model a axial blade design. Im using Solidworks for the design and trying to use ansys workbench for fluid flow, velocity, pressure, torque and rotational velocity generated by an injected fluid. My problem is setting up turbogrid for rotational data. My design consists of blades connected to a common shaft(hub) but the blades do not have a shroud side they connect to the next blade in front of them. The design consists of only blades and shaft no housing around them. I have designed both the blades and fluid domians, I used the fluid domains to model the steady state flow, velocity and pressure in Advanced CFD but im not sure how to obtain rotational data.
1. What program is best for this situation (turbogrid, CFX or ICEM CFD)?
2. How do I set this up.
Any help would be greatly appriciated.
|May 17, 2010, 09:24||
Choose your weapons...
Retired from CFD Online
Join Date: Mar 2009
Location: Ann Arbor, MI
Blog Entries: 1Rep Power: 39
I am not a turbogrid expert, but I understand that it makes certain assumptions about your blade design. If you were doing something fairly standard, I would recomend it highly because it automatically blocks and smooths the model like a pro. However, it sounds like your design may not fit with these assumptions, so you may need to try a more flexible tool.
CFX is a solver, so I assume you mean CFX Mesh, which is now part of the ANSYS Meshing tool that comes with Fluent or CFX now. Yes, that should make it pretty easy for you, but will produce a Tetra/prism or perhaps swept Hexa mesh (if your geometry is sweepable). This will be good for intial numbers, but may not detect small improvements in the design. ANSYS Meshing is parametric and persistent, so changes to your solidworks model can be easily remeshed. It also works easily with ANSYS DesignXplorer, allowing you to easily run thru a number of design points and build good understanding of your design.
ICEM CFD takes a bit more time to learn (more options) but offers block structured hexa. ICEM CFD Hexa is what the big blade manufacturers (such as GE) use to generate the meshes for their blades. It is very flexible and scriptable. If you plan to do a lot of runs, it will really pay off in terms of rapid remesh, reduced solver time (due to the pure hexa mesh) and better accuracy, but it will take you longer to get going. You can hook it up with any optimizer that can run a script.
|Thread||Thread Starter||Forum||Replies||Last Post|
|multiphaseInterFoam for RAS turbulence model||chiven||OpenFOAM Bugs||7||August 23, 2011 02:52|
|2 stage axial turbine model convergence issues||sherifkadry||CFX||2||September 7, 2009 20:51|
|multi fluid mixture model issue||rystokes||CFX||3||August 9, 2009 19:13|
|algebraic slip model with stirred suspension?||Manuel Krause||CFX||0||April 15, 2005 09:10|
|Advanced Turbulence Modeling in Fluent, Realizable k-epsilon Model||Jonas Larsson||FLUENT||5||March 13, 2000 04:27|