[Crank-Shaft]

Hello fellow CFD Onliners,

I have been attempting to model an open flow domain including a backward facing ramp and a vane type vortex generator. The relative inlet boundary velocity is currently set to 4.5 m/s however, this will be increased later in the study. I also included a pressure outlet boundary with a zero gauge pressure and the top boundary is a free-shear wall. The sides of the fluid domain have been modelled as symmetric boundaries. The main objective is to analyse the downstream flow features, implement and test vortex detection algorithms.

There is another older thread with a description and images of the geometry of interest - http://www.cfd-online.com/Forums/ans...generator.html

The geometry was modelled in ANSYS DesignModeler and Fluent was used as the main solver. CFX Post was used for all the data collection and visualisation, often with scripts and session files.

The angle of attack of the vortex generator is a design variable and this needs to be altered over a range of 4 values - 5, 15, 30, 60 degrees. Since the mesh for the entire flow domain has been thoroughly studied and refined, it is preferable to avoid repeating the entire pre-processing and meshing procedure for each variable. Currently it is aligned stream-wise with zero angle of attack relative to flow domain and all the blocks and mesh biased sizing features were generated to suit this.

I would like to learn about some alternative techniques or approaches which could be used to model this. I have thought about varying the inlet flow velocity components, to match the desired angles however (Attachment), I am not sure whether this adequately represents an open flow over a surface with a vortex generator. Changing the inlet flow angle may also lead to other problems such as the free-stream 4.5m/s fluid completing missing the VG and exiting through the symmetry boundary before reaching the VG.

I would greatly appreciate it if someone could share their opinions on the above proposal and also add some suggestions or ideas to properly model the variable angles of attack. Remeshing the entire domain is extremely time-consuming especially due to the large number of design iterations involved with this.

Thank you and I look forward to a helpful discussion.

[Crank-Shaft]

Hey everyone,

Any ideas on this so far?

I have thought about the above proposal and the inlet velocity components in the X and Z directions respectively will not only change where the fluid stream exits, but also change the angle of incidence of the ramp and the VG together.

I really need some ideas on how to isolate the incidence angle changes to the vane itself. Surely someone out there attempted a similar simulation case and I would love to hear about any suggestions you have.