Gyroid representative volume element (REV) periodic boundary setup

- March 29, 2024 at 1:34 am
  
  sean.gao
  Subscriber
  
  Hi,
  
  I need to run Fluent for a heat exchanger with a gyroid core. There are three pieces, one is the solid gyroid wall, and two are the hot and cold fluid domains. Because I need to simplify the model and reduce the CFD running time, I need to use the representative volume element (REV) piece and apply the periodic boundary conditions. I used SpaceClaim to prepare the three pieces separately. The workflow I am using in the FLuent mesh is the Fault-tolerance workflow because the gyroid's STL file is inaccurate. However, I am not able to find how to define the periodic boundary condition. Anyone can help me with that even with a different workflow?
  
  Best regards,
  Sean
- April 2, 2024 at 4:26 pm
  
  Rob
  Forum Moderator
  
  Fluent will only allow one set of periodics and additionally, solid zones can't bridge the periodic zone. What is the objective of the model?
- April 2, 2024 at 6:17 pm
  
  sean.gao
  Subscriber
  
  Hi Rob,
  
  Thanks for replying! The objective of the model is to simulate the heat exchanger (HEX) with a gyroid core, the whole HEX is shown below.
  I heard Fluent only allows one set of periodic BC, but I do find a couple of papers stating that they are able to set up the periodic BCs for similar cases with solid and fluid domains, I also attached a screenshot of that.
  Could you please help to find out how to implement that?
- April 3, 2024 at 9:07 am
  
  Rob
  Forum Moderator
  
  I think the paper cheated and added fluid over the whole periodic surface: look at the a) image very closely. That avoids the solid restriction, and there shouldn't be any flow over a single layer of cells so the bulk result should be OK, but that's for you to determine.
- April 3, 2024 at 5:15 pm
  
  sean.gao
  Subscriber
  
  Thanks for replying, Rob.
  I want to do so, could you show me how to implement that? If needed, I can share with you the STL files.
  
  Best regards,
  Sean
- April 3, 2024 at 5:23 pm
  
  sean.gao
  Subscriber
  
  I am confused, because there are two fluids in my case, and the two fluids will be mixed if I add a single layer of fluid over the solid piece. Could you explain that in detail?
  
  Best regards,
  Sean
- April 4, 2024 at 9:56 am
  
  Rob
  Forum Moderator
  
  The example you showed has a single fluid, and they're bypassed the problem of solids bridging the periodic by making the periodic zone just fluid.
- April 4, 2024 at 5:50 pm
  
  sean.gao
  Subscriber
  
  Hi Rib,
  
  The paper attached simulated the same structure with the periodic boundary condition on the top and bottom surfaces so I believe there must be a way to set it up. Could you please help me to work on that by any chance?
- April 5, 2024 at 10:03 am
  
  Rob
  Forum Moderator
  
  It's possible to link the boundaries via UDF or similar, but looking at the image the periodic section is in the 85mm length direction. If that is the case what's flowing/conducting in that direction?
  I'm only permitted to offer general comments on here as it's a public forum.
- April 5, 2024 at 8:14 pm
  
  sean.gao
  Subscriber
  
  Thanks, Rob,
  
  I am still confused about what to do. If you are not able to offer more detailed information here, can discuss this via email? If possible, I can share my email.
  
  Best regards,
  Sean
- April 8, 2024 at 9:04 am
  
  Rob
  Forum Moderator
  
  If you're using licenced software the University should have someone (possibly you or your Professor) who can contact Ansys or the Channel Partner and the account manager can discuss options. As the Forum is public I'm very limited by the rules imposed to keep me on the right side of the US export laws: I'm always bending some of the rules, but never mess with Laws, or Legal....