Porous "Solid" Zone Interfacing with a Solid Zone - Boundary Condition - LTNE

RyMorRyMor Member

Hello, 

I am wondering about how to setup or create the proper boundary conditions for a conjugate heat transfer simulation that features steel rods embedded in a porous medium with air flowing through the porous medium. The simulation considers the LTNE porous media model. 

The porous medium (essentially sand) is at a high temperature, and the air influx through the sand is meant to convectively cool the sand. Steel rods have been added to the model in such a way as to possibly influence faster sand cooling times, and the overall purpose of this simulation is to gauge the effectiveness of this approach. 

While I can establish thermal interaction (i.e. heat transfer) between the fluid (air) and steel rod zones (through either interface or coupled wall boundary condition), when the LTNE model is active (creating the "sand" zone) I have not been able to find a way to couple the "sand" zone to the steel rod zone. 

I would expect that contact between the high temperature sand zone and the steel rods would have a considerable thermal impact on the steel rods in terms of heat transfer. As such, having these two zones coupled is imperative to my model. 

Any suggestions on how to couple the fluid (air) & steel rod (solid) zones as well as the sand ("porous medium") & steel rod (solid) zones would be appreciated. 

 

Thanks, 

Ryan

Comments

  • KremellaKremella Admin
    edited June 12

    You should be able to find the walls you would like to connect. The first step would be to fuse the walls to form an interior zone. Please use the TUI command

    /mesh/modify-zone/fuse-face-zone

    This will create an interior zone. Then you right-click on this interior zone and change this to the wall. This will create a coupled wall - wall-shadow. This way, you should be able to connect your steel and sand boundaries to get your CHT problem set-up correctly.

    I hope this helps.

    Karthik

  • RyMorRyMor Member
    edited June 12

    Hi Karthik, 

    Thanks for the reply. I have completed this approach to some respect, however, my issue is towards the possibility of having 3 zones (i.e. sand, air, steel) interface at one boundary. I guess one could refer to it conceptually as a "tripled" wall instead of coupled.

    From my experience thus far (specifically using the LTNE model), this does not seem possible. Would you be able to confirm this? In the event that it is possible, how can it be completed with Fluent?

    Otherwise, I guess it would be a fair assumption to only couple the sand and steel zones. I would just wonder about the convective effects of air on the rod (however, that effect might be insignificant compared to the interactions between the sand and steel). 

    Thanks!

    Ryan

  • KremellaKremella Admin
    edited June 12

    Ryan,

    This is a good point. I'd think that you will have three sets of coupled walls in your case: Air-sand, air-steel, and steel-sand. 

    This is how Fluent will be coupling the shared walls / surfaces between these cell zones in your model. 

    On top of this, since you are using the thermal non-equilibrium model, you will have a solid as well as a corresponding fluid zone. So, this will be adding an additional set of commonly shared surfaces.

    This is a tricky problem to set-up and you will need to be careful with how you are going about changing the conditions. I'd strongly recommend that you test the workflow on converting walls to coupled walls on a test problem with three different cell zone conditions (one of which is going to be the porous zone). This will at least point you in the right direction on how to go about this conversion.

    Regarding your second question, it really depends on the Peclet number in your problem. If you are really solving a conduction dominant heat transfer, then perhaps the coupling between steel and air might not matter. Air will simply be an insulator around the steel rods. However, if your Reynolds number is high, there is a good chance that you might have to resolve the convective effects. Again, I'd strongly recommend that you estimate the Peclet number in the problem to obtain clarity on this.

    I hope this helps.

    Thanks.

    Karthik

  • RyMorRyMor Member
    edited June 15

    Hi again Karthik, 

    Thanks for the detailed advice. 

    In regard to your first point, issues come about when trying to couple steel to air and steel to sand. An error is produced in the Mesh Interface interface that states that (if say steel is already interface with air) that steel has already been used in an interface and cannot be used again. Would you happen to know any work around to this?

    If not, I think that your initial point about simply coupling the sand and steel together is a valid assumption for my simulation scenario. With that said, if you are able to answer my question above, I feel that it would be helpful for anyone who also might be in my position but cannot assume that the interaction between steel and sand is negligible (or more generally, the interface of a "third" material with a "first" material which is already interfaced with the "second" material). 

    Thanks, 

    Ryan

  • KremellaKremella Admin
    edited June 16

    Hey Ryan,

    I wouldn't try and use the Mesh Interface interface for this. I'd try to use the TUI commands instead. Please try the commands from my earlier post and see if they help.

    Please let me know if this helps.

    Thanks.

    Karthik

  • RyMorRyMor Member
    edited June 17

    Hi Karthik, 

    I have tried to complete the commands you suggested in regards to using the TUI commands instead the Mesh Interface interface. Unfortunately, I run into the same issue. 

    The issue being that "steel" seems to be only able to be fused with air OR sand not both as this will produce an error. I have decide to only couple the sand and steel regions as this seems like a valid assumption for my application and also, the simulation results thus far seem reasonable. I will continue with this approach until it is apparent that it is actually not valid. 

    With that said, and as I mentioned in my earlier post, I do think for those looking to create a "tripled" wall (instead of coupled), whether or whether not ANSYS is capable of doing this remains unanswered. 

    Therefore, if you have any further suggestion regarding this topic, please feel free to share them for the benefit of anyone else who might end up in the same situation as me but unable to make the assumptions I have been able to make. 

    Otherwise, thank you for your help!

    Ryan

Sign In or Register to comment.