Bondary conditions for heat transfer between two fluid regions

buffi · July 10, 2015, 05:49

I want to simulate heat transfer between two channel flows which are separated by two walls, like so:

Code:

----------------------- cyclic
        fluid A <-
======================= wall
        fluid B ->
======================= wall
        fluid A <-
----------------------- cyclic

I guess I need chtMultiRegionSimpleFoam for that, or would a different solver be better suited? The wall can be modeled as infinitely thin, i.e. without any thermal resistance.

The two channels exist as two separate meshes which I have created and tested individually. The meshes match geometrically, but the faces do not align perfectly. Do they have to? My problem is that I don't know how to set up such a coupled case. As far as I understand it, chtMultiRegionSimpleFoam accepts one mesh per region as in

Code:

constant/
|-regionA/
| |-polyMesh/
| | |- ...
| |- thermophysicalProperties
| |- ...
|-regionB/
| |-polyMesh/
| | |- ...
| |- thermophysicalProperties
| |- ...

but how am I supposed to couple them? The chtMultiRegionSimpleFoam/heatTransfer tutorial did not enlighten me (and I don't think it is close to what I want to do).

The chtMultiRegionFoam/multiRegionLiquidHeater tutorial seems to be closer, but I don't really understand how the coupling is done. Taking a look at that tutorial's BCs it seems that it is enough to set the patch type to

leftSolid/T:

Code:

    leftSolid_to_bottomWater
    {
        type            compressible::turbulentTemperatureCoupledBaffleMixed;
        value           uniform 300;
        neighbourFieldName T;
        kappa           solidThermo;
        kappaName       none;
    }

bottomWater/T:

Code:

    bottomWater_to_leftSolid
    {
        type            compressible::turbulentTemperatureCoupledBaffleMixed;
        value           uniform 300;
        neighbourFieldName T;
        kappa           fluidThermo;
        kappaName       none;
    }

So in my case I'd use

regionA/T:

Code:

    bottomWall
    {
        type            compressible::turbulentTemperatureCoupledBaffleMixed;
        value           uniform 300;
        neighbourFieldName T;
        kappa           fluidThermo;
        kappaName       none;
    }

regionB/T:

Code:

    topWall
    {
        type            compressible::turbulentTemperatureCoupledBaffleMixed;
        value           uniform 300;
        neighbourFieldName T;
        kappa           fluidThermo;
        kappaName       none;
    }

and that's it? Again, I don't need any kind of conduction modeling in the wall. If it is that simple I guess that those baffles simply confused me.

buffi · July 10, 2015, 08:21

Well chtMultiRegionSimpleFoam is now running, but first I had to couple the regions as described in this thread:
http://www.cfd-online.com/Forums/ope...-boundary.html

That took care of the coupling and also of patches not aligning perfectly. However, convergence is still causing some headache.

July 10, 2015, 05:49	Bondary conditions for heat transfer between two fluid regions	#1
buffi Member Join Date: Jan 2011 Posts: 45 Rep Power: 15	I want to simulate heat transfer between two channel flows which are separated by two walls, like so: Code: ----------------------- cyclic fluid A <- ======================= wall fluid B -> ======================= wall fluid A <- ----------------------- cyclic I guess I need chtMultiRegionSimpleFoam for that, or would a different solver be better suited? The wall can be modeled as infinitely thin, i.e. without any thermal resistance. The two channels exist as two separate meshes which I have created and tested individually. The meshes match geometrically, but the faces do not align perfectly. Do they have to? My problem is that I don't know how to set up such a coupled case. As far as I understand it, chtMultiRegionSimpleFoam accepts one mesh per region as in Code: constant/ \|-regionA/ \| \|-polyMesh/ \| \| \|- ... \| \|- thermophysicalProperties \| \|- ... \|-regionB/ \| \|-polyMesh/ \| \| \|- ... \| \|- thermophysicalProperties \| \|- ... but how am I supposed to couple them? The chtMultiRegionSimpleFoam/heatTransfer tutorial did not enlighten me (and I don't think it is close to what I want to do). The chtMultiRegionFoam/multiRegionLiquidHeater tutorial seems to be closer, but I don't really understand how the coupling is done. Taking a look at that tutorial's BCs it seems that it is enough to set the patch type to leftSolid/T: Code: leftSolid_to_bottomWater { type compressible::turbulentTemperatureCoupledBaffleMixed; value uniform 300; neighbourFieldName T; kappa solidThermo; kappaName none; } bottomWater/T: Code: bottomWater_to_leftSolid { type compressible::turbulentTemperatureCoupledBaffleMixed; value uniform 300; neighbourFieldName T; kappa fluidThermo; kappaName none; } So in my case I'd use regionA/T: Code: bottomWall { type compressible::turbulentTemperatureCoupledBaffleMixed; value uniform 300; neighbourFieldName T; kappa fluidThermo; kappaName none; } regionB/T: Code: topWall { type compressible::turbulentTemperatureCoupledBaffleMixed; value uniform 300; neighbourFieldName T; kappa fluidThermo; kappaName none; } and that's it? Again, I don't need any kind of conduction modeling in the wall. If it is that simple I guess that those baffles simply confused me.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Difficulty In Setting Boundary Conditions	Moinul Haque	CFX	4	November 25, 2014 17:30
Setting rotating frame of referece.	RPFigueiredo	CFX	3	October 28, 2014 04:59
ansys cfx solver exit with return code 1!!!!!	mhabibnia	CFX	7	August 19, 2013 03:53
Error finding variable "THERMX"	sunilpatil	CFX	8	April 26, 2013 07:00
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 09:11

July 10, 2015, 08:21		#2
buffi Member Join Date: Jan 2011 Posts: 45 Rep Power: 15	Well chtMultiRegionSimpleFoam is now running, but first I had to couple the regions as described in this thread: http://www.cfd-online.com/Forums/ope...-boundary.html That took care of the coupling and also of patches not aligning perfectly. However, convergence is still causing some headache.