CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > OpenFOAM > OpenFOAM Bugs

Probable improper implementation of optional thermal layers in mixed multi-region BC

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   February 26, 2015, 07:48
Default Probable improper implementation of optional thermal layers in mixed multi-region BC
  #1
New Member
 
Volker
Join Date: Aug 2014
Location: Germany
Posts: 15
Rep Power: 10
volker1 is on a distinguished road
Hi all,

The feature I am concerned about is the option to insert additional thermal layers at a boundary. It is implemented in the turbulentTemperatureCoupledBaffleMixed BC which may be applied for heat transfer between different regions e.g. when the solver chtMultiRegionFoam is used.
I have some doubts on the calculation of the overall (series) heat transfer coefficient (contactRes_) in version 2.3.1.


I cite the corresponding lines in the file:


turbulentTemperatureCoupledBaffleMixedFvPatchScala rField.C

Code:
  113  if (dict.found("thicknessLayers"))
  114     {
  115     dict.lookup("thicknessLayers") >> thicknessLayers_;
  116     dict.lookup("kappaLayers") >> kappaLayers_;
  117 
  118      if (thicknessLayers_.size() > 0)
  119       {
  120      forAll (thicknessLayers_, iLayer)
  121        {
  122      const scalar l = thicknessLayers_[iLayer];
  123     if (l > 0.0)
  124          {
  125      contactRes_ += kappaLayers_[iLayer]/l;
  126           }
  127        }
  128       }
  129    }
This calculation sums up thermal conductivities of the layers in the following way


contactRes_ = kappa1/Layer1 + kappa2/Layer2 + .


This would mean something like additional, parallel heat paths.
However from my understanding the layers were intended as series heat transfer and consequently the inverse values should be summed up:


1/contactRes_ = Layer1/kappa1 + Layer2/kappa2 +


Interestingly the second formulation may be found in the very similar BC file

turbulentTemperatureRadCoupledMixedFvPatchScalarFi eld.C
Code:
  119  if (dict.found("thicknessLayers"))
  120     {
  121      dict.lookup("thicknessLayers") >> thicknessLayers_;
  122      dict.lookup("kappaLayers") >> kappaLayers_;
  123 
  124       if (thicknessLayers_.size() > 0)
  125       {
  126  // total thermal transmittance by harmonic averaging
  127        forAll (thicknessLayers_, iLayer)
  128        {
  129        const scalar l = thicknessLayers_[iLayer];
  130        if (l > 0.0)
  131         {
  132         contactRes_ += l/kappaLayers_[iLayer]; // inverse sum
  133         }
  134        }
  135      contactRes_ = 1.0/contactRes_; // new total inverse
  136      }
  137     }
In case you make use of thermal layers in the turbulentTemperatureCoupledBaffleMixed BC, better have a look.
I certainly would prefer the second version for a layered problem.


Regards, Volker
volker1 is offline   Reply With Quote

Old   September 2, 2021, 06:06
Default
  #2
New Member
 
Fab
Join Date: Apr 2019
Posts: 24
Rep Power: 5
Rotidpor is on a distinguished road
I'm bumping this old thread just to share my experiments with kappaLayers and coupled Bcs in multiregion solvers.


I have strange issues that look a lot like what is described above.


So I set up a test case where I have an interface solid/fluid where the solid heats the fluid. I have found that if you put kappaLayers only on the fluif side, instead of reducing the heat flux passing through the interface, you increase it !


So my guess is that it is a parallel flux kind of thing implementation that adds a thermal path if you put kappalayers only on one side of the interface... Or the sign of the heat flux matters...


If you put kappalayers on both sides or on solid side (where the flux come from), the heat flux is effectively reduced.


So my question is, do you need to put h/2 on both sides to have a thermal insulation layer of effective height h or just put h on both sides ? I think you should put the same height h on both sides and it will result in an effective height of h.
Rotidpor is offline   Reply With Quote

Old   January 28, 2022, 15:45
Default
  #3
New Member
 
Balaji
Join Date: May 2013
Posts: 21
Rep Power: 11
Nealcaffrey is on a distinguished road
Quote:
Originally Posted by Rotidpor View Post
I'm bumping this old thread just to share my experiments with kappaLayers and coupled Bcs in multiregion solvers.


I have strange issues that look a lot like what is described above.


So I set up a test case where I have an interface solid/fluid where the solid heats the fluid. I have found that if you put kappaLayers only on the fluif side, instead of reducing the heat flux passing through the interface, you increase it !


So my guess is that it is a parallel flux kind of thing implementation that adds a thermal path if you put kappalayers only on one side of the interface... Or the sign of the heat flux matters...


If you put kappalayers on both sides or on solid side (where the flux come from), the heat flux is effectively reduced.


So my question is, do you need to put h/2 on both sides to have a thermal insulation layer of effective height h or just put h on both sides ? I think you should put the same height h on both sides and it will result in an effective height of h.
Hi,

Did you find the solution to this problem?

Does it adds up in parallel or in series, the efective thermal conductivites?

For example

// windshield 2 1 & 2 mm
thicknessLayers (0.002 0.001 0.002);
kappaLayers (0.8 0.2 0.8);
}

I have a windshield of three layers with 2, 1 and 2 mm respectively, i would like to use the effective thermal conductivities of 0.8, 0.2, 0.8 so that the heat transfer is reduced progressively.

Is the turbulentTemperatureRadCoupledMixedFvPatchScalarFi eld.C file replaced with proper calculation?

Any update would be helpful.

Thanks
Nealcaffrey is offline   Reply With Quote

Old   February 4, 2022, 05:11
Default
  #4
New Member
 
Fab
Join Date: Apr 2019
Posts: 24
Rep Power: 5
Rotidpor is on a distinguished road
Well, I must admit that I've been super lazy recently and my solution to this problem is to just create and mesh the layers.
It is as bad as it sounds but at least it works without any trouble
Rotidpor is offline   Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Unstabil Simulation with chtMultiRegionFoam mbay101 OpenFOAM Running, Solving & CFD 13 December 28, 2013 14:12
[snappyHexMesh] Multi region meshing & recovering the original patch names fluidpath OpenFOAM Meshing & Mesh Conversion 4 May 19, 2013 20:13
[Commercial meshers] Using starToFoam clo OpenFOAM Meshing & Mesh Conversion 33 September 26, 2012 05:04
[Other] StarToFoam error Kart OpenFOAM Meshing & Mesh Conversion 1 February 4, 2010 05:38
[Gmsh] Import gmsh msh to Foam adorean OpenFOAM Meshing & Mesh Conversion 24 April 27, 2005 09:19


All times are GMT -4. The time now is 19:12.