# Heat conservation in a channel

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

May 27, 2011, 13:31
Heat conservation in a channel
#1
Senior Member

jeff osborne
Join Date: Mar 2010
Posts: 108
Rep Power: 7
Hello all,

I've been having a reoccurring problem that I am desperate for some help with. I have flow through a channel with a cylindrical obstacle with constant heat flux coming through the top and bottom walls of the channel with periodic boundaries in both the spanwise and streamwise directions. I have attached a picture of my solution domain so you can get an idea of my problem. Looking at the picture, the top and bottom boundaries are walls set with a constant temperature gradient, and the front/back and inlet/outlet are all cyclical.

In order to make the streamwise boundary cyclical for the temperature field, what I do is subtract a constant number from the outlet to get the inlet. In my 0/T file, my boundary for inletOutlet looks like:
Code:
```inOut
{
type fan;
patchtype cyclic;
f List<scalar> 1(-6);
}```
where the number in the brackets (-6) is the value that I add to the outlet to get the inlet. The number 6 is calculated from looking at the total heat that comes into the channel between the inlet and outlet boundaries.

My issue is that I cannot get my temperature field to stay constant, it always increases over time. That is, the bulk temperature of the fluid is always increasing. I've tried numerous cases and here are some things I've found out:

1) Although it increases it time, if you look at the temperature difference between inlet and outlet the difference is the number you put in the boundary (ie. 6 in the example above)
2) It only increases in time when the flow is unsteady. When you reduce the Reynolds number enough such that the flow is steady, then it does not increase in time
3) The rate of increasing bulk temperature is constant, it always increases at the same rate
4) I have also tried fixing a point in the domain to force the temperature to stay constant. What ends up happening is the value in the cell that I "fixed" just increases with the rest of the solution. What's more is that for an otherwise steady case (ie. when the Reynolds number is low enough) when you fix a cell the bulk temperature still increases. In other words, for a steady case it doesn't increase when you don't fix a cell, but increases when you do fix a cell.

So, now I'm completely lost. I would really appreciate anyone that can give me some insight into this one, it would be a huge help.Thanks a million!

Jeff
Attached Images
 Solution Domain.png (30.0 KB, 2 views)

 May 27, 2011, 16:01 #2 Senior Member   jeff osborne Join Date: Mar 2010 Posts: 108 Rep Power: 7 New information to the puzzle: It doesn't matter what I change the value to in the fan BC, the results are identical. OF is not looking at this boundary in order to calculate the inlet temperature from the outlet.

 May 27, 2011, 19:35 #3 Senior Member   jeff osborne Join Date: Mar 2010 Posts: 108 Rep Power: 7 More information: When I try a really simple test channel flow case, I can get it to work. I can see a jump across the boundaries and the temperature field remains stationary. However, when I use my more complex geometry case, it still does not. There seems to be no relation between what I put in BC for inOut in my 0/T file and what is actually solved. The geometry for this case (with the cylinder) I made in Gambit and imported into OF following the steps outlined here (Problem with cyclic boundaries in Openfoam 1.5) Does anyone have any thoughts? I would really appreciate the help Jeff

 Thread Tools Display Modes Linear Mode

 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 OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Aloex FLUENT 1 February 22, 2011 13:28 svens OpenFOAM Pre-Processing 7 June 23, 2009 10:13 Mark CFX 6 November 15, 2004 16:55 src FLUENT 1 August 6, 2004 07:13

All times are GMT -4. The time now is 02:37.