Getting (Nusselt vs Re) & (friction vs Re) in a ribbed duct
Hello
I have a ribbed duct as blow : http://imageshack.us/photo/my-images/109/51677694.jpg/ With the following boundary conditions: Duct height (H)=20 mm Rib height (e)=3 mm (square rib) p/e=13.3 Inlet length=245 mm Uniform heat at bottom surface=1100 W/m2 (the surface below a rib is considered insulated) Aspect ratio (AR)=5 Pitch p=40 mm Length of test section=280 mm Outlet length=115 mm Width of duct=100 mm Reynolds number=8900. How can I get the following curves in? http://imageshack.us/photo/my-images/209/17778613.jpg/ http://imageshack.us/photo/my-images/23/35025581.jpg/ *I use (Fluent 6.3.26)* Thank you in advance Shayan |
To get Nu vs Re, you need to run multiple simulations (multiple Re).
For each Re, calculate Nu and then save it into your favorite data manipulator (excel, etc) and make the plot from there. |
Quote:
In Fluent , I can get the "Surface Nusselt Number vs Position". You mean that I should write the reports of that for my specific Re number and calculate the average of those Nusselt number? In order to change the Re number, is it needed to change the "inlet velocity" only or any other terms should be considered? Lastly,the above curve is called "Ave(Nu)-Re" or "Nu-Re"? Regards |
You need to define Re. And then change the appropriate boundary condition to achieve the desired change in Re.
The curve is labeled Nu but it is undecipherable from the plot what Nu is. It could mean a lot of things. How Nu is defined will determine what you need to do in order to plot it. Fluent can calculate the overall average Nu (area weighted average) so you do not need to plot Nu vs x and then average yourself. Fluent can calculate those directly. |
Quote:
http://imageshack.us/photo/my-images/209/17778613.jpg/ Thanks |
It's up to you.
The overall average Nusselt number is one way to get one Nusselt number. But there are many many many non-trivial ways to define average Nusselt number or just Nusselt number. |
Hi,
It seems the best you can do is changing velocity to have different Re and calculate the Nu for each model. Even you can do it changing viscosity or density, whatever you want if you keep consistency with the Re. The only part I see a bit more difficult is how deal with this low Re number around 2000-10000. You will have to try different turbulence methods and find out which is able to handle this low Re region. Rgds. |
All times are GMT -4. The time now is 10:47. |