Nusselt number is decreasing when using nanofluid using single phase phase approach
 

Hi,
i have used single phase approach for calculating the enhancement in HTC in a flow through a channel ,for the same first i have calculated the HTC (both local along the wall surface and Avg) for plane water without any nanofluid thereafter i have inputted the thermophysical properties of water+nanoparticle properties but the HTC is decreasing which is contrary to literature.
the BC are :
inlet v=
temp 293 k
outlet =outflow
wall temp=310 k
in reference values i have changed the length value to be the hydraulic dia
and for temp=the tbulk i have approached two method
1)made a centerline in the channel and calculated the weighted ag along the line
2)(outlet+inlet)/2

but in both these cases the HTC and NU is decreasing
Plz help in telling why there is an anomaly of the current analysis with literature.....

thanks in advance

You have to notice that, even in constant wall temperature flows, the bulk temperature will change along the tube, approaching the wall temperature at infinite length. If you set a specific temperature in the reference, the default Nusselt number and HTC calculated by Fluent will be, if you are really really lucky, only correct at one point, but it's going to be wrong everywhere else in your domain.

Calculating HTC and Nusselt numbers in duct flows in Fluent is rather tricky. I don't think you can do it in Fluent itself, so you'd have to use a post-processing software such as CFD-Post. In CFD-Post, you have to create a plane or a line (depending if you're solving 2D or 3D) perpendicular to your tube, calculate the bulk temperature along that line/plane and manually define a new variable that will compute the Nusselt number. If you are interested in plotting Nusselt or HTC along the tube, you will have to use what is called a session file, which makes use of Perl as a scripting language. The way I do it, I tell this session file to produce several lines and wall-points along the tube and compute the bulk average, then I probe the properties I want (conductivity, etc) at those wall-points to calculate Nusselt. The Nusselt is written in a data file and I read it with MATLAB.