Why Fluent give me wrong heat transfer coefficient?
 

Hello I did simple heat transfer case represented by water flowing in a tube with uniform heat flux at tube wall I run the case in Fluent but it gives me incorrect values for heat transfer coefficient from reports>surface integrals>wall fluxes>surface heat transfer coefficient with unknown reason? I fixed all reference values but the heat transfer coefficient values still incorrect??

Mariam

No one know to answer my question?

I want to add if i have a fluid flowing over a surface with high temperature and i dont know the heat flux of the system rather i want to find out from the flow. Is it possible ? because most of the ANSYS tutorials on heat transfer assumes a heat flux then it just shows how the heat is being distributed. I know the temperature difference and my main interest is how the flow will occur and when its occuring how much will be the the heat flux.

how can you solve that equation? If you don't have a temperature and a flow rate you only got the heat coefficent. You cannot solve that equation manually.
The FAQ for boundaries for a stable/soluable simulation say give boundary conditions you can reproduce in a lab.

Steffen595 I have known heat flux on the pipe wall but fluent give me incorrect heat transfer coefficient when I compare with experiment may you tell me the reason why this happen with me?

mariam

sorry, cannot help you on that

I asked general question?? really no one know to answer it??

Don’t you have to define a HTC in the 1st place if you calculate heat transfer? So then there will be problems with what you defined and what is happening?
The heat flow is not evenly distributed I guess, so there may be problems? I.e. constant wall thickness, but different flow, how to average the HTC?
Did you check anything else, heat and mass coming in and out?
Is your simulation fully converged?

My geometry is so simple its just 2D tube the mesh quality is good the solution is converge and the prove for the solution is correct I get correct temperature distribution. I mentioned in my first post HTC predicted from surface integral (see my first post above) but its values seems incorrect when compared to experiments??

that sound like a recent post. In 2d and the velocity was way off. It had to do with, in 2d the depth is set as 1m. So if you calculate with surface integral, maybe assume the surface with 1m deep?
There are way less people doing CFD than CAD, so this forum is less busy. And some problems I guess are pretty rare.

My 10 minutes test shows no problem in calculating Nusselt/h values in 2d axisymmetric pipe.

Once you have finished your simulation if you want an avarage value of Nusselt or h just go to reports and perform area-weight surface integral of surface heat transfer coefficient on the surface of interest.
This will give you an average h, if you want Nusselt number just multiply this h by Diameter/(thermal conductivity).

In my test:
Nusselt calculated with:
0.023*Re^0.8*Pr^0.4 = 83.32 (Re=13000, smooth wall)

Nusselt from Fluent = 85.40

Daniele