I have attached a figure where heat transfer fluid (HTF) is flowing through a rectangular duct. Three lateral surfaces of the duct are subjected to insulated boundary conditions, however, top surface is subjected to constant heat flux boundary condition.
I considered conjugate heat transfer module of COMSOL with turbulent flow and fine meshing.
I am calculating heat transfer coefficient between top surface and fluid with respect to change in mass flow rate of HTF. For this calculation, I am observing average temperature of top surface, volumetric average of fluid temperature and heat flux at top surface (Q).
Then, I am calculating heat transfer coefficient (h) by using formula:
h = Q/(Ttop surface - Tfluid)
Here, I have attached my results. But I am surprised after getting so high value of heat transfer coefficient. Are my results feasible? Can heat transfer coefficient be so high?
Can anyone suggest me, where am I committing mistake? Or what is the reason behind this high value of heat transfer coefficient? Or COMSOL does not have any control over heat transfer coefficient ?
Please suggest your valuable comments. I will be thankful to you.
-Manoj Sharma
IIT Kanpur, India
Dear Manoj Kumar Sharma
You have a system include two boundary condition, one heat flux and another one is insulation. You also have a fluid between them.
Point is, you have assumed the heat coming from boundary is just consumes for fluid temperature change. But if heat flux be very large, then fluid temperature will increase and finally you will have phase change. Then just by easy formula like h =q/dT, you can not calculate the heat transfer coefficient.
Let me explain more, in this formula h = Q/(Ttop surface - Tfluid) , Ttop is temperature at boundary (heat flux side) but how did you set Tfluid? this formula can be valid for a very thick fluid (or high volume) where the fluid temperature at far distance from boundary still remain without any change.
Dear Amir Reza Ansari Dezfoli , Thank you very much for your valuable answer. I have explanation for my question as you asked for that.
1. I used heat transfer fluid that has its boiling point at 350 deg C . Therefore, no need to consider latent heat transfer. (I forgot to mention it earlier)
2. I measure Tmean by considering the volumetric average of fluid domain's temperature. That should exist between the temperature of fluid at inlet and at exit. I have attached a figure that represents all the temperatures used to measure heat transfer coefficient (h). This figure demonstrates that Tmean exist between the temperature of fluid at inlet and at exit.
Please suggest me the possibility behind this high value of heat transfer coefficient (h). As in literature, I read that h can not exceed from 50,000 W/m2 .
Anything is possible in a simulation.
If you want to check whether or not your simulation is physically realistic, its better to start looking at your model inputs, rather then the outputs. For example: what is your Reynolds number? What pressure would be required to drive the fluid through this channel at the given velocity? Can you find experimental results that operate under comparable conditions (heat flux, mass flux)?
It seems to me like your mass flux is unrealistically high. This would also explain why the temperature of the fluid at the outlet is nearly identical to the inlet temperature.
- Badges
- Science topic