Learning Forum

[naada]

I’m working on a heat sink, I applied a heat source at the bottom of the heat sink and the model is turbulence but the issue I don’t get how I should use the built- in function to get the heat transfer coefficient do I use the surface average for the heat transfer coefficient if yes, should I consider the whole geometry even the bottom of the heat sink?. I have a conjugate heat transfer and I don’t know how the heat transfer coefficient is calculated in the software or how should I calculate it?. Please help me with this I would be very thankful.

[peteroznewman]

If your model is in Mechanical, using a Steady State Thermal analysis, then you apply a Convection BC to all the surfaces that are touching air, so all except the bottom surface which is receiving heat from a source below.  In this type of model, you need to provide the convective heat transfer film coefficient.

If your model is in Fluent, then you need to mesh the air as well as the solid and let natural convection carry away the heat and you don't need a convective heat transfer film coefficient, the solution computes the motion of the air that carries away the heat.

[naada]

Thank you, Yes i did the second way, but how can I get the report of the heat transfer coefficient?. I used a surface average and I chose all the surfaces except the bottom of the heat sink but I got totally different heat transfer coefficient than the expected one. Should I include the bottom surface ( the applied heat source surface) as well?. But even the surface temperature I got it somehow different I don’t know what is wrong in the simulation?.

[peteroznewman]

A surface average of what?  Please show details of exactly what you did to compute a heat transfer coefficient.  What is the value you got, what value did you expect and where did that number come from?  You said the model includes turbulence, which I assume means forced convection. Show a cross section of your mesh with the inflation layer at the air-wall boundary. What is the first layer element thickness? How do you know your solution is adequately converged?

[naada]

I used the surface average report to get the average heat transfer coefficient. No the model is a natural convection. I got the value of 2 and i’m Expecting 9. I’m not working in ansys i’m Working in star-CCM+ but they have the same concept as I think. Even for the temperate i’m expecting 30 the difference between the heat sink and the ambient and i’m Getting 49. The mesh is very fine but I don’t know where is the mistake?.

[peteroznewman]

Download the free ANSYS Student 2019 R2 and install it. Build the model in Fluent. Ask questions here about your Fluent model.