I am planning to run a simulation to investigate the problem of natural convection over a heat sink and I have some doubts regarding which analysis type I should choose in order to do so. I have been reading that CHT would be a possible choice but I am not sure. Thus, I have two questions: 1- Is it…

Hi @gcleite96 , so although the inlet-outlet will not be driving the flow, they still need to be defined to tell the simulation where air can naturally come in and out, otherwise, as you have experienced, air will circulate within the domain, continuously increasing in temperature. I would define a p…

Thank you very much for the explanation and advice, @1318980 ! I will set up the simulation right now, applying pressure inlet and outlet boundary conditions, as you recommended. Hopefully, we will achieve good results. I will let you know when it finishes! Gabriel

Hi @1318980 ! Unfortunately, I have bad news. Apparently, the results for temperature did not converge again. I may be doing something wrong, but I really can’t see it. I created a copy of the project in order to do this new simulation. As can be seen in the second simulation (the one called “Natur…

Hi @gcleite96 , yes both using subdomain temperature is recommended regardless of having a good guess, this is so that the simulation starts with a temp difference between the solids. A longer run time might be required, first though, its best to check to make sure the setup is solid. At the moment I…

Hi @1318980 ! I think the gravity direction is right, although you probably spotted my mistake. See if it makes sense: The orientation of the geometry is this one: [Geometry%20Orientation] So, the gravity direction should, indeed, be pointing to -y. But, when I was setting the pressure inlet and…

Ok, I am struggling a bit, so the flow is flowing from Y- to Y+ boundary faces? if so I would suggest altering the shape of the domain, to be in the vertical orientation, what we want is to follow the rule of thumb that a bounding box (presuming it isn’t defined by some experimental setup) should be…

Well, when I set the forced convection simulation, the air was indeed flowing from +Z to -Z boundary faces. But, when I “shut down” this flow and let only natural convection takes place, the fluid, due to buoyancy effects, starts to flow from -Y to +Y, if I am thinking right. So, I was going to ask…

[image] gcleite96: So, I was going to ask you exactly what you’ve just said: it is better to invert the dimensions of the fluid region, right? Yes, this makes sense. Also in the interest of not skewing, alter the dimensions as above and ensure that the walls that are bounding are slip walls. …

Ok Darren! I will try the new setup and let you know how everything goes. Thank you very much for all the help, Gabriel

Hi @1318980 ! I finally achieved results close to the ones I was expecting. I think I just have to make some improvements now and everything will be good. This was the first part of the simulation. The last one, and even more challenging, is to do the same but with a heat sink immersed in oil rather…