I am trying to do a simulation for a transformer radiator, and in this simulation I am using the conjugate heat transfer v2 simulation type. My goal is to find out how much the temperature drops in the lower outlet pipe after passing through the vertically positioned radiator fins, with a heat input temperature from the upper inlet pipe of 80 degrees Celsius. The problem is, during the simulation I am constrained by the results obtained because I used the default initial conditions set at 19 degrees Celsius. What should I do to get more accurate results and what type of simulation should I use? Since this is my first experience using CFD, I am doing it to fulfill my undergraduate graduation requirements.
I am trying to do a simulation for a transformer radiator, and in this simulation I am using the conjugate heat transfer v2 simulation type. My goal is to find out how much the temperature drops in the lower outlet pipe after passing through the vertically positioned radiator fins, with a heat input temperature from the upper inlet pipe of 80 degrees Celsius. The problem is, during the simulation I am constrained by the results obtained because I used the default initial conditions set at 19 degrees Celsius. What should I do to get more accurate results and what type of simulation should I use? Since this is my first experience using CFD, I am doing it to fulfill my undergraduate graduation requirements.
and this is my project
Translated with DeepL.com (free version)
Hi @heriyanto, thanks for posting on the forum and welcome to the SimScale Community 
Please share a project link so we can have a closer look at your simulation. Here’s a guide teaching you how: Guide to Clear and Understandable Forum Posting
From your explanation, a few comments:
- Yes, since you’re simulating a heat exchanger, the CHT v2 analysis type should be the way to go (here’s a tutorial that might help)
- Many times, simulation errors can be caused by a faulty/unprepared CAD models. Please take a look:
- Meshing is also a hot source of errors:
Cheers
Igor
I have sent my project to your email, thank you for your help.
Hi @heriyanto,
I assume you’re talking about this project.
Indeed, looking at the meshing log, it seems like your Non-orthogonality is a bit too high (see mesh quality resource I’ve shared to fix this problem):
A greater problem in my view, however, is the fact that you didn’t include a convective wall heat flux to the enclosure (see here for an example). If the walls are left unassigned, keep in mind they will behave as adiabatic walls, which doesn’t seem interesting since you seem to have these plates as parts of a heat exchanger of some sort.
Cheers