Definition of the rotation axis

[mercierg]

Hello every body,

I'm currently working on water turbine farm. As a preliminary study, I try to simulate several (two, so far) rotating cylinders in a bigger domain with sat/sat coupling. You can have a look at the simple geometry in the attached file.

The coupling simulation between the 3 instance of Code_Saturne works fine. But I want to define both rotor as a rotating mesh. I think the use of usini1 user routine simply allow you to define the direction of rotation axis. In my case, I also need to set the center of the rotation.

Do someone see a solution using 'icorio=1', or maybe an way to simulate it?

Thanks for your time,

Guillaume

ps:I use v2.2 or v2.1.6

[Yvan Fournier]

Hello,

Looking at the code, it seems that this case is not planned for.

It would probably be a relatively simple development, adding a vector defining the axis to cstphy.f90, initializing it in iniini.f90 and usini1.f90, then adjusting the terms in which omegax, omegay, and omegaz appear in the code.

I'll try to see if we can add that in version 2.3 (this summer) or 3.0 (december), but in the meantime, I'm afraid you need to do it on your own (in versions 2.1 and 2.2, "grep omegax" in the src/base directory prints about 50 lines, in less than 10 files, so adapting the code could be less than 1 day's work if you do it carefully). I'll try to talk about this with developers who know this part of the code better than I do, to check there is no "hidden difficulty" in doing this, and provide suggestions if necessary.

Best regards,

Yvan

[mercierg]

Hello Yvan,

Thank you for your answer, you confirmed what I imagined. I'll try to make the modification myself and tell you what comes out. I also stay tuned for more input from your part.

Best regards,

Guillaume