Questions about CGS units

Submitted by pipegaldames on Tue, 10/01/2013 - 22:32

I tried to work with diameter of particle about 20-100 microns, but I read here there are a limit because a precision machine issue and just I could work with 124 microns of diameter min in SI units. I changed my problem to CGS units, but I have some questions. In the example of tutorial ErgunTestMPI_cgs, the script Allrun.sh calls to controlDict_cgs_2.2.x and copy in other directory. It is necessary? What is it doing exactly ? .

In the other hand, I took exactly the same case and i converted all the units to cgs system, but the results are not the same as the SI. they are close but not equal. Does anybody knows why i'm watching differences?

Cheers,
Felipe.

Forums:

alice's picture

alice | Wed, 10/02/2013 - 14:22

Hi Felipe,
well the reason for copying the controlDict_cgs_2.2.x to the given directory makes you system use cgs units. You canjust open it in a text editor and have a look at the settings.
About the differences in the results: are you sure that you changed all the parameters (including viscosity etc)? There are quite some locations where changes have to made, one easily misses one... Otherwise, if you follow the routine in the Allrun-script, the results sould be identical (given that your particles are not so small that cgs is required anyways...)
Cheers,
Alice

pipegaldames | Thu, 10/03/2013 - 18:39

Thanks Alice, in past I ran same case without the commands that calls to controlDict_cgs_2.2.x and later with this and I don´t see many differences, for that i had questions. In relation to change of parameters, i checked all files both CFD and DEM, properties, boundary conditions, mesh, etc. There will be something else that I am forgetting?

Cheers,
Felipe.

alice's picture

alice | Mon, 10/07/2013 - 14:12

Hello Felipe,
well, as long as the particles in your test case are big enough for being treated with si units there should be no differences in the results. Did you run the test cases that come with the latest version? Here you find an si and a cgs version of the same case. You'll see that the results are in accordance to each other...
Cheers,
Alice