Hello, I am trying to change si scale to cgs. I made the change in liggghts side and in OF part, but I am having some different behaviours. I run the settlingTestMPI using two different values for the kinematic viscosity in both cases and plotted v_y velociy as a function of time.
In liggghts side I changed unit, diameter, density, gravity and young modulus.
In OF side I changed gravity, kinetic viscosity (transport prop), rho field and the scale in blockMeshDict. I checked with the ErgunTestMPI_cgs and it looks like I did all the changes I needed but results are still not the same.
Could anyone please tell me if I should change something else in order to get equivalent systems? I am attaching a plot of that. Thank you very much in advance.
| Attachment | Size |
|---|---|
 comparision.png | 24.47 KB |
j-kerbl | Mon, 02/06/2017 - 10:56
Hi paulaalejandrayo,
Hi paulaalejandrayo,
the most important change on CFD side is to add the DimensionedConstants to the controlDict. Are these entries in there?
Have you changed the region extents on the LIGGGHTS side as well?
Cheers,
josef
paulaalejandrayo | Mon, 02/06/2017 - 12:29
Hello josef, thank you for
Hello josef, thank you for your answer. Yes, I did. I copied the DimensionedConstants part that was at the end of the controlDict fine in ErgunTestMPI_cgs exmple . But I also tried without that and I did not see changes. Is there any extra e way in which I should activate cgs units in CFD side?
This is the head of what I copied
"//libs ( "libgroovyBC.so" );
DimensionedConstants
{
unitSet CGS; // SI; // USCS; //
CGSCoeffs
{
universal (...)
" and I copied this until the end.
I changed the liggghts region too - I multiplied every variable with distance units by 100 in the DEM side and changed densitiy, gravity, etc. I was comparing files in both sides -si and cgs- and really do not know were else should I change. The boundary conditions are all set to zerograd or zero so I do not see any variable I could adjust there. Any idea? Thank you very much again.
NTT1508 | Mon, 02/06/2017 - 23:07
Start with Tutorial First
Hi mate,
I suppose what you should do first is to re-run again the tutorial for SI and CGS, try to change number of particles, the fluid velocity... and observe whether the tutorial give the same results using different unit systems. I have played with this game and at this level, I found a good agreement. See my previous post below:
http://cfdem.com/forums/cgs-unit-coefficients
Regards,
Nathan,
paulaalejandrayo | Wed, 02/08/2017 - 15:02
Hello Nathan,
Hello Nathan,
Thank you for your answer, I had seen your post before, this was the way in which I found the modifications needed for the controlDict in first place :)
My case is really simple, it is just a particle falling down with gravity and drag forces acting. But if you see the attachment in the post there are differences between the results I am having for si and cgs. They are in the same range of values, there is not a crazy difference, what makes me think that I am not doing completely wrong, but the limit speed I am having has different values for the different units and in the case with low viscosity the difference is really big although they have the same slope at the beginning. I think I am maybe missing to change some variable, but I am still not able to find what I am doing wrong. any idea? Thank you very much.
Best,
Paula