Potential causes for increase in water height

Hi all,

I was wondering if any of you could give me a quick tip what could cause a increase of the water lever over time? Basically I am running a periodic case with a constant flow in a particular direction. However, with time the water height is increasing even the flow speed is maintained... Any ideas ??? Best H

Comments

  • You should take care with the use of deltaSPH. In some cases that negative effect can occur. Are you using deltaSPH?
  • Hi Alex,

    thank you so much for your answer. Yes I use deltaSPH and the 0.1 value. I am running now a couple of benchmark cases without deltaSPH to find out if I get rid of that increase... I let you know as soon as my pc has the results ...
  • Hi Alex,

    thanks again for your help.

    Removing the deltaSPH did not relly help actually I observed a increase of the height over time for both cases. I uploaded a small video of my case for better visualisation to my gdrive. The top case is the one with deltaSPH enabled.

    https://drive.google.com/open?id=1LBChQ5v2cvHuPYFJK0qsW_KUEaI9rO3X

    This is a 2D case and simulates almost 20s. If you compare earlier with later steps you seet that the height is increasing.

    Do you think that might be a 2D issue ??? I run a similar case in 3D and it seems better. However I still wait for the results ...

    If you have any idea I am more than happy for your answer...

  • That is strange since we have only experience the level increase with delta-SPH.
    Can you please send the XML to us to check the options you are using. Please email us to dualsphysics@gmail.com

    Alex
  • It can be also related with time-algorithm since Verlet is only suggested for preliminary tests, but not for accurate simulations...

    Our suggestions are:
    - wendland kernel
    - symplectic time algorithm
    - coefh:1.2
    - coefsound: 20
    - no deltaSPH (for the first try)

    Alex
  • Hi Alex thanks for your kind help I forwarded you my case !!! Thanks

  • Shifting should be only used for internal flows or when gaps that appear in the middle of the fluid can be a problem.
    However in the cases you are trying there is no need to apply shifting, which is creating here the effect of level increase since particles close to the surface try to move to "empty spaces" above the surface...

    Regards

    Alex
  • Dear Alex,

    just a quick follow up on that. I just analyzed the 3D cases and it seems a bit diffenent as in 2D ones. In 2D without shifting and deltasph enabled the water height is not increasing anymore. However, in 3D the water height is increasing without shifting and deltasph enabled. So as you indicated in your presious post the deltasph is in my 3D case causing a increase in the water height over time.

    It is cool to know that, however, when I disable the deltasph and the shifting, the case maintains the water level over time, but the velocity field becomes somewhat checkerd over time. Do you have any idea to fix this? I just try out a case in 3D with a lower cfl number to get a better timestepping. But I dont know if that will fix it.
    What do you think ?

    Thank you so much ...
Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!