Set frictional contact in Non-Newtonain flow

yzhao469yzhao469
in DualSPHysics v5.0

Hi, does anyone know how to set frictional coefficient for fluid-boundary interaction in SPH? Particularlly, I am using Non-Newtonian solver and "ViscoTreatment" = Constitutive eq.

I understand that if I use Newtonian solver, choose "ViscoTreatment" = "artificial viscosity" and set "ViscoBoundFactor" to a value larger than 1, then I am kind of having created a frictional boundary. However, I don't know how to set this friction in "Non-Newtonian & Constitutive eq".


Is there modification I can do in dynamic boundary condition? But I didn't find anything in DualSPHysics wiki about that. I also tried to couple it with CHRONO since CHRONO has Coulomb frictional contact defined in it. However, since there is no floating object inside my model, the error reads:

image.png

So I guess CHRONO is also not a correct way to solve the problem.


Thanks everybody who join the discussion in advance!

Comments

  • AlexAlex

    @gfourtakas ?

  • raphael_zuegerraphael_zueger

    I had to change the parameter "VelocityGradientType" to FDA. The difference can be seen in the image (same parameters, exept the velocity gradient type):

    image.png

    Here you can see, that the velocity at the boundary is a bit lower. Then I tried different changes (values from 0.1 to 0.9) in the "ViscoBoundFactor" to modify this "friction". The result was always the same. I had not tested it with values above 1 (I don't no why not). May you can test values above 1 and if this solves the problem it would be nice to leave a coment here on the forum.

    So I hope this will bring you a bit further.

    Regards

    Raphael

  • yzhao469yzhao469
    https://forums.dual.sphysics.org/discussion/comment/4241#Comment_4241

    Hi, thank you for your comment.

    I tried the following 4 cases:

    1. Viscotreatment = constitutive eq; velocitygradienttype = SPH, viscoboundfactor = 1;
    2. Viscotreatment = constitutive eq; velocitygradienttype = SPH, viscoboundfactor = 20;
    3. Viscotreatment = constitutive eq; velocitygradienttype = FDA, viscoboundfactor = 1;
    4. Viscotreatment = constitutive eq; velocitygradienttype = FDA, viscoboundfactor = 20;


    All the 4 cases give very similar result. So this doesn't work out.

  • AlexAlex

    Meanwhile @gfourtakas can not help here, I will invite you to check the info at: https://dual.sphysics.org/5thusersworkshop/user/pages/02.workshop/07.programme/Fourtakas_5DUW.pdf

    Regards

  • yzhao469yzhao469
    edited March 2021


    https://forums.dual.sphysics.org/discussion/comment/4244#Comment_4244

    Thank you @Alex for sharing the newest document. This looks like not include any discussion on the frictional contact. I'd appreciate it if the development team could consider including the "Coulomb frictional contact" into fluid -nonfloating rigid body interaction. If I am correct, currently the Coulomb frictional contact can only be applied through CHRONO to fluid-floating rigid body interaction. If the rigid body is not floating, the CHRONO coupling will be ignored.

    Is there any specific place to leave suggestions to the development team or may be here is a good enough place?

    Thank you for all the developers! Salute!

Sign In or Register to comment.

© Vanilla Keystone Theme 2024