Force gauge vs. Computeforces

Hello

We are a couple of guys trying use DualSPHysics for confined internal flow without free surfaces using the outlet/inlet boundary conditions.

So far we have done two benchmark studies on poisseuille flow and flow around cylinder, but we do not get the same forces as when done with Finite-Volume Method. And theres even a big difference if we use the Force gauge vs. Compute forces to measure the forces on the same mkbound.

Does anyone know why this seems to be the case? What is the difference between the two force measure methods?

Comments

  • Hello

    Consider showing with pictures you case and how you placed gauges etc. Makes it easier to discuss/help you out.

    If you are not permitted legally, to do this (i.e. confidential etc.), then consider making a simple case and show us that differences occur.

    Kind regards

  • Dear all


    The cases you want to compute are not the adequate for the v4.4 of DualSPHysics

    We are now working on new version 5.0 using new modified boundary conditions that will give more accurate results solving poisseuille flow and flow around cylinder.

    The code will be ready March 2020


    Regards

  • Sure. The following case is poiseuille flow, viscosity=0.0534, inlet velocity =1 m/s. SimulationTime=10s.

    Shifting is =3, Full and shift coef is -5. (The shifting coefficient also seems to have a big influence on the force measurements)

    Measurements are on the lower DBC: mkbound 0


    Force gauge, last second of simulation:

    Compute forces, last second of simulation:

    As you can see they differ even though they are measuring the same mkbound.

    Is there any inherent difference in the way the two methods measure forces?

    Please let me know, if you require other information about the simulation setup.

  • @Alex

    Thanks for the fast feedback.

    Very much appreciated.

    regards

  • @Rahca

    Do not open the files in Excel, since it is probably understanding the "comma" incorrectly. "E+15" is very wrong for our kind of simulations. Try comparing the results using for an example notepad++, to view the files.

    I did not know that a force gauge feature actually existed, so I cannot say more about that sorry.

    Kind regards

  • edited February 17

    @Asalih3d

    We are aware that "E+15" is not the real output but rather "E+03". We mainly use excel for plotting how the force evolves through time.

    But thanks for the heads up

    best regards

Sign In or Register to comment.