Fluids

Fluids

DPM unsteady particle tracking issues

    • sandeepjaggarapu
      Subscriber

      Hello Ansys Community,

      I am solving a DPM model for unsteady particle tracking of lubricating-oil with the unsteady continuous fluid as air.

      I am trying to simulate a two-phase flow(air and oil) in a circular pipe (Dia. 5mm and length 20mm) using DPM.

      The problem is that I want to export the particle history data but when I export, it says "DPM Iteration..." in the console and no data is being exported. The same "DPM iteration" message is being shown for tracking the velocity of the particle as shown in the following image. I cannot even get the sample trajectories data (Reports->Discrete Phase -> Sample) i.e. when I start the data to be saved it says stop and no end to it.


      Here is my setup:

      Time-> Transient

      Gravity-> Y: -9.81 m/s^2

      Viscous-> k-epsilon, Realizable, standard wall function.


      DPM - Interactions with continuous phase is turned on and updated for every two iteration intervals.

      Discrete phase on (rosin rammler particle distribution with surface injection)

      The injection start time=0 and

      stop time = 0.004 (this I have measured by the basic formula speed = distance/time)


      Boundary Conditions:

      Inlet velocity BC & outlet: pressure(0 Pa) DPM: escape

      Wall DPM: trap


      Method: coupled, Courant Number: 0.8

      Initialize: standard

      Run Calculation:

      Type-> Fixed

      Number of Time Steps: 1000

      Time Step Size: 4e-6

      Max Iterations: 20

      Reporting interval: 1

      Profile update interval: 1

    • Surya Deb
      Ansys Employee
      Hello, nSince this is unsteady particle tracking, the sampling is initiated by pressing the Start button. It creates the file and writes the header information. nBut you will need to progress the simulation and evolve the particle tracks in time by running it. This will then keep writing the particle information to the sampling file as and when a particle crosses the sampled plane. nYou can stop this sampling anytime by pressing Stop button.nI hope this helps.nRegards,nSDn
    • sandeepjaggarapu
      Subscriber
      Hello sdeb,nSir, are you suggesting to start the sampling data before running initialization and then start calculating? nOr else first run the calculation and then start the sampling data.nnThanks in advance.n
    • DrAmine
      Ansys Employee
      Yes.n

      Hello sdeb,Sir, are you suggesting to start the sampling data before running initialization and then start calculating? Or else first run the calculation and then start the sampling data.Thanks in advance.https://forum.ansys.com/discussion/comment/107335#Comment_107335

      n
    • diaelhag
      Subscriber
      Hello nif you couldnot save the files you can use UDF to save the output file with all information.nnregradsn
    • DrAmine
      Ansys Employee

      Hello if you couldnot save the files you can use UDF to save the output file with all information.regradshttps://forum.ansys.com/discussion/comment/107368#Comment_107368

      Not really helpful when one wants to sample particle tracks.n
    • sandeepjaggarapu
      Subscriber
      hello sir,nThank you, sir. I have got the sampling data of the particles.n

      Hello, Since this is unsteady particle tracking, the sampling is initiated by pressing the "Start" button. It creates the file and writes the header information. But you will need to progress the simulation and evolve the particle tracks in time by running it. This will then keep writing the particle information to the sampling file as and when a particle crosses the sampled plane. You can stop this sampling anytime by pressing "Stop" button.I hope this helps.Regards,SDhttps://forum.ansys.com/discussion/comment/107294#Comment_107294

      nn
    • sandeepjaggarapu
      Subscriber

      Hello Ansys Community,I am solving a DPM model for unsteady particle tracking of lubricating-oil with the unsteady continuous fluid as air. I am trying to simulate a two-phase flow(air and oil) in a circular pipe (Dia. 5mm and length 20mm) using DPM. The problem is that I want to export the particle history data but when I export, it says "DPM Iteration..." in the console and no data is being exported. The same "DPM iteration" message is being shown for tracking the velocity of the particle as shown in the following image. I cannot even get the sample trajectories data (Reports->Discrete Phase -> Sample) i.e. when I start the data to be saved it says stop and no end to it. Here is my setup:Time-> TransientGravity-> Y: -9.81 m/s^2Viscous-> k-epsilon, Realizable, standard wall function.DPM - Interactions with continuous phase is turned on and updated for every two iteration intervals.Discrete phase on (rosin rammler particle distribution with surface injection)The injection start time=0 and stop time = 0.004 (this I have measured by the basic formula speed = distance/time)Boundary Conditions:Inlet velocity BC & outlet: pressure(0 Pa) DPM: escapeWall DPM: trapMethod: coupled, Courant Number: 0.8Initialize: standardRun Calculation:Type-> FixedNumber of Time Steps: 1000Time Step Size: 4e-6Max Iterations: 20Reporting interval: 1Profile update interval: 1 https://us.v-cdn.net/6032193/uploads/26DMVZ82DW45/screenshot-28301-29.pnghttps://us.v-cdn.net/6032193/uploads/7VR0F5H6F2SZ/image.pngEdit @Rob correcting header typo to assist in future searches. https://forum.ansys.com/discussion/24654/dpm-unsteady-particle-tracking-issues

      Hello sir,nI have another issue regarding the same mentioned problem.nSir, I'm solving a two phase flow simulation flowing along a pipe. Pipe diameter of 5mm and length of 20mm. The velocity of the air and oil are considered the same as about 5m/s.nThe inputs were the same as I mentioned in the quoted text, sir. nThe residuals - 1e-6nThe problem doesn't converge sir and the scaled residuals graph were as follows. I have calculated the time step size, number of time steps and courant number. There are 7 lakh grid elements considered for the small pipe simulation but still, the convergence issue rises. Can anyone help me whether the obtained residual graph is okay or else still some changes are to be done for the convergence criterion? nn
Viewing 7 reply threads
  • You must be logged in to reply to this topic.