General Mechanical

General Mechanical

1-way FSI for structure with multiple bends

    • i.tamunodienye2
      Subscriber

      Hi,

      I want to perform 1-way FSI on the elbows of a pipe structure with multiple bends. See figure below.

      I have completed CFD and i want to apply the calculated time signals of the flow induced forces to the inner surface of the elbow control volume of the X and Y components. And finally extract the resultant effect of the combined forces. I calculated the forces for X and Y in excel spreadsheet. I have inserted six forces in the analysis tree and pasted the force values in them to run the simulation. I have issues with selecting inside the elbows separately to apply the forces. I also had errors when i ran the simulation. Pls also advice on the best way to set-up my simulation especially time selections(end time, start time etc)

      please i need your help.

    • i.tamunodienye2
      Subscriber
      This is the error that's been coming up
    • i.tamunodienye2
      Subscriber

    • i.tamunodienye2
      Subscriber

    • Emperor
      Subscriber
      there was a time when I had to do a 1 way FSI.
      The principle was the following =: we were only interested in the fluid part, but the fluid mesh had to be deformed by the solid without being interested in the deformation of the solid (hence the interest of the 1 way fsi).
      This type of analysis is totally done with the fluent solver (hence the name 1 way FSI) when we couple the solvers (mechanical and fluent for example) we talk about 2 way FSI (that's how I perceived it).
      If I see the mechanical solver on your study, it seems to me that you are doing a 2-way FSI. Am I wrong?
      Put your study in the fluid category, I think you will have more answer.
    • i.tamunodienye2
      Subscriber
      Thanks for your response.
      I am actually doing 1-way FSI.
      Ansys has a tutorial for 1-way FSI using a workbench.
      In my case, I already ran a CFD study and calculated the forces in an excel spreadsheet.
      I copied and pasted the forces in their various components (x and y) but the issue I am having is applying the time-steps. These are where I am getting errors.
      Please do you know how to go about applying the current step number, No. of steps, step end time, initial time step, min time step, and max time step in transient structure

      Thanks
    • Emperor
      Subscriber
      I haven't done a transient analysis for a long time. But to find these parameters, if I remember my vibrational courses, you have to have the first natural frequency of the system (thanks to a modal analysis), then you have to know how long the effort involved lasts (actually).
      The end time must be a small multiple of the period of the first natural frequency. The period P = 1/f in seconds. For example, if the natural frequency is 20 Hz, the period is 0.05 seconds. If you want 5 periods, the end time would be 0.25 seconds.
      The initial time step should be a small fraction of the duration of the pressure pulse.
      If the duration of the pressure pulse was 0.025 seconds, you want at least 10 time steps to follow that pulse, so an initial time step of 0.0025 seconds.
    • Emperor
      Subscriber
      why do you want to make the transitional ?
      can surely bring an insight into your subject.

    • i.tamunodienye2
      Subscriber
      thanks for your help.
      I want to obtain the time-histories of the force signals of the fluid acting on the pipe bend which will be used to plot the PSD for further analysis.
      I am trying to use the document: Customer portal KM ID: 2040255 to demonstrate similar situation. But mine is a bit different. I have a copied and pasted the time series for 80sec. The time step set up is the main issue at the moment. You have explained I should perform modal analysis first. I will try to do that.
Viewing 8 reply threads
  • You must be logged in to reply to this topic.