Rocket nozzle solution for parameter study in Fluent fails without obvious reason

    • nerror

      I try to solve a problem for a university class where I have a parameter study of a rocket nozzle. The idea is to change choke- and nozzle outlet diameter and compare the change in velocity at the outlet. I created a 2D nozzle in SpaceClaim with 2 parameters to change the nozzle. The computation works fine and the results seem reasonable for the standard geometry. The solution is found with the Fluent solver and a 2D axisymmetric approach.

      Now, whenever I create a new design point and change the nozzle diameter, the update of the solution fails. The error message is not helpful either as it only says "Update Solution Failed". 
      When I set the newly created design point as "current", the geometry creation and meshing worked but the solution failed, saying "User Input required". Opening Fluent, the "Fluent launcher (Setting Edit only)" opens. However, when fluent is open, I can compute the solution straight away without making any other adjustments to the settings, only clicking on "ok" to initialize the problem. The solver runs fine and the result looks good. 

      I am really confused about what the problem here is and what I need to do to make the computation run automatically as I don't do anything other than to click on "compute". 

      (Also note that after I manually clicked on compute for the new design point, all other design points in the parameter overview in workbench have the yellow lightning indicating that there was a change in the setting but I really can't figure out what it might be.)


      I included the project files here.

    • Rob
      Ansys Employee

      Check the solver: is it set for 2d-axi when you open it? Also check the model is on the z=0 plane and that y>=0 (y=0 is hard coded as the axis).

      Staff are not permitted to open/download attachments so I'll need images.

    • nerror

      Hello rwoolhou,

      I didn't understand what exactly happend but after setting up the entire fluent solver again, the problem fixed itself and the parametric study runs without issue. 

      Thank you for your reply, though! 


Viewing 2 reply threads
  • You must be logged in to reply to this topic.