Topics relate to HFSS, Maxwell, SIwave, Icepak, Electronics Enterprise and more

Error while using Magnetostatics in workbench

    • Manas Seth

      while analyzing a solenoid, how to find the nodal solutions of magnetostatic scalar potential



    • GLUO
      Ansys Employee


      This section is for AEDT. Please also post on Mechanical section.



    • Ashish Khemka
      Forum Moderator


      Please see if the following link helps:


      How to access the ANSYS Online Help

      Thanks and Regards,

      Ashish Khemka


    • Bill Bulat
      Ansys Employee

      Hi Manas,

      By default, the finite element models created by the Magnetostatic analysis system use something called the edge-flux DOF to calculate results such as B field and magnetic forces. There are magnetic element types in the element library of the MAPDL solver that use the magnetic scalar potential (these are used in the example in the link that Ashish sent you). However, MSP technology is not natively exposed in Mechanical. Unless you have used a command object to highjack the model created by Mechanical and convert element types to those that use the MSP (e.g., SOLID96), there are no magnetic scalar results calculated by your Mechanical model. The elements used by default in Magnetostatic analysis systems use an entirely different technology to obtain results.

      In case you aren't already aware, the MSP is restricted to magnetostatic calculations - transient effects such as velocity-induced back EMF and induced eddy currents cannot be obtained with the MSP. In my experience, these phenomena have a modest effect on the performance of a solenoid during actuation. If your solenoid geometry is, like many, axisymmetric, and you want to account for transient effects that usually accompany actuation, it is usually easiest to setup the model using 2D coupled field elements in MAPDL (PLANE233). The electromagnetic options of this element type are not yet natively exposed in Mechanical, and setup with command object will probably be pretty tedius.



Viewing 3 reply threads
  • You must be logged in to reply to this topic.