General Mechanical

General Mechanical

Importing mechanical loadings for X-axis linear motor

    • Nikai Ryosuke

      I am currently doing a thermalmechanical simulation of X-axis of linear motor (consisting of movable parts such as the coil slider, cooling plates and the X-slider). The picture is shown below. Transient structural analysis is used to see the thermal deformations of the linear motor based on the imported temperature data from the transient thermal analysis. 

      All 5 parts are screwed together, so I defined them in Connections with translational joints instead of bonded type manual contacts.

      All parts should move synchrnously together with this given velocity profile in one duty cycle (see below).

      In the transient structural, I have at first defined the velocities and remote displacements (Y, Z and rotation axes constant zero, while X position/velocities with some calculated values) and also defined frictionless support between two/three in-contact surfaces of the adjacent bodies. However, at the end, the linear motor seems to be moving far and far away instead of showing the zig-zag motion (it should because the velocities/displacments have fluctuating values, see the table/graph for velocity profile below). Only one value is shown (no maximum and minimum) in the thermal deformation results and result fails to converge.

      So what I have done instead is putting zero constant for all translational/rotation axes in remote displacement, suppressing Velocity and then using acceleration with calculated values to define the mechanical inertial loadings (see figure below). I think by doing so is wrong because I am now assuming that the linear motor is static/fixed in one position and the accelerations become like a force acting on all the bodies of linear motor, resulting in static thermal deformation of a beam.


      So my question is what should I do to appropriately define the mechanical loadings, so that I can see plausible thermal deformations and see the linear motor moving in quasi zig-zag motion in the results section.

      Are connection settings correct, should I change/added specific types of connections?



    • Ashish Khemka
      Ansys Employee


      If all the parts are screwed together and cannot move relative to one another then bonded contact should be the choice instead of a translational joint. Try simplifying the model by applying one load at a time and see if you get the zig-zag motion with the application of displacement.


      Ashish Khemka

Viewing 1 reply thread
  • You must be logged in to reply to this topic.