hexagon logo
Register Sign in
  • State Not Answered
  • Replies 2 replies
  • Subscribers 5 subscribers
  • Views 82 views
  • Users 0 members are here
Options
Related

Initial temperature dip down before rising up when heat flux is applied

stiwari
I'm using Marc to simulate the moving heat flux applied to the plate. The maximum temperature reached and cooling curve seems reasonable but there is an initial dip down in the temperature just before rising up. Initially, I was using first-order elements hex8 at the beginning. I tried using second-order elements hex20 but then the external heat flux is so weird. And there is a whole spectrum of flux moving up and down as waves. I also tried making the mesh finer, the time period for the dip in the temperature is less but there is still a dip.
Attached Files (3)
Parents
  • 0
    Basically this dip in temperature is caused by a point hat load to a grid point, it is less pronounce when you apply the heat load over an area (heat flux) analogous to pressure loading versus concentrated force at one grid point.
    I can duplicate this problem using MSC/NASTRAN thermal analysis or any FEM model as shown in my powerpoint.(heat_spike.pptx)
    Basically the spike has to do with your thermal conductivity of the material ,and the time steps as well.
    In slide 2 when I used epoxy material with thermal conductivity around 2 watt/m.K, you see the dip in temperature, a drop off of 50 degree C in the negative direction. Now if you increase the thermal conductivity by a factor of 100, like 200 Watt/m.K, (aluminum material), and slide 4 shows that there is no more dipping in the temperature which means that if you have refined the mesh then you will see no dipping using epoxy material as well.
    Slide 4 shows that for alum material the dipping is no longer there, and you can see color fringe plot on slide 5.
    Capture
    Capture
    Attached Files (1)
Reply
  • 0
    Basically this dip in temperature is caused by a point hat load to a grid point, it is less pronounce when you apply the heat load over an area (heat flux) analogous to pressure loading versus concentrated force at one grid point.
    I can duplicate this problem using MSC/NASTRAN thermal analysis or any FEM model as shown in my powerpoint.(heat_spike.pptx)
    Basically the spike has to do with your thermal conductivity of the material ,and the time steps as well.
    In slide 2 when I used epoxy material with thermal conductivity around 2 watt/m.K, you see the dip in temperature, a drop off of 50 degree C in the negative direction. Now if you increase the thermal conductivity by a factor of 100, like 200 Watt/m.K, (aluminum material), and slide 4 shows that there is no more dipping in the temperature which means that if you have refined the mesh then you will see no dipping using epoxy material as well.
    Slide 4 shows that for alum material the dipping is no longer there, and you can see color fringe plot on slide 5.
    Capture
    Capture
    Attached Files (1)
Children
No Data
© 2025 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement