Topics related to Lumerical and more

difference between Pabs_total calculated by analysis group

    • ferruccio.pisanello


      I have confusion related to total absorbed power calculated by two different analysis groups.

      I used analysis group "cross-section" to calculate Pabs_total as done in the Mie scattering example. This analysis group calculates power flow in/out of a box.

      Also, I used the analysis group "Power absorbed (advanced)" to calculate Pabs_total as done in the Photothermal heating example. This analysis group uses the Joule heating formula.

      My question/confusion is, are they both not supposed to give the same Pabs_total? However, I am getting different values of Pasb_total. Both analysis groups have the same dimensions in my simple problem.

      for example for a 20nm dia gold sphere in water at 1mW/um2 power intensity at 530 nm,

      the total absorbed power (pabs_total) is:

      8e-11 W with the analysis group "Power absorbed (advanced)"

      2.5e-7 W with the analysis group "cross-section"

      as per literature, the value calculated by the analysis group "cross-section" is more reliable.

      Can anyone please comment on this?

    • Guilin Sun
      Ansys Employee
      Thank you for bringing this issue.
      You are right that the two methods do not give exactly the same result, even though in physics they should. In particular, when the absorbing object is small, the difference can be large.
      I prefer to believe the 1-R-T (plane monitors) is more reliable (the cross-section analysis group) , provided that the mesh is uniform for both groups. The reason is: planar power monitor has built-in feature to calculate accurately the Poynting vector and accurate integration to get transmission. In the absorption group, it has two steps to get the absorption with 3D power monitor: calculate the absorption spatial density, with multiplication of E-square and imaginary permittivity, which has more grid pints. and then integrate in space to get the final absorption ratio. The more post processing the more numerical error it can introduce.
      To further improve the accuracy using the planar monitor, make sure it has enough area to get better integration accuracy. This can be done with larger monitor size and finer mesh.
Viewing 1 reply thread
  • You must be logged in to reply to this topic.