[Getdp] Electromagnetic 3D model - Crash info (bug?)

[Lucas]

Thanks for the answer, Lars! That solved the problem.

That's right, I'm starting to work with 3D photonic crystal structures and I
believe GetDP is
general enough to allow me to produce the band diagrams.
However I'm not sure how to formulate the problem.
Since you are also working around the same subject, could you point me in
the right direction?

Constraint {
   { Name BoundCond; Type Link;
     Case {
       { Region fxp; RegionRef fxm; Coefficient Complex[Cos[kx*ax],
-Sin[kx*ax]]; Function Vector[$X-ax,$Y,$Z]; }
       { Region fyp; RegionRef fym; Coefficient Complex[Cos[ky*ay],
-Sin[ky*ay]]; Function Vector[$X,$Y-ay,$Z]; }
       { Region fzp; RegionRef fzm; Coefficient Complex[Cos[kz*az],
-Sin[kz*az]]; Function Vector[$X,$Y,$Z-az]; }
     }
   }
}

FunctionSpace {
  { Name Curl_3D; Type Form1;
    BasisFunction {
      { Name W1; NameOfCoef e1; Function BF_Edge; Support Domain; Entity
EdgesOf[All]; }
    }
    Constraint {
      { NameOfCoef e1; EntityType EdgesOf; NameOfConstraint BoundCond; }
    }
  }
}

Formulation {
  { Name EMW_3D_Eigen; Type FemEquation;
    Quantity {
      { Name e; Type Local; NameOfSpace Curl_3D; }
    }
    Equation {
      Galerkin { [ -Dof{Curl e} , {Curl e} ];
                 In Domain; Jacobian JVol; Integration I2; }
      Galerkin { DtDt [ mu[] * eps[] * Dof{e} , {e} ];
                 In Domain; Jacobian JVol; Integration I2; }
    }
  }
}


Thank you for your help,
Lucas



On Wed, Jul 30, 2008 at 7:13 AM, Lars Rindorf <Lars.Rindorf at teknologisk.dk>
wrote:
>
> Hi Lucas
>
> You need to tell getdp the specific name of your domain. I.e. you should
set
> Group {
>   Domain = Region[4000];
> }
> instead of
> Group {
>   Domain = Region[All];
> }
>
> I'm also working on 3D electromagnetic bloch simulations, and I can help
you if have questions.
>
> Are you work on band structure calc of photonic crystals?
>
> Kind regards
> Lars
>
> ________________________________
> Fra: getdp-bounces at geuz.org [mailto:getdp-bounces at geuz.org] På vegne af
Lucas
> Sendt: 29. juli 2008 21:51
> Til: getdp at geuz.org
> Emne: [Getdp] Electromagnetic 3D model - Crash info (bug?)
>
> Hi all,
>
> I've just started working with GetPD and I'm not sure I'm doing the right
thing here.
> The attached files should solve an 3D eigenvalue EM problem (no boundary
conditions or
> sources yet) but I can't seem to make them work:
>
> getdp test -solve EMW_3D_Eigen -pos EMW_3D_Eigen
>
> Loading   : Problem definition 'em\test.pro'
> Loading   : Problem definition 'em\em.pro'
> Info      : Selected Resolution 'EMW_3D_Eigen'
> Loading   : Geometric data 'em\test.msh'
> Info      : System 'A' : Complex, Frequency = 0 Hz
> P r e - P r o c e s s i n g . . .
> Operation : Treatment Formulation 'EMW_3D_Eigen'
> Error     : Segmentation violation (invalid memory reference)
>             ------------------------------------------------------
>             You have discovered a bug in GetDP! You may report it
>             by e-mail (together with any helpful data permitting to
>             reproduce it) to <getdp at geuz.org>
>
> The geometry file is transfinite just cause I plan to use Bloch periodic
conditions later (any
> help in that matter would also be welcome).
>
> If I can manage to understand how to correctly create a model in GetDP, I
would say it's one
> of the best sw packages for FEM, but I couldn't do it following only the
documentation.
>
> Thanks for any help,
> Lucas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.geuz.org/pipermail/getdp/attachments/20080730/6f039a8f/attachment.html>