40m QIL Cryo_Lab CTN SUS_Lab CAML OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  COMSOL elog  Not logged in ELOG logo
Entry  Sat Jul 5 13:04:32 2014, Sam Moore, Optics, , Heinert Model TR Noise Verification threeD_cylinderTRnoise.pngoneDcylinderTRnoise_copy.png
    Reply  Mon Jul 7 19:47:00 2014, Koji, Optics, , Heinert Model TR Noise Verification 
       Reply  Fri Jul 11 10:58:18 2014, not Koji, Optics, , Heinert Model TR Noise Verification heinert_analyticalTest_residual_threeD.epsheinert_analyticTest_residual_oneD.eps
Message ID: 91     Entry time: Sat Jul 5 13:04:32 2014     Reply to this: 92
Author: Sam Moore 
Type: Optics 
Category:  
Subject: Heinert Model TR Noise Verification 

Agreement with Heinert's paper for cylindrical TR noise has now been achieved.  Using the stationary state assumption to calculate the temperature profile, the computation time was reduced compared to the previous time-dependent approach. Here are the plots showing the agreement.  I have shown the plots for a 1D axisymmetric model, in addition to a full 3D model in COMSOL.  Both give the same result.

 
What went wrong?  In the 1D axisymmetric case, it turns out that COMSOL has the incorrect cylindrical coordinate Laplacian for the coefficient form PDE interface.  I corrected for this by expanding the Laplacian with the product rule, giving an extra -kappa/r ''convection coefficient'' in the PDE interface.  Meanwhile, the 3D case worked from the beginning, since its Laplacian was in cartesian coordinates.
 
The next goal is to use a stationary state method to calculate TE noise for a test case.
Attachment 1: threeD_cylinderTRnoise.png  4 kB  | Show | Hide all | Show all
Attachment 2: oneDcylinderTRnoise_copy.png  4 kB  | Hide | Hide all | Show all
oneDcylinderTRnoise_copy.png
ELOG V3.1.3-