40m QIL Cryo_Lab CTN SUS_Lab CAML OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  COMSOL elog, Page 2 of 3  Not logged in ELOG logo
ID Date Author Typeup Category Subject Text Attachments
  121   Tue Dec 5 10:50:54 2017 aaronMechanicsPonderSqueezeFEA on optimusI had some trouble running this on optimus.

Optimus has COMSOL 5.1 installed,
but I made these files in 5.3. I downloaded
  
  122   Tue Dec 5 19:50:47 2017 aaronMechanicsPonderSqueezeFEA on optimusGautam advised me against trying to install
version 5.3, lest it break version 5.1--I
had already gone through the install, but
  
  123   Tue Dec 12 11:50:12 2017 aaronMechanicsPonderSqueezeFEA on optimusSimulation results

First
run

Gautam ran the
COMSOL model on sandbox1 since we were trying
 SERaLIGO.jpg 
  125   Mon Jan 22 21:13:25 2018 aaronMechanicsPonderSqueezetips from ShoaibI talked with Shoaib about some changes
I could make to the FEA model to improve
convergence and reduce memory usage. Summary:
  
  126   Mon Jan 29 23:02:13 2018 aaronMechanicsPonderSqueezetips from ShoaibI started implemented some of these changes:


Started the mesh with a
boundary free quad mesh on the interface
  
  129   Sun Aug 26 16:42:54 2018 ranaMechanicsAnalysisTest Mass Thermal Noise: Consistency ChecksFor the Voyager test masses, we have been
considering a barrel coating to increase
the IR emissivity to increase the radiative
  
  131   Thu Feb 14 12:38:51 2019 Ching PinMechanics comsol modellingSo I did a simple comsol model of laser
heating of a silicon disk, with only radiation,
to see the temperature variation at steady
 Screenshot_from_2019-02-14_12-47-11.pngScreenshot_from_2019-02-14_15-07-40.pngScreenshot_from_2019-02-14_15-41-41.pnggraph.pdf 
  132   Fri Feb 15 21:05:31 2019 Ching PinMechanics comsol modelling 

So I got the time dependence to
work, but I'm not sure what went wrong
 Screenshot_from_2019-02-15_21-40-01.png 
  133   Tue Feb 19 19:52:53 2019 Ching PinMechanics comsol modellingThe time step response to heating via laser
(22.5 mW) is given in the attached picture,
for 2 starting temperatures, 122.5 K and
 Screenshot_from_2019-02-19_19-52-10.png 
  134   Fri Mar 1 19:33:40 2019 Ching PinMechanics comsol modellingI've changed the heating to be from
two heat sources, to better model the situation
with a heater and a laser. The heater deposits
 Screenshot_from_2019-03-01_19-35-32.pngScreenshot_from_2019-03-01_19-36-02.pngScreenshot_from_2019-03-01_19-46-25.png 
  135   Mon Mar 4 17:22:07 2019 Ching PinMechanics comsol modellingI've updated the material properties
to vary with temperature, mainly in the range
of 90-140 K. Using the parametric sweep function
 Screenshot_from_2019-03-04_19-13-28.pngScreenshot_from_2019-03-04_19-14-41.pngScreenshot_from_2019-03-04_19-15-37.pngScreenshot_from_2019-03-04_19-15-56.png 
  136   Wed Mar 6 09:51:18 2019 Ching PinMechanics comsol modellingSo I tried adding the sapphire lens to
the comsol model, and I am having teething
issues. I can't seem to get the solver
 Screenshot_from_2019-03-06_09-31-24.png 
  137   Thu Mar 7 10:10:37 2019 Ching PinMechanics comsol modellingThere are no issues with the thermal side
of the modeling, the issue seems to be with
the structural mechanics side. I'm not
 Screenshot_from_2019-03-07_10-10-47.png 
  3   Tue Apr 2 19:24:07 2013 Emory BrownOptics First steps producing a Finite Element Model to find the internal Brownian noise of a LIGO test mass 
Motivation:
Reduction
  
  5   Mon Jun 3 21:01:46 2013 Emory BrownOpticsGeneralAnalytic Calculation of Thermal Noise due to Brownian Motion using Levin and Thorne's MethodsMotivation:
Reduction of Brownian thermal noise
in future gravitational wave detectors is
 ComparisonPowerSpectralDensitySurfaceFluctuationsVsFrequency.pngDifferencePlot.pngThorneComparisonPowerSpectralDensitySurfaceFluctuationsVsFrequency.nb 
  6   Tue Jun 4 17:25:11 2013 Emory BrownOpticsGeneralComparison of First Results from COMSOL Model to the Analytic Solution Motivation: Before we can attempt
to modify the mirror designs to reduce the
thermal noise due to Brownian motion in them,
 ComsolComparisonPowerSpectralDensitySurfaceFluctuationsVsFrequency.png 
  7   Wed Jun 5 13:31:41 2013 Matt AOpticsGeneralAnalytic Calculation of Thermal Noise due to Brownian Motion using Levin and Thorne's MethodsNice Work Emory,
Can you make a plot with both of
the lines on the same plot, with a legend,
  
  8   Wed Jun 5 13:49:13 2013 Deep ChatterjeeOpticsGeneralDifference in the Levin and Liu & Thorne results for thermal noiseThe analytical expressions for thermal
noise, as calculated by Liu & Thorne
and Levin, was plotted as a function of frequency
 Levin_throne_comparison.pdfLevin_throne_comparison.png 
  9   Wed Jun 5 16:56:53 2013 Matt A.OpticsGeneralDifference in the Levin and Liu & Thorne results for thermal noiseGood Work Deep,
Can you include the equations that
you used to calculated these expressions?
  
  10   Wed Jun 5 16:59:51 2013 Matt A.OpticsGeneralMatlab code for calculating the zeros of Bessel functions from GWINC For calculating the Liu and Thorn
U_0 + DU, you need to sum over the first
N zeros of the first-order Bessel function.
 besselzero.m 
  11   Wed Jun 5 20:39:48 2013 Deep ChatterjeeOpticsGeneralConventional Thermal noise (Sec V) from Liu & Thorne>The linearized PSD plots are created
for the case of thermal noise in finite test
mass (Sec. V) of Liu and Thorne.
 Thorne_thermal_noise.pngPercentage_difference.pngThorne_thermal_noise.pdfPercentage_difference.pdf 
  13   Thu Jun 6 12:40:07 2013 Matt A.OpticsGeneralConventional Thermal noise (Sec V) from Liu & ThorneGood work Deep,
Can you write more on what this is
and why you're doing it? We want our elog
 Thorne_thermal_noise.pngPercentage_difference.png 
  14   Thu Jun 6 17:00:11 2013 Deep ChatterjeeOpticsGeneralComparison of ITM and FTM geometries (FIG. 3 Liu & Thorne)In section V of their paper on thermoelastic
and homogenous thermal noise, Liu and Thorne
have corrected the result of power spectral
 C_FTM.pngC_FTM.pdfFig3_Liu&Thorn.bmp 
  19   Wed Jun 12 13:52:44 2013 Deep ChatterjeeOpticsGeneralDifference in COMSOL and Analytic solution for TR noiseIn this post, I am reporting of the relative
difference between the analytic and COMSOL
results.
 analytic~COMSOL.png 
  21   Mon Jun 17 14:59:20 2013 Emory BrownOpticsGeneralImproved Meshing in a COMSOL ModelIn order to make computations more efficient
and possibly allow the set of boundary conditions
based on Liu and Thorne's suggestion of a
 COMSOLMeshing.png 
  22   Tue Jun 18 15:26:50 2013 Emory BrownOpticsGeneralAttempting to Implement Liu and Thorne's Boundary ConditionsLiu and Thorne demonstrated in their paper
that the optimal set of boundary conditions
for a problem like the one I am working on
  
  27   Thu Jun 20 16:29:38 2013 Deep ChatterjeeOpticsGeneralDifferences on using a coarser mesh in calculating TR noiseIn the code for evaluating TR noise, the
parameter for automatic meshing (which can
take values from 1 to 9 -  1 being the
 difference_in_mesh.png 
  29   Sun Jun 23 14:29:04 2013 KojiOpticsGeneralDifferences on using a coarser mesh in calculating TR noiseWhat about the dependence on the size of
the time slice?
It is unclear what the unit fo the
  
  30   Mon Jun 24 16:02:43 2013 Emory BrownOpticsGeneralRebuilding a minimal version of our COMSOL model I rebuilt our COMSOL model in order
to see if I could find any errors in it and
to clean out unnessesary features and parameters.
  
  31   Mon Jun 24 18:00:26 2013 Deep ChatterjeeOpticsGeneralThe problem of TE Noise calculation for a beam passing through a cylindrical substrateIn this post, I mention of my immediate
plan for the week and also describe the problem
I am looking at in brief.
  
  32   Tue Jun 25 15:16:59 2013 Arnaldo RodriguezOpticsGeneralSetting Up Looped Simulation for PID ControllerAfter setting up a COMSOL model that includes
the heat flux from the laser and the ring
heater, I've made the model solve over time
 Defocus.pngITMPlusRing.m 
  33   Tue Jun 25 15:18:41 2013 Emory BrownOpticsGeneralA new boundary condition to attempt to avoid the relative residual error I contacted COMSOL support about
our difficulties with the relative residual
error and was told "Typically for solid
  
  34   Wed Jun 26 13:52:06 2013 Arnaldo RodriguezOpticsGeneralVerifying Relative Error in Defocus for Regular and Manual-Loop SimulationsTo verify the validity of the solutions
produced by the manual simulation, I've calculated
the relative error between the results from
 Defocus_Error.png 
  35   Wed Jun 26 15:23:55 2013 Arnaldo RodriguezOpticsGeneralSolving Time per Loop in Manual Dynamic SimulationI've attempted to determine the solving
time per loop as a function of the simulation
time, in an attempt to identify any trends
 SolvingTimePerLoop.png 
  36   Wed Jun 26 16:46:32 2013 Emory BrownOpticsGeneralA convergent stationary solution By adding additional constraints
as described in the previous eLog to the
model I was able to get a convergent solution
 ConvergingModelMinimal-6-26-13.mph 
  37   Thu Jun 27 11:26:18 2013 Emory BrownOpticsGeneralMissing factor of 4 and a comparison of analytic and simulated Brownian noise    
  38   Thu Jun 27 15:06:13 2013 Arnaldo RodriguezOpticsGeneralPID Function in Manual Simulation I have inserted a rudimentary PID
function into the manual simulation code
as a way to test whether or not the PID function
 ITMPlusRingPlusPID.mPID.m 
  39   Thu Jun 27 17:12:33 2013 Arnaldo RodriguezOpticsGeneralDefocusing in P-Controlled Manual Simulation To test the effectiveness of the
PID loop, I've only given a nonzero value
to the K_p constant and varied it over a
 Defocus_(P_Active).png 
  40   Fri Jun 28 11:54:10 2013 Emory BrownOpticsGeneralMaintaining a constant volume while modifying the radii of the test mass faces When changing the dimensions of the
mirror, we want to keep a constant volume.
 To do so, we must modify the height of the
  
  42   Mon Jul 1 13:01:58 2013 Arnaldo RodriguezOpticsGeneralBrief Parameter Study on P-active Control Loop The following is a plot of the defocus
as a function of time with only the proportional
gain being nonzero. Only three cases were
 K_p.png 
  43   Mon Jul 1 17:31:40 2013 Arnaldo RodriguezOpticsGeneralParameter Study on Sampling Time for P-active Loop The following is a plot of the defocus
as a function of time for a fixed proportionality
gain of 150, and different listed sampling
 Deltat.png 
  44   Tue Jul 2 13:32:30 2013 Emory BrownOpticsGeneralUsing Matlab to run our COMSOL simulations We want to be able to test a number
of values for the ratio of the radii of the frustum
which will be much easier once we can get
 model_7_1_13.mlooper.m 
  45   Wed Jul 3 00:04:46 2013 KojiOpticsGeneralUsing Matlab to run our COMSOL simulationsPlease use zipped files instead of uploading
extremely long text files on the elog. 
Or use wiki or svn instead of elog
  
  46   Thu Jul 4 15:23:47 2013 Deep ChatterjeeOpticsGeneralAttempt to compute TE noise following Liu and ThorneLiu and Thorne, in their paper have analytically
solved for the TE noise spectra considering
the test mass to be a half infinite space.
 Liu&Thorne_type.bmpsomething_like_TE_1.bmpsomething_like_TE_1_10times.bmpsomething_like_TE_1_1000times.bmp2D_almost_infinte_TE.mph 
  47   Fri Jul 5 14:06:24 2013 Deep ChatterjeeOpticsGeneralAttempt to compute TE noise following Liu and Thorne


   
        
            
 comparison.bmpL&T_COMSOL_difference.bmp 
  48   Mon Jul 8 15:47:23 2013 Emory BrownOpticsGeneralMeshing error when varying geometry I attempted to run the Matlab code
built last week, but received an error that
the mesh failed to generate on the inner
  
  49   Mon Jul 8 18:39:31 2013 Matt A.OpticsGeneralMeshing error when varying geometry

    

       
            
  
  50   Mon Jul 8 19:04:35 2013 Matt A.OpticsGeneralMeshing error when varying geometry

    

       
            
 SimpleCylinder.mph 
  51   Tue Jul 9 14:45:04 2013 Emory BrownOpticsGeneralMeshing error when varying geometryI tried removing the constraints and applied
force and was able to get close to the same
frequencies for the butterfly modes as the
 Eigenfrequency.png 
  52   Wed Jul 10 14:15:20 2013 Emory BrownOpticsGeneralComparison of Umax and the frequency of the lowest eigenmode against ratioR I was able to run the looper program
designed last week over the interval ratioR=[0.5,
0.9] with a step size of 0.01 and obtained
 UmaxValuesVsRatioR.pngFrequencyLowestEigenmodes.pngdata.tar.gz 
ELOG V3.1.3-