40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  COMSOL elog, Page 3 of 3  Not logged in ELOG logo
ID Date Authordown Type Category Subject
  58   Sun Jul 14 15:13:46 2013 Deep ChatterjeeGeneralGeneralTE noise problem - Matlab script

Quote:

Quote:

The present aim is the calculation of the TE noise like Liu and Thorne. Although, the results of Liu and Thorne are for the case of an infinite mirror, in our case, we try
to model a test mass with large dimensions in an attempt to get closer to the result due to Liu and Thorne

Following my previous post - The presents results given by the COMSOL simulations gave the same profile as Liu and Thorne's result which has a frequency dependence
of 1/omega^2 but was displaced from the analytic result by a constant factor which was dependent on the applied pressure on the face of the test mass which should not be
the case.

Following the steps followed Liu and Thorne, we had constructed the test mass as a cylinder, large compared to the beam spot size. An oscillating pressure was applied
to one of the faces and the from the temperature gradient generated in the process due to strains, one can calculate the work dissipated. The process involves integrating
the gradient of temperature over the entire geometry and taking the time avrage. However, this did not give the correct results, so it was decide to extract the Fourier coefficient
of the signal and perform the integration on it, as done in the case of TR niose by Koji Arai.

I mention the steps it was done -

> The data extracted was stored in a 3D array in matlab. using mphinterp - each dimension for r, z, t

> The number of radial slices and z slices is defined by the user previously

> For each r,z value there was a time signal.

> The fourier coefficient corresponding to the pressure oscillation frequency was extracted from the last three cycles
(In theory only one cycle should suffice)

> The above step was done so that random data( which is small in magnitude) generated by COMSOL is avoided.

> The fourier coefficients are stored in a 2D array corresponding to r and z values.

> The integration was performed using trapz twice once on each dimension to get the total volume integral

> The rest of the calculation is same as the previous script - plugging in the prefactors in the formulae and plotting

 

**The primary problem with this script is that the data extraction takes significant amount of time - For a mirror with 200 times the radius of the beam spot and 200 radial slices, it takes close to 6 minutes
to evaluate work dissipated for each frequency. The script solves for 16 frequency values. The results for small number of radial slices does not follow the straight line profile. For larger number of slices
the program takes a longer time. The results have not been checked yet.

 I am replying to report of the modification made to the script from the last one. The problem of too many radial and z slices have been avoided by using a slicing based on a gaussian
distribution both along the r and z directions - i.e. the slicing is heavy in the region around the surface is applied and much thinner towards the edges of the cylinder. The slicing has
been separately handled by a function named giveSlices() which does the slicing and returns the values as two arrays corresponding to r and z.
The extent to which the fineness of slicing is controlled in the geometry is by controlling the parameter SD in the code which is the standard deviation of the gaussian according to which
the slicing is controlled as being fine throughout or fine in the centre and thin with increasing r.

This method reduces the number of slices by almost 2 orders for significant large values of the dimensions of the test mass while it is expected to give similar results since slicing the
test mass finely at the edge is not required as the gradient of temperature almost falls to zero.

It is suspected that the long time of simulation is attributed to calling mphinterp() command a large number of times. In my next modification, it will be tried to use the command just once
followed by the proper data extraction from the output given by the command.

 

 Following the previous reply regarding the use of mphinterp(), the appropriate changes in code were made. The overhead time due to the call to COMSOL was responsible for the longer
simulation time. Right now, for relatively large no. of total slices (about 27000) just the extraction of data happens within a few seconds. However, for the larger number of slices, the codes 
will probably still take some time to complete. The codes are attached.

  60   Mon Jul 15 16:05:29 2013 Deep ChatterjeeGeneralGeneralTE noise problem - Matlab script

Quote:

Quote:

Quote:

The present aim is the calculation of the TE noise like Liu and Thorne. Although, the results of Liu and Thorne are for the case of an infinite mirror, in our case, we try
to model a test mass with large dimensions in an attempt to get closer to the result due to Liu and Thorne

Following my previous post - The presents results given by the COMSOL simulations gave the same profile as Liu and Thorne's result which has a frequency dependence
of 1/omega^2 but was displaced from the analytic result by a constant factor which was dependent on the applied pressure on the face of the test mass which should not be
the case.

Following the steps followed Liu and Thorne, we had constructed the test mass as a cylinder, large compared to the beam spot size. An oscillating pressure was applied
to one of the faces and the from the temperature gradient generated in the process due to strains, one can calculate the work dissipated. The process involves integrating
the gradient of temperature over the entire geometry and taking the time avrage. However, this did not give the correct results, so it was decide to extract the Fourier coefficient
of the signal and perform the integration on it, as done in the case of TR niose by Koji Arai.

I mention the steps it was done -

> The data extracted was stored in a 3D array in matlab. using mphinterp - each dimension for r, z, t

> The number of radial slices and z slices is defined by the user previously

> For each r,z value there was a time signal.

> The fourier coefficient corresponding to the pressure oscillation frequency was extracted from the last three cycles
(In theory only one cycle should suffice)

> The above step was done so that random data( which is small in magnitude) generated by COMSOL is avoided.

> The fourier coefficients are stored in a 2D array corresponding to r and z values.

> The integration was performed using trapz twice once on each dimension to get the total volume integral

> The rest of the calculation is same as the previous script - plugging in the prefactors in the formulae and plotting

 

**The primary problem with this script is that the data extraction takes significant amount of time - For a mirror with 200 times the radius of the beam spot and 200 radial slices, it takes close to 6 minutes
to evaluate work dissipated for each frequency. The script solves for 16 frequency values. The results for small number of radial slices does not follow the straight line profile. For larger number of slices
the program takes a longer time. The results have not been checked yet.

 I am replying to report of the modification made to the script from the last one. The problem of too many radial and z slices have been avoided by using a slicing based on a gaussian
distribution both along the r and z directions - i.e. the slicing is heavy in the region around the surface is applied and much thinner towards the edges of the cylinder. The slicing has
been separately handled by a function named giveSlices() which does the slicing and returns the values as two arrays corresponding to r and z.
The extent to which the fineness of slicing is controlled in the geometry is by controlling the parameter SD in the code which is the standard deviation of the gaussian according to which
the slicing is controlled as being fine throughout or fine in the centre and thin with increasing r.

This method reduces the number of slices by almost 2 orders for significant large values of the dimensions of the test mass while it is expected to give similar results since slicing the
test mass finely at the edge is not required as the gradient of temperature almost falls to zero.

It is suspected that the long time of simulation is attributed to calling mphinterp() command a large number of times. In my next modification, it will be tried to use the command just once
followed by the proper data extraction from the output given by the command.

 

 Following the previous reply regarding the use of mphinterp(), the appropriate changes in code were made. The overhead time due to the call to COMSOL was responsible for the longer
simulation time. Right now, for relatively large no. of total slices (about 27000) just the extraction of data happens within a few seconds. However, for the larger number of slices, the codes 
will probably still take some time to complete. The codes are attached.

 The dependence of the Power Spectral Density on the parameters F0, which is the pressure amplitude, has been somewhat corrected for in the codes attached. The mistake lay in the fact that
extraction of the Fourier coefficients was being done for the ht.gradTmag, the Fourier coefficients of which are not the correct ones to be used in this case since ht.gradTmag is not the correct time signal.
In the present case the Fourier coefficients are extracted from the r and z components, squared and added to carry out the calculation. However, the results are still off from the L&T results by order of 5.
The codes and the plots are attached.

Jul_15.bmp

The red values are the L&T result while the blue sketch is the simulation

  61   Tue Jul 23 18:26:04 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


  67   Tue Jul 23 20:53:37 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  66   Tue Jul 23 20:53:37 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  65   Tue Jul 23 20:53:37 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  64   Tue Jul 23 20:53:37 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  63   Tue Jul 23 20:53:37 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  62   Tue Jul 23 20:53:37 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  68   Tue Jul 23 20:53:45 2013 Deep ChatterjeeOpticsGeneralComparison between Liu and Thorne Results and COMSOL results for TE noise

Quote:

In this post I report of the results of TE noise simulated by COMSOL for the TE noise of Infinite test masses.

The aim was to follow the procedure by Liu and Thorne in their analytic calculations so that the same model could be used for the other
geometries.

The simulation is done in a different way than the TR simulations. It was observed that the output given by COMSOL by the use of commands
like mphinterp() or taking an export resulted in certain discrepancies between the results computed in COMSOL and that read by MATLAB.

Thus, the volume integration of the temperature gradient is performed in COMSOL itself and the results of the integration for each time
are sent to files. Matlab read these values and time averages them to get the result as in the paper (Sec. 2 of Liu and Thorne).

The errors expected are
> Fourier analysis is not done at all. This would have involved exporting data which, as mentioned before is giving errors

> The numerical errors by COMSOL are therefore not filtered off.

> The plot differs from the analytic solution for larger frequencies over 3000 Hz.

> It is to be noted from the paper by Liu and Thorne that the TE noise for the finite and infinite case are not very different. In
fact the correction factor goes as O(1). Thus, differences between finite and inifinte cases are unlikely to be prominent
in the log scale plots

 

 The codes are put as a zip file. Corrections made to the codes will be uploaded as a reply.

Jul_23.bmp


 Here is another plot with a mesh size slightly finer than the default Extra fine mesh in COMSOL.

One may notice that the value for the final frequency i.e. 10000Hz is different from the previous plot. 
It maybe that the error for the higher frequencies is a result of the FEA. However, it may also be that
the appropriate boundary conditions required for an infinite model break down at high frequencies.

  69   Wed Jul 24 21:08:24 2013 Deep ChatterjeeOpticsGeneralTR results for different dimensions

In this post I simulate the procedure of calculating the TR noise for finite cavities as proposed by Heinert and check for a
match.

The technique of performing all necessary calculations in COMSOL and exporting the results was applied to the TR codes.
It was seen that the codes gives similar output as the technique of extraction of Fourier coefficients in place of time averaging
as has been done in the codes of Koji Arai. One can see the output as the present code runs to be similar to the previous ones
found in the SVN.

However, the results in the present case were off by a constant factor close to 100. This maybe due to some 'm' - 'cm' or similar difference between
analytic calculations and COMSOL values of parameters. Although, it has not been found yet, the correction is hopeful to be
found soon.

The codes give results similar to the analytic result for other values of the mirror radius and beam radii (apart from the constant
factor I have mentioned above). One may have a look at the trend of the graphs between analytic and simulated values in the plots
attached. These plots are for the case when the mirror radius = 25m while the beam radius = 9 cm i.e. the original radii were 0.25m
and 9cm respectively i.e. the ratio has been changed by a order of 2.

As mentioned before the reason for the constant factor difference will be looked into.

  71   Thu Jul 25 13:21:46 2013 Deep ChatterjeeOpticsGeneralTR results for different dimensions

Quote:

In this post I simulate the procedure of calculating the TR noise for finite cavities as proposed by Heinert and check for a
match.

The technique of performing all necessary calculations in COMSOL and exporting the results was applied to the TR codes.
It was seen that the codes gives similar output as the technique of extraction of Fourier coefficients in place of time averaging
as has been done in the codes of Koji Arai. One can see the output as the present code runs to be similar to the previous ones
found in the SVN.

However, the results in the present case were off by a constant factor close to 100. This maybe due to some 'm' - 'cm' or similar difference between
analytic calculations and COMSOL values of parameters. Although, it has not been found yet, the correction is hopeful to be
found soon.

The codes give results similar to the analytic result for other values of the mirror radius and beam radii (apart from the constant
factor I have mentioned above). One may have a look at the trend of the graphs between analytic and simulated values in the plots
attached. These plots are for the case when the mirror radius = 25m while the beam radius = 9 cm i.e. the original radii were 0.25m
and 9cm respectively i.e. the ratio has been changed by a order of 2.

As mentioned before the reason for the constant factor difference will be looked into.

 

 

The discrepancy related to the difference between the analytic and COMSOL results has been partially addressed. Attached is another
plot showing the comparison. The ratio this time between the COMSOL results and the analytic results is between 0.7 - 0.8. This difference
will be looked into. It is, however, observed that the difference is not a constant factor - it has to do with the model file.

  72   Thu Jul 25 15:54:58 2013 Deep ChatterjeeOpticsGeneralTR results for different dimensions

Quote:

Quote:

In this post I simulate the procedure of calculating the TR noise for finite cavities as proposed by Heinert and check for a
match.

The technique of performing all necessary calculations in COMSOL and exporting the results was applied to the TR codes.
It was seen that the codes gives similar output as the technique of extraction of Fourier coefficients in place of time averaging
as has been done in the codes of Koji Arai. One can see the output as the present code runs to be similar to the previous ones
found in the SVN.

However, the results in the present case were off by a constant factor close to 100. This maybe due to some 'm' - 'cm' or similar difference between
analytic calculations and COMSOL values of parameters. Although, it has not been found yet, the correction is hopeful to be
found soon.

The codes give results similar to the analytic result for other values of the mirror radius and beam radii (apart from the constant
factor I have mentioned above). One may have a look at the trend of the graphs between analytic and simulated values in the plots
attached. These plots are for the case when the mirror radius = 25m while the beam radius = 9 cm i.e. the original radii were 0.25m
and 9cm respectively i.e. the ratio has been changed by a order of 2.

As mentioned before the reason for the constant factor difference will be looked into.

 

 

The discrepancy related to the difference between the analytic and COMSOL results has been partially addressed. Attached is another
plot showing the comparison. The ratio this time between the COMSOL results and the analytic results is between 0.7 - 0.8. This difference
will be looked into. It is, however, observed that the difference is not a constant factor - it has to do with the model file.

 

 

 

The issue related to the difference between the analytic and simulated values has been resolved. The codes seems to give reasonable match
between the analytic and simulated case. There is, however, a difference between the formulas being used from the previous cases. Note that
the 1/2 in front of Eq.(15) of Heinert is a because the time average has already been considered. However, in the present codes, the volume
integral of grad_T is evaluated in COMSOL and exported as a function of time. It is then averaged in MATLAB. Thus the factor of 1/2 is to be
omitted in this case(see Liu and Thorne Eq.(5). The presence of this extra factor of 1/2 was giving error in the last upoaded plots. From the
relative difference plot, one can see the maximum difference between COMSOL and analytic results go upto 7% but for most of the graph
it is close to 1% which is a fair result.

  73   Mon Jul 29 22:42:57 2013 Deep ChatterjeeOpticsGeneralAvoiding transient solutions in the Computation of TE/TR noise

An error was being encountered in the computation of the TR noise lately. It was observed that while running the simulations in the case of the materials which have a lower value of the thermal diffusivity (silicon / sapphire at room temperature), the simulated result were slightly off from the analytic result. On the other hand, if the simulation was run with a material of higher diffusivity(same materials at lower temperature), the match would be better. The reason being the transient solution not dying off significantly during the period of the simulation. Since a time average was being taken of the quantity integral{grad_T ^2},  the transient contributed to the integral. To get the correct value, the fourier coefficient of the time signal of integral{grad_T ^2},  was extracted at twice the frequency of the pressure oscillation. The reason being that the signal was squared. Extracting the response at this frequency after the integration is logical since the integration is over space while the response we extract is over time.

The same procedure was also applied to the TE noise calculation. However, this time we obtained similar result as the case where this procedure was not applied but a simple time averaging was performed. The tail of the plot, at high frequency, is still seen to deviate from the analytic result of Liu and Thorne as was the case previously. A plot is attached showing the spectrum for Fused silica at 290K. The conclusion being that the transients do not affect the TE noise calculation - the plot stayed the same even after filtering them out. This is probably because unlike the TR case which has a heat source present along the cylinder axis, the TE noise calculation involves applying pressure only on the face of the cylinder, and the transient do not contribute much to the volume integral.

 

  74   Wed Jul 31 15:39:11 2013 Deep ChatterjeeOpticsGeneralFirst try with paramter optimization for TE and TR noise profiles

After the simulations have been found to match to a fair extent with the analytic results by Heinert and Liu and Thorne, the attempt is check out the parameters for which the TE and TR noise are
close to each other. This was done with the analytic results. The frequency range 10 - 1000 Hz was looked into. In between this range, the quantity that was minimized was the absolite value
of the logarithm of the ratio between the TR and TE noise. The fminsearch function was used to minimize the mentioned quantity. The parameters that were changed were - conductivity, thermo
optic coefficient and coefficient of linear expansion. The reason for choosing these three were -

> TR noise is independent of coefficient of linear expansion

> TE noise is independent of thermo optic coefficient

> The power law dependence on conductivity is different for the TE and TR cases as can be seen from the analytic expressions

once the code returned the optimized parameters, these values were plugged in and the results were plotted.

**Note that the minimization was done for frequencies between 10 to 1000 Hz

  75   Thu Aug 8 17:17:19 2013 Deep ChatterjeeOpticsGeneralSomething like cancellation

For the material parameters of Sapphire at 300K, the TE and TR Noise profiles, though not very close, lie close to within an order of magnitude. Sapphire has a positive coefficient of linear expansion. We just inverted the sign of this quantity
and the ran the codes that puts heat and pressure simultaneously to the test mass. The total noise looks to be lower than the TR noise which is greater.

Mirror radius 0.25[m]

Mirror height 0.46[m]

Beam Radius 0.09[m]

 

If we have physical parameters which make the two Noise sources come closer to each other and then flip the sign of alpha, we may be able to see some noise reduction to a greater extent.

  80   Fri Jan 24 17:26:38 2014 Chris CoustéOpticsAnalysisMount Analysis Functional!

 The comsol eigenmode analysis is complete, and the only thing left to do on this part of the project is to run the analysis on a range of different configurations of optical mounts as well as a range of materials. This compilation will be posted on this elog in the next few weeks, due to the fairly long runtime of the analysis software.

  81   Sun Feb 16 21:10:56 2014 Chris CoustéOpticsAnalysisOptical Mount data compilation 1: Aluminum

 The time is finally here! this is the highest displacement of each mount in its lowest few eigenfrequencies, using 6061 Aluminum as a material. pictures will be added in a future log, because I'm going to make them into one file. Other materials will also be tested to see if there is variance in these findings, but only relevant data will be posted.

 

tl;dr findings: thick-open is the best combination because it has less displacement in eigenmodes and its eigenmodes are all very high/spread apart in frequency.

 

long:

 

(key) Eigenfrequency (Hz), highest displacement (mm), direction

 

Thin base, open mount:

315, 8.07, leaning back (laser will angle up)

316, 7.94, leaning to the side (laser won't change direction if the optic is flat, i.e. point of contact moves along surface of optic)

924, 10.50, twist to the side (laser will angle to the side)

1949, 8.89, twist about optic, slight lean forward (point of contact will move slightly, laser may angle down)

2144, 10.47, another leaning forward mode

 

thin base, open mount:

322, 13.29, leaning forward and translating to the side

328, 13.36, same as above

994, 17.66, twisting to the side (laser will angle to the side)

1943, 13.86, intense leaning forward

2100, 14.27, twist about optic

 

thick base, closed mount:

1329, 13.23, leaning back

1347, 13.32, leaning back and to the side

3577, 15.46, twisting to the side

 

thick base, open mount:

1326, 9.69, leaning to the side

1335, 9.77, leaning back

3504, 13.88, twisting to the side

 

------

All of these frequencies were checked for all four models, and the max displacements in those cases ranged from 100 picometers to 10 femtometers, so it's pretty reasonable to base decisions off of the displacements at the eigenmodes.

 

Based on this, it seems that for both thick and thin, open is the best, considering it displaces significantly less than closed (which is reasonable because there is less mass at the top to be thrown around). The choice between thick and thin depends on what frequencies we believe the mounts will be subjected to. If we are probably not going to have much sound at frequencies above 1000Hz, then thick is the better option (which is nice because it is the stock stalk instead of the custom made one). However, if there will be plenty of high noise, the thick is still probably the better option because it has fewer eigenmodes in the same range of frequencies.

 

here's a sample of two pictures of the relevant eigenmodes of the winner.

 

 

  77   Wed Nov 13 23:55:44 2013 Chris CousteOpticsAnalysisOptical Mount Vibrational Analysis

Project: Vibrational Analysis of Optical Mounts

Goal: Use COMSOL to run finite element analysis on simplified models of different types of optical mounts available to us, in order to find which shape/style/material reacts the least to external sound pollution. Once the few best candidates have been identified, develop test to experimentally determine optimal mount configuration and material.

 

Part 1: FEM

Process: Simultaneously build models in SolidWorks and design analysis in COMSOL

Solidworks: base/shaft models based on measurements of actual optical mounts, optic holder models matching mass and moment of inertia from downloaded models from their manufacturers. ~Progress: base/shaft done, optic holders almost done.

COMSOL: Design analysis that first does a general eigenfrequency analysis to find general vicinity of modes, then does a full modal frequency analysis. ~Progress: finished, ready for models to be imported.

  78   Mon Nov 18 15:56:59 2013 Chris CousteOpticsAnalysisRepresentative Models

The simpler models of the optical mounts are finished, they will be run through the comsol analysis software soon.

 

see pictures below:

 

  79   Fri Nov 22 21:04:53 2013 Chris CousteOpticsAnalysisanalysis

 The analysis is making nice eigenmode and stress mode models, but the displacement experiment needs work. Should be fixed by monday.

  131   Thu Feb 14 12:38:51 2019 Ching PinMechanics comsol modelling

So I did a simple comsol model of laser heating of a silicon disk, with only radiation, to see the temperature variation at steady state, which could be the limiting factor for high Q at 123 K, due to the thermalelastic effect. 

The model just uses a simple 2 inch disc, at 0.028 cm thick, with the flats not incorporated in yet. 

I had to search for silicon thermal conductivity and heat capacity at low temperatures, settling with k= 800 W/(m K) and C_p= 300 j/(kg K) from refering to papers. Will check LIGO documents for more accurate versions.

I put an arbitary boundary condition of constant temperature of 123 K on a spot .2 mm in radius, to simulate a beam.

Other arbitary values include 77 K for ambient and a surface emissivity of 0.5.

The laser is off center, because that it where the laser will enter the current setup.

We can see that the power required is .02 W, which seems reasonable.

The model is consistent with the analytic model I made with the laser beam at the centre of the disc. See last two figures.

 

I'm still trying to get the time dependence to work, as it is just giving me nonsense right now. 

 

Some thoughts: beam radius affects the temperature variation quite significantly, with a fat beam (1 mm radius) having half the temperature variation as a beam of .2 mm radius

I think the halo is just a trick of the eye, but I could be wrong. 

 

Things to do: 

Find the time scale of the system, as we want to modulate the laser to adjust the temperature, which will then be run though the mode ringer to measure Q to find the zero crossing

Change the heat source to be an actual laser

Add in the solid mechanics part

Add in the sapphire lens underneath

  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 in the first time anyways. I'll trying to get a sense of how long it takes for the temerature to semi-equlibrate, and coming to grips with comsol as a whole. There seems to be some inaccuracies when the timing increases, so I'm having to figure out how to increase accuracy without drastically increasing compute time. On the bright side, I switched the model to heating via deposited beam, for a more accurate model.

 

Attached is comsol's handling of a deposited beam modulated sinusoidally with a frequency of 0.1 Hz and screwing up badly.  Y axis is the average surface tempurature across the whole disc.

 

  133   Tue Feb 19 19:52:53 2019 Ching PinMechanics comsol modelling

The time step response to heating via laser (22.5 mW) is given in the attached picture, for 2 starting temperatures, 122.5 K and 122.8 K. We see that it takes fairly long to equlibrate, with a time constant of about 500 s, and is consistent across both temps. The y axis is average temperature across the surface of the disc, and the x axis is time. I believe that the heat distribution profile would be very similar with time, simply because of how much faster conduction is compared to radiation

  134   Fri Mar 1 19:33:40 2019 Ching PinMechanics comsol modelling

I've changed the heating to be from two heat sources, to better model the situation with a heater and a laser. The heater deposits 22 mW, with the laser deposting .5 mW. The overall temperature distribution is smaller then before, as expected, but doesn't really change much. The heater is simulated with a deposited beam with a gassian beam profile with a standard deviation (s.d.) of 8 mm. The laser to the size has a .3 mm s.d. for contrast. I learned that while the deposited beam power doesn't care for emisitivity, it cares about the area the beam is incident with, so for example, if you increase the s.d. too much, you get less power deposited then what you enter.

 

 I've also got modes to appear using solid mechanics, and I'm trying to see if there is a good way to get the frequency dependence with temperature nicely simulated. Changing the parameters of the simulation does give me my different frequencies, but I trying to find a way to do that over the time evolution of the model. I also got to check if the frequency shift is in line with what we are measuring. 

  135   Mon Mar 4 17:22:07 2019 Ching PinMechanics comsol modelling

I've updated the material properties to vary with temperature, mainly in the range of 90-140 K. Using the parametric sweep function to vary the input power of the heater, we get the eigenfreqencies' dependence on temperature to show up. The fractional dependence of 1.3e-5 /K around 123 K matches with what Aaron calculated in this elog entry, which is always a good sign that nothing is going horribly wrong. I've also added the flats to the silicon disc, for better accuracy. See the screenshots below showing the frequency shift with temperature.

  136   Wed Mar 6 09:51:18 2019 Ching PinMechanics comsol modelling

So I tried adding the sapphire lens to the comsol model, and I am having teething issues. I can't seem to get the solver to converge, but I'm working on it.

  137   Thu Mar 7 10:10:37 2019 Ching PinMechanics comsol modelling

There are no issues with the thermal side of the modeling, the issue seems to be with the structural mechanics side. I'm not sure what I'm doing wrong though, but it just isn't converging. In any case, seeing that this is my last day here, I'll just point out that the version without the lens is saved in cvs/cds/caltech/users/cp/current working model.mph, while the model with the lens is saved in the same folder under the file name testing with lens.mph, using optimus. There is also a small file edition of current working model, with a file name that is self evident. I'll leave it to aaron to upload that to git.

 

In any case, let me just put down some documentation and thoughts on this model: The physical parameters on the model are generally what we do know of silicon at these temperatures, with the exception of emissitivity, which was randomly given a parameter of .5. The model is currently absorbing 22 mW from the heater and .5 mW from the laser, which implies that the heater should be able to have 45 mW incident on the disc, which would in turn suggest that you would want it to at least dissapate 100 mW to account for the lack of direction from radiation. Because comsol's deposit beam power function does not care for emissitivity, it must be modified in tandem with it.

  32   Tue Jun 25 15:16:59 2013 Arnaldo RodriguezOpticsGeneralSetting Up Looped Simulation for PID Controller

After 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 manually by performing the solution process over a loop in MATLAB.

This allows for the future insertion of the PID controller object in the solving process, and the dynamic manipulation of the applied heat loads.

The following is an automatically generated plot of defocus effects as a function of time at 1 beam radius (54 mm), included in the program, with only the ring heater turned on in a top-hat emission configuration and the total power being 5 watts.

The linear projection values needed to calculate the defocus effect are extracted directly from COMSOL, with no output files required through the use of the mphinterp command.

The behavior appears to be physical and is of the correct order of magnitude.

Defocus.png

I've also attached the code that produced it for verification. (It requires COMSOL+Matlab to run.)

 

  34   Wed Jun 26 13:52:06 2013 Arnaldo RodriguezOpticsGeneralVerifying Relative Error in Defocus for Regular and Manual-Loop Simulations

To verify the validity of the solutions produced by the manual simulation, I've calculated the relative error between the results from the manual code and the results produced by COMSOL normally.

The plot for the relative error in the defocus at r = 0 and r = 54 mm is shown below, in the case where only the ring heater is turned on at a total power of 5 W.

Defocus_Error.png

The following indicates that the maximum error is less than 0.01% (in percent error format).

 

 

 

 

 

  35   Wed Jun 26 15:23:55 2013 Arnaldo RodriguezOpticsGeneralSolving Time per Loop in Manual Dynamic Simulation

I've attempted to determine the solving time per loop as a function of the simulation time, in an attempt to identify any trends in the solving time for a constantly dynamic load.

The following is a plot of the solving time per loop as a function of the simulation time for a load which is constantly dynamic (sinusoidal in time, in this particular run).

The mesh size is normal (default in COMSOL) with heat fluxes from both the beam and the ring heater (as in the real case).

 SolvingTimePerLoop.png

It is difficult to identify any particular behavior or trend due to the large amount of "noise" other than a trivial general increase after ~4 s. Mesh quality does not appear to influence solving time per loop significantly.

Work must be done to reduce the total solving time for the simulation, which in this case amounted to 18 and a half minutes (1.1109e+03 seconds).

 

  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 is changing the defocus values in the desired manner.

I am currently determining the ratio of ring heater power to the steady-state defocus as a way of measuring the scale of the response.

This ought to give a good way of measuring the scale needed to convert the calculated actuator response into an actual load.

I've attached the rudimentary code below. (The actuator isn't feeding into the heater at the moment, but inserting the "actuation" variable into the load expression is all that is required.)

 

  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 series of runs to verify the response by the controller.

The following is a plot of the baseline compared to the P-active simulations for different proportionality gain constants. I've yet to determine whether all the trends in behavior are physical.

Each different case has an evident droop (steady-state error), and seem to be destabilizing as K_p increases. However, most cases reach steady-state quicker than the baseline.

Defocus_(P_Active).png

 It should be noted that the ring heater is not correcting any effects from the beam, as it is turned off. The ring heater is correcting itself; or rather, its own initial heat pulse of 5 W * delta_t (600s) = 3000 J before the controller becomes active.

  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 run before the program crashed.

K_p.png

The dotted line indicates the setpoint.

One can clearly see the loop becomes unstable and oscillates out of control at a proportional gain of 240, and can see the characteristic steady-state offset of the P-controller from the set point develop in the other cases.

  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 times.

 Deltat.png

The discrepancy between the behaviors is enough to warrant a proper mathematical explanation; the max. overshoot, for example, is inversely proportional to the sampling rate.

 

 

  115   Thu Nov 2 17:23:56 2017 AaronMechanicsPonderSqueezeModelling suspension noise

aLIGO Suspensions Toy Model

On Wednesday I started making my own model of the aLIGO suspensions, with the top of the silica fibers attached to ears that are fixed rather attached to an additional suspension stage (so this will be a one stage suspension).

I grabbed the aLIGO ear design from the DCC: LIGO-D080751-v4

I am almost done with the model, should have it working tomorrow and will add it to the experimental gravity github in an appropriate place.

  116   Fri Nov 3 15:03:10 2017 AaronMechanicsPonderSqueezeModelling suspension noise

Model Geometry

Test Mass

I found the dimension of the test mass flat in the drawings of the mock test mass design here: LIGO-D080687.

Fibers

I modelled the fibers with the profile described in LIGO-D080751, fig 3.7.

Ears

I grabbed the d-values from LIGO-T1000545, but since the d-value is defined as the distance from the center of mass (of the penultimate mass (PUM) or the E/ITM) to the bend point (BP) of the fiber (I believe the point on the fiber with maximal flexure in the fundamental mode), I did not go through the effort of figuring out where the bend point is but rather grabbed the horn-CM distance from LIGO-T1100407

I wanted to get the real aLIGO parameters for the first version of this model, and have parametrized the model in such a way that I can define all of the parameters that need to change (surface area of the ear-TM bond, length of the fibers, thickness and profile of fibers, d-value, etc) and scale them with mass in some way for future iterations on this design.

I need to pare down the number of parameters, because I started by fully defining the ears and now am importing a 3D model of the ears and planning to scale these with mass.

Materials

For the material of the entire test mass and suspension, I used the fused silica that is specified as [solid,NIST SRM 739 - Type I]. I wasn't sure the difference between the types of silica, but this one said SRM so I thought it might have been defined on my distribution of COMSOL by a LIGO person. A quick google search showed me that person may have been rana?? https://labcit.ligo.caltech.edu/~rana/research/etm.html

Physics

I'm using a solid mechanics model.

Fixed Boundary Constraint

I fixed the position of the bonding surfaces for the PUM ears, so it is as if they are contacted to a completely fixed PUM (the PUM is not included in the model, but the upper ears are included, so the constraint is on the ears not the fiber. See drawing).

Gravity

I added gravity to all parts of the model. Apparently, it is not trivial to use gravity in a frequency domain study in COMSOL, as described in this presentation here. Fortunately, the presentation in the link is interested in the transfer function for a mass on a string also, so I follow the simulation steps they describe below.

Boundary Load

I add a boundary load that will vary sinusoidally for the frequency domain study.

Mesh

I have not yet messed with the meshing for these models. Obviously the points with more flexture and smaller parts (like at the horns of the ears, the tapering parts of the fibers, etc) will require a finer mesh.

Study

I need to incorporate the advice on how to build this study described in the link above. The following might also be useful, though I haven't looked through them yet:

https://www.comsol.com/model/dynamics-of-double-pendulum-14021

https://www.comsol.se/forum/thread/4843/pendulum-response?last=2010-04-27T01:48:26Z

  117   Wed Nov 15 14:05:12 2017 AaronMechanicsPonderSqueezeModelling suspension noise

Model Geometry

I pared down the number of parameters in the model to only the necessary ones. These are the ones that remain:
 
TM_radius: Radius of the test mass
TM_width: Width of the test mass
TM_flats: length of TM flats
ear_length: length of the ear
horn_spacing: length of the ear
horn_gap: gap between the top of the horn and the TM on the near side
d_val: distance from the CM to the bend point
horn_BP: distance from the horn to the bend point
ear_height_tot_nominal: nominal total height of the ear and horn for the unscaled (aLIGO) design (this name made more sense in a previous version of the model)
fiber_stock_length: length of the fiber's stock
fiber_neck_length: length of the fiber's neck
fiber_thick_length: length of the thick section of the fiber
fiber_main_length: length of the main section of fiber (the thinnest part)
fiber_taper_length: length of the tapering section of fiber
fiber_stock_radius: radius of the fiber stock
fiber_thick_radius: radius of the thick section of fiber
fiber_main_radius: radius of the main section of fiber
F_load: force of the boundary load used for the excitation in the frequency domain study
ear_scale_height: scale the height of the ear
ear_scale_length: scale the length of the ear
ear_scale_width: scale the width of the ear

Materials

For the material of the entire test mass and suspension, I used the fused silica that is specified as [solid,NIST SRM 739 - Type I]. I wasn't sure the difference between the types of silica, but this one said SRM so I thought it might have been defined on my distribution of COMSOL by a LIGO person. A quick google search showed me that person may have been rana?? https://labcit.ligo.caltech.edu/~rana/research/etm.html

Physics

Rana suggests that for the purpose of this study, it is not necessary to actually have COMSOL handle gravity as a restoring force... I'm not sure if I understand why this is yet? It seems that if we are interested in the relative strain energy in different parts of the wire compared to other parts of the system, it is important that the wire be under tension. If we have no gravity, the wire is effectively not under tension.

Mesh

I have not yet messed with the meshing for these models. Obviously the points with more flexture and smaller parts (like at the horns of the ears, the tapering parts of the fibers, etc) will require a finer mesh.

Study

I need to incorporate the advice on how to build this study described in the link above. The following might also be useful, though I haven't looked through them yet:

https://www.comsol.com/model/dynamics-of-double-pendulum-14021

https://www.comsol.se/forum/thread/4843/pendulum-response?last=2010-04-27T01:48:26Z

ELOG V3.1.3-