40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log, Page 160 of 344  Not logged in ELOG logo
ID Date Authordown Type Category Subject
  6307   Thu Feb 23 02:20:07 2012 ZachUpdateSUSwacky state of SUS input matrices

This reminds me that the whole Dr. SUS situation never got taken care of. Where I left off, I was having issues pulling 40m data with NDS2 (which is what all the diagonalization scripts use).

What is the deal with 40m+NDS2? If it is till no-go, can we have a consensus on whether this is too important to wait for? If so, I will rewrite the scripts to use NDS and we can upgrade to NDS2 once we can prove we know how to use it.

 

Quote:

While Kiwamu and I were trying to investigate the the vertex glitches we were noticing excess noise in ITMX, which Kiwamu blamed on some sort of bad diagonalization.  Sure enough, the ITMX input matrix is in the default state [0], not a properly diagonalized state.  Looking through the rest of the suspensions, I found PRM also in the default state, not diagonalized.

We should do another round of suspension diagonalization.

Kiwamu (or whoever is here last tonight): please run the free-swing/kick script (/opt/rtcds/caltech/c1/scripts/SUS/freeswing) before you leave, and I'll check the matrices and update the suspensions tomorrow morning.

[0]

0.25 0.25 0.25 0.25 0
1.66 1.66 -1.66 1.66 0
1.66 -1.66 -1.66 1.66 0
0 0 0 0 1

 

  6399   Sat Mar 10 15:29:47 2012 ZachHowToComputer Scripts / ProgramsModeMatchr

For your mode matching pleasure, I have added a tool called "ModeMatchr" to the SVN under /trunk/zach/tools/modematchr/

It uses the usual fminsearch approach, but tolerates a fully astigmatic input (i.e., w0ix ≠ w0iy, z0ix ≠ z0iy) and allows for transforming to an elliptical waist  (i.e., w0fx ≠ w0fy, but z0fx = z0fy). It would be straightforward to allow for z0fx ≠ z0fy, but I have never seen a case when we actually wanted this. On the other hand, the elliptical output ability is nice for coupling to wide-angle ring cavities.

It also does the looping through available lenses for you , and retains the best solution for each lens combination in an output cell, which can then be combed with another function (getOtherSol). fminsearch is incredibly fast: with a 10-lens bank, it finds all 100 best solutions on my crappy MacBook in <10s.

I have also included the functionality to constrain the length of the total MMT to within some percentage of the optimal distance, which helps to sift through the muck .

MMrLogo.png

  6415   Wed Mar 14 13:27:15 2012 ZachHowToComputer Scripts / ProgramsModeMatchr

I have added to ModeMatchr the capability to fix the total MMT distance. This is nice if you are coupling to a cavity some fixed distance away. The blurb from the help:

% Note: for any total length constraint dtot_tol > 0, ModeMatchr will use
% fminsearch to find the best solutions near your nominal dtot, and then
% omit solutions whose dtot lie outside your tolerance. For dtot_tol = 0,
% ModeMatchr actively constrains dtot to your value, and then finds the
% best solution. Therefore, set dtot_tol = 0 if you have a fixed distance
% into which to put a MMT.

Quote:

For your mode matching pleasure, I have added a tool called "ModeMatchr" to the SVN under /trunk/zach/tools/modematchr/

It uses the usual fminsearch approach, but tolerates a fully astigmatic input (i.e., w0ix ≠ w0iy, z0ix ≠ z0iy) and allows for transforming to an elliptical waist  (i.e., w0fx ≠ w0fy, but z0fx = z0fy). It would be straightforward to allow for z0fx ≠ z0fy, but I have never seen a case when we actually wanted this. On the other hand, the elliptical output ability is nice for coupling to wide-angle ring cavities.

It also does the looping through available lenses for you , and retains the best solution for each lens combination in an output cell, which can then be combed with another function (getOtherSol). fminsearch is incredibly fast: with a 10-lens bank, it finds all 100 best solutions on my crappy MacBook in <10s.

I have also included the functionality to constrain the length of the total MMT to within some percentage of the optimal distance, which helps to sift through the muck .

MMrLogo.png

 

  6443   Mon Mar 26 12:50:24 2012 ZachUpdateelogrestarted with script

On the plus side, it was the first time I've had to do it in a while..

  6478   Tue Apr 3 01:52:15 2012 ZachUpdateLSCRAM simulation for Full ifo

Quote:

 I also would like to extend this script to use the DC readout, but don't know how to calculate the postion offset for AS_DC because the error signal is not zero-crossing for AS_DC anymore. Do you have any suggestions for me?

 I don't think I understand the question. AS_DC should not have a zero crossing, correct?

  6550   Thu Apr 19 16:21:04 2012 ZachUpdateComputer Scripts / ProgramsArbcav updated, made badass

I have modified Arbcav to be way cooler than it used to be.

Main modifications:

  • Can now truly model an arbitrary cavity geometry
    • The previous version could only handle a few different topologies. In each case, it would unfold the cavity into the equivalent linear cavity and use the g-parameter method to calculate gouy phases, etc.
    • The new model uses the closed cavity propagation matrix to find the supported mode, and then explicitly calculates the accumulated gouy phase by propagating the beam through the full cavity. This is done analytically with zR, so there is negligible slow-down.
  • Now plots a diagram of the cavity geometry, both to help you and for you to verify that it is calculating the right thing (<-- this is the cool part)
    • Plots the beam path and mirror locations
    • Specifies whether mirrors are curved or flat
    • Prints mirror parameters next to them
    • Finds all intracavity waist locations and plots them
    • Gives waist information (size in X, Y)

Since the information is already there, I will have the output structure include things like the input beam q parameter, which could then be fed directly to mode matching tools like ModeMatchr.

The function takes as input the same arguments as before. Example for a square cavity:

out = arbcav([200e-6 50e-6 200e-6 50e-6],[0.75 0.75 0.75 0.75],[1e10 9 1e10 9],[45 45 45 45],29.189e6,10e-6,1064e-9,1000);

i.e.,

out = arbcav(transmissivity_list, length_list, RoC_list, angle_list, modulation_freq, loss_list_or_loss_per_mirror, wavelength, num_pts_for_plot);

If you don't give it a modulation frequency, it will just plot carrier HOMs. If you don't give it RoCs and angles, it will just plot the transmission spectrum.

 

I'm still fine-tuning some functionality, but I should have it up on the SVN relatively soon. Comments or suggestions are welcome!

 

Some screenshots:

Cavity geometry plots (linear, triangular, square, bowtie):

linear.pngtriangular.pnggyro.pngbowtie.png

 

Transmission and HOM spectra (these correspond to the square cavity at lower left, above):

gyro_spect.pnggyro_HOM.png

  6823   Sat Jun 16 12:03:41 2012 ZachUpdateGreen Lockingscanned Y arm for 5FSR

Is that time stamp really correct? I wanted to look at the signal closely to see if I could get any feeling for why it would look so different when positive vs. negative, but I do not see a triangle anywhere near this time (1023780144)...

Quote:

Interesting. It seems for me that there is a dependence of the noisiness as the beat frequency is scanned.

As you increase (or decrease?) the offset, C1:ALS-BEATY-COARSE_I_IN1 becomes bigger and more crisp.
The resulting out-of-loop stability also seems to be improved as you can see from the crispness of the PDH signal.

Do you find why this happens? Is this because the beat S/N depends on the beat frequency due to the PD noise?

 

 

  6912   Wed Jul 4 18:25:44 2012 ZachUpdateComputersNDS2 client now working on Ubuntu machines

After plenty of work, NDS2 can now be used to get site data within MATLAB using the following machines:

  • allegra
  • megatron
  • ottavia
  • pianosa
  • rosalba
  • rossa

What I did

NDS2 was not working on any of the machines, so the first thing I did was simply to install the newest version. I downloaded the latest tarball (0.9.1) from the LDAS Wiki, unzipped and installed it

/users/zach $ tar -xvf nds2-client-0.9.1.tar

/users/zach $ cd nds2-client-0.9.1

/users/zach $ sudo ./configure --prefix=/cvs/cds/caltech/apps/linux64 --with-matlab=/cvs/cds/caltech/apps/linux64/matlab/bin/matlab

/users/zach $ sudo make

/users/zach $ sudo make install

 

Even with the new version, it still didn't work.

Solution: The main problem was that the cyrus-sasl-gssapi authentication protocol was not installed on these machines, so that even with a kerberos ticket the datalink could not be established. Using information from the LDAS Wiki, I used aptitude to install it as:

$ sudo aptitude install lscsoft-auth

This group installs both the SASL protocol and the package python-kerberos

 

I also needed to update the kerberos config file for each machine, which is located at /etc/krb5.conf. I found that ottavia had a nice one with many realms, so I copied that one over to the other machines. In any case where there was an old config file overwritten, it is now /etc/krb5.conf.old.

Finally, the matlab path for NDS2 was still set to the old 2010a directory (/cvs/cds/caltech/apps/linux64/lib/matlab2010a) that was created by the NDS2 install when Rana originally did it. The new install I made above created the appropriate 2010b mexa64 files, so I changed the matlab path within matlab to this one:

>> rmpath /cvs/cds/caltech/apps/linux64/lib/matlab2010a

>> addpath /cvs/cds/caltech/apps/linux64/lib/matlab2010b

>> savepath

 

Now everything works fine on all these machines. As in Rana's original post, you get data in the following way:

$ kinit albert.einstein %then enter password

$ matlab -nosplash -nodesktop

>> d = NDS2_GetData({'H1:LSC-NPTRX_OUT16.mean'},963968415,6000,'nds.ligo.caltech.edu:31200')

 

d = 


            name: 'H1:LSC-NPTRX_OUT16.mean' 

            chan_type: 'm-trend'             

            rate: 0.0167       

            data_type: 'real_8'     

            signal_gain: 1   

            signal_offset: 0     

            signal_slope: 1     

            signal_units: ''   

            start_gps_sec: 963968415     

            duration_sec: 6000             

            data: [100x1 double]           

            exists: 1

 

>> quit % since you've seen that the data is really here

$ kdestroy % so that no one uses your credentials

 

Some thoughts

  • I would like to extend this to the 32-bit machines, but I have to figure out the best way to install the proper NDS2 client without interfering with the 64-bit version. I think it is just a matter of specifying the matlabroot in the .../linux/ instead of .../linux64/
  • It would be nice to find a way that the nice tool gps('MM/DD/YYYY XX:XX:XX UTC'), which calls the ligotool executable tconvert, can be automatically usable when calling NDS2 functions. Right now, there seems to be an issue preventing that: even though tconvert can be run in the terminal, gps() returns an error and even directly running unix('tconvert now') or !tconvert returns the same error. I have emailed Peter Shawhan to see if he has any advice.

 

 

  6921   Thu Jul 5 13:12:12 2012 ZachUpdateComputersNDS2 client now working on Ubuntu machines

From my conversations with JZ and Leo, it seemed there was no package that generated the appropriate mex files. It was clear that the right ones weren't there from the absence of a /cvs/cds/caltech/apps/linux64/lib/matlab2010b directory. I'm sorry if I screwed anything up with pynds, but I have repeatedly asked for help with NDS2+matlab and no one has done anything.

It would be nice to do it via apt if there indeed is a versioned package that can make the mexs. Sorry again if I jumped the gun, but I didn't think anyone was going to do anything.

I guess the only 32-bit machine I can think of is mafalda.

About tconvert, I think the best solution is to make a new wrapper M-file. gps was just a convenient remnant of mDV, but all that we need is some matlab function that can output a GPS time given a date/time string. We can use whatever command-line utility you want.

  7136   Thu Aug 9 12:55:15 2012 ZachUpdateelogelog was being a pain in the ass; I restarted it

The elog was not responding. I attempted to restart it by running .../start-elog.csh, but this didn't work (even after the usual ~2 times it needs).

Somehow pkill did not kill the daemon, so I kill -9'd it and that did the trick. I ran the start script once more and it came back online.

  7782   Tue Dec 4 11:30:36 2012 ZachUpdatePSLPMC calibration for MC_F calibration

In order to have less unknown, you can calibrate the PMC PZT separately. Lock the PMC and take a transfer function from either the NPRO PZT input or the FSS AOM VCO input to the PMC control signal. The VCO is better, since the calibration should be much better known, but I am not sure what the current setup of the 40m PSL is, so I don't know if the FSS is normally locked.

Since you know the NPRO PZT or VCO actuation coefficients, you can assume the PMC loop (where the OL gain is high enough) is correcting for the frequency fluctuations. So, simply multiply the known coefficient by the transfer function to get the PMC PZT gain.

Then, you can re-do your PMC PZT sweep measurement and be confident of the calibration. The FSR must be right, so you can get the finesse with confidence.

Quote:

Hmm... Does anyone find falses in my measurement?
If not, the finesse can be 4 times smaller than the value which was 5 years ago?

 

  7965   Wed Jan 30 14:37:01 2013 ZachUpdateISSISS Design and Prototyping

Quote:

Unfortunately, as noted in the file, not all of the elements are possible to analyze in liso, such as any type of op-amp with more than two inputs and one output (AD602 used in this design has 16 pins with two distinct amplifiers contained within).

Typically, you can still find a way to model the important parts of the stages that are not as simply added. In the case of the differential input stage, in particular, it is important to include it because it will usually set the input noise level of the circuit. In this case, the noise is the same as the second stage (U5) and it has a gain of 1, so there is essentially no difference (up to factors of sqrt(2) or 2).

You can edit the opamp.lib file and add in custom components. For the input stage, you can just pretend it is a simple non-inverting amplifier with the specified noise characteristics from the datasheet: un = 1.3n, uc = 50 Hz (see below).

For dual op amps, you can usually just model each part separately. For example, the OPA2604 is a dual op amp that is included in the opamp.lib and can be treated as a single one in a model.

Screen_Shot_2013-01-30_at_4.22.46_PM.png

 

  7966   Wed Jan 30 15:45:09 2013 ZachUpdateLockingMode spacing calc

Quote:

Thus, the modes should be well separated

=>  spacing is 4.3 MHz while FWHM is 0.311 MHz  (cavity fsr = 34.5 MHz)

Something looks fishy. I calculate a transverse mode spacing of 2.21 MHz---is there a factor of two missing somewhere in your analytical calculation?

delta_f = (1/2/pi) * w01 - w00 = (1/2/pi) * acos(±sqrt(0.96)) /pi *2 * pi * c /2 /L = 2.21 MHz

I guess that's still OK, but if you are using 11-MHz sidebands, there is a n+m=5 mode within one linewidth of resonance. Can you use 55?

-------------

May I suggest my arbcav() tool for things like this? I think it's pretty handy for just this sort of calculations. I'm actually hoping to revamp the I/O to make it much cleaner and more intuitive.

>> T = [0.055 20e-6];

>> L = [4.34 4.34];

>> RoC = [115.5 1e10];

>> theta = [0 0];

>> fmod = 11e6;

>> lambda = 1064e-9;

>> num_pts = 1000;

>> loss = 50e-6;

>> [fin,coefs,df] = arbcav(T,L,RoC,theta,fmod,loss,lambda,num_pts);

>> fmod = 55e6;

>> [fin,coefs,df] = arbcav(T,L,RoC,theta,fmod,loss,lambda,num_pts);

 

HOM11.png HOM55.png

  7975   Thu Jan 31 15:20:46 2013 ZachUpdateLockingMode spacing calc

I should mention that I just found a bug in how it treats odd-mirror-number cavities. For such cavities, HG modes with odd horizontal indices should receive an extra roundtrip phase of pi/2 (due to the rotation by the cavity). Because of a numbering convention issue, arbcav actually used to apply this phase shift to even-order modes. Essentially, the only difference is that the fundamental mode was shifted to anti-resonance. Everywhere else, there are modes at both corresponding locations in frequency space, and so it does not back a big difference in terms of cavity design.

Thanks to this IMC modeling we are doing at the workshop, I caught it! It has been fixed in the SVN.

Quote:

I have calculated (using Zach's sweet software) the expected mode content for the various possible PRCs that we can make. 

Also, Zach was right about the factor of 2.  I see now that I was calculating the mode spacing between a plane wave and a HOM, so the guoy phase had a factor of (n+m+1).  The right thing to do is to get the spacing between the 00 mode and HOMs, so the guoy phase just has (n+m).  Switching from n+m+1=2 to n+m=1, that fixes the factor of 2 problem.

 I attach my results as a pdf, since I'm listing out 5 configurations.  Each config has a cartoon, with a small (hard to read) HOM plot, and then at the end, each HOM plot is shown again, but larger.  Also, "TM" is the "test mirror", the flat G&H that we're using as the cavity end mirror.

 

  7982   Fri Feb 1 12:22:27 2013 ZachSummaryASCOptics lit

It's OK; even Siegman got it wrong---48 times.

RA: NO, stil not OK.

Quote:

 Gouy not Guoy:

http://www.rp-photonics.com/gouy_phase_shift.html

pronounced Goo-eee, with the emphasis on the second syllable.

 

  8097   Mon Feb 18 00:03:46 2013 ZachUpdateComputer Scripts / ProgramsARBCAV v3.0

I have uploaded ARBCAV v3.0 to the SVN. The major change in this release, as I mentioned, is the input/output handling. The input and output are now contained in a single 'model' structure. To define the cavity, you fill in the substructure 'model.in' (e.g., model.in.T = [0.01 10e-6 0.01]; etc.) and call the function as:

model = arbcav(model);

Note: the old syntax is maintained as legacy for back-compatibility, and the function automatically creates a ".in" substructure in the output, so that the user can still use the single-line calling, which can be convenient. Then, any individual parameter can be changed by changing the appropriate field, and the function can be rerun using the new, simpler syntax from then on.

The function then somewhat intelligently decides what to compute based on what information you give it. Using a simple option string as a second argument, you can choose what you want plotted (or not) when you call. Alternatively, you can program the desired functionality into a sub-substructure 'model.in.funct'.

The outputs are created as substructures of the output object. Here is an example:

 

>> th = 0.5*acos(266/271) *180 /pi;

OMC.in.theta = [-th -th th th];

OMC.in.L = [0.266 0.284 0.275 0.271];

OMC.in.RoC = [1e10 2 1e10 2];

OMC.in.lambda = 1064e-9;

OMC.in.T = 1e-6 * [8368 25 8297 33];

OMC.in.f_mod = 24.5e6;

>> OMC

OMC = 

    in: [1x1 struct]

>> OMC = arbcav(OMC,'noplot')

Warning: No loss given--assuming lossless mirrors 

> In arbcav at 274 

OMC = 

         in: [1x1 struct]

        FSR: 2.7353e+08

        Lrt: 1.0960

    finesse: 374.1568

    buildup: 119.6956

         df: [1000x1 double]

      coefs: [1000x4 double]

        HOM: [1x1 struct]

>> OMC.HOM

ans = 

      f: [1x1 struct]

    pwr: [1x1 struct]

>> OMC.HOM.pwr

ans = 

    carr: [15x15 double]

     SBp: [15x15 double]

     SBm: [15x15 double]

 

Some other notes:

  • The annoying Mdo.m has been internalized; it is no longer needed.
  • For the next release, I am working on including:
    • Finite mirror thickness/intracavity refractive elements - If, for god knows what reason, you decide to put a mirror substrate within a cavity 
    • Mode overlap - Calculating the overlap of an input beam to the cavity
    • Mode matching - Calculating a mode matching telescope into the cavity for some defined input beam
    • Anything else?

I have added lots of information to the help header, so check there for more details. As always, your feedback is greatly appreciated.

  8125   Wed Feb 20 23:25:50 2013 ZachSummaryElectronicsReplacement for the AD743: OPA140 and OPA827

I have found two great FET input chips that rival the storied, discontinued AD743. In some ways, they are even better. These parts are the OPA140 and the OPA827.

Below is a plot of the input-referred voltage noise of the two op amps with Rsource = 0, along with several others for comparison. The smooth traces are LISO models. The LT1128 and AD797 are BJT-input parts, so their voltage noise is naturally better. However, the performance you see here for the FET parts is the same you would expect for very large source impedances, due to their extremely low current noise by comparison. I have included the BJTs so that you can see what their performance is like in an absolute sense. I have also included a "measured" trace of the LT1128, since in practice their low-frequency noise can be quite higher than the spec (see, for example, Rana's evaluation of the Busby Box). The ADA4627 is another part I was looking into before, the LT1012 is a less-than-great FET chip, and the AD797 a less-than-great BJT.

As you can see, the OPA140 actually outperforms the AD743 at low frequencies, though it is ~2x worse at high frequencies. The OPA827 comes close to the AD743 at high frequencies, but is a bit worse at low ones. Both the OPA140 and OPA827 have the same low-frequency RMS spec, so I was hoping it would be a better all-around part, but, unfortunately, it seems not to be.

The TI chips also have a few more things on the AD743:

  • Input current noise @ 1kHz
    • AD743: 6.9 fA/rtHz
    • OPA827: 2.2 fA/rtHz
    • OPA140: 0.8 fA/rtHz (!)
  • Input bias (offset) current, typ
    • AD743: 30 pA (40 pA) --- only for Vsupply = ±5 V
    • OPA827: ±3 pA (±3 pA) --- up to ±18V
    • OPA140: ±0.5 pA (±0.5 pA) (!) --- up to ±18V
  • Supply
    • Both OPA140 and OPA827 can be fed single supplies up to 36V absolute maximum
    • The OPA140 is a rail-to-rail op amp

These characteristics make both parts exceptionally well suited for very-high source impedance applications, such as very-low-frequency AC-coupling preamplifiers or ultra-low-noise current sources.

 ULN_opamp_comparison_2_20_13.png

(Apologies---the SR785 I was using had some annoying non-stationary peaks coming in. I verified that they did not affect the broadband floor).

R.I.P., AD743

  8151   Sat Feb 23 18:01:38 2013 ZachSummaryElectronicsReplacement for the AD743: OPA140 and OPA827

Rana suggested that I measure the OPA827 and OPA140 noise with high source impedance so as to see if we could find the low-frequency current noise corner. Below is a plot of both parts with Rs = 0, 10k, and 100k.

As you can see, both parts are thermal noise limited down to 0.1 Hz for up to Rs = 100k or greater. Given that the broadband current noise level for each part is ~0.5-1 fA/rtHz, this puts an upper limit to the 1/f corner of <100 Hz. This is where the AD743 corner is, so that sounds reasonable. Perhaps I will check with even higher impedance to see if I can find it. I am not sure yet what to make of the ~10-20 kHz instability with high source impedance.

OPA140_OPA827_noise_vs_Rs.png

EDIT: The datasheets claim that they are Johnson noise limited up to 1 Mohm, but this is only for the broadband floor, I'd guess, so it doesn't really say anything about the low frequency corner.

Screen_Shot_2013-02-24_at_12.12.23_PM.png Screen_Shot_2013-02-24_at_12.12.43_PM.png

Quote:

I have found two great FET input chips that rival the storied, discontinued AD743. In some ways, they are even better. These parts are the OPA140 and the OPA827.

 

  8420   Sun Apr 7 20:49:19 2013 ZachUpdateGeneralRestarted elog

with the script, as it was down.

  8428   Tue Apr 9 01:46:40 2013 ZachUpdateGeneralRestarted elog

Again.

Quote:

with the script, as it was down.

 

  10019   Tue Jun 10 11:54:36 2014 ZachConfigurationWikiDokuWikis are still down

Not sure if someone is already on this, but the nodus DokuWikis are still down (AIC, ATF, Cryo, etc.)

I get:

DokuWiki Setup Error

The datadir ('pages') does not exist, isn't accessible or writable. You should check your config and permission settings. Or maybe you want to run the installer

  10022   Wed Jun 11 05:16:14 2014 ZachConfigurationWikiDokuWikis are back up

It looks like auth is broken on the AIC wiki (though working fine on ATF and Cryo). I did some poking around but can't see how anything we did could have broken it.

Quote:

As of this writing, the DokuWiki appears to be working.

 

  10029   Wed Jun 11 20:09:37 2014 ZachConfigurationWikiDokuWiki situation

I have been looking into the DokuWiki situation, with no great progress so far.

The problem is with authentication. When you try to access the wiki, you get: "User authentication is temporarily unavailable. If this situation persists, please inform your Wiki Admin."

 As Evan points out, this goes away if you turn off ACL (Access Control Lists), but---as he also mentions---that just means that authentication is disabled, so the wiki becomes open. All signs point to this being a problem with the authentication mechanism. We use the 'authplain' plaintext method, where the usernames and hashed passwords are stored in a plaintext file.

Evan and I have both done plenty of testing with the config settings to see if the problem goes away. Even if you restore everything to default (but enable ACL using authplain and the existing user file), you still get an error.

I went as far as installing a brand new wiki on the web server, and, surprisingly, this also exhibits the problem. Interestingly, if I install an old enough version (2012-10-13), it works fine. After this revision, they transitioned their authentication methods from "backends" to "plugins", so this is a clue. As a side note, the other wikis on nodus (ATF and Cryo) are running earlier versions of DokuWiki, so they have no problems.

As it stood, our options were:

  1. No wiki (not even read-only, since the issue prevents logging in and also prevents anonymous reading, somehow)
  2. No ACL = open wiki. Also not good.

So, I created a temporary read-only version of the wiki using the aforementioned earlier DokuWiki release. It has a soft link to the real wiki data, but I deleted the user database so no one can log in and edit (I also disabled registration). It can be found at https://nodus.ligo.caltech.edu:30889/wiki_ro/.

I also created a backup of the real wiki as wiki_bak to avoid any potential disasters.

The simplest thing to do would be to just roll the wiki back to this working version, but that doesn't sound so smart. Clearly, it was working with the plugin structure before our snafu, and somehow our fixing everything else has broken it.

Quote:

Quote:

It looks like auth is broken on the AIC wiki (though working fine on ATF and Cryo). I did some poking around but can't see how anything we did could have broken it.

I went into local.php and changed $conf['useacl'] = 1; to $conf['useacl'] = 0; and it looks like the auth issue goes away (I've changed it back). This isn't a fix (we want to use access control), but it gives us a clue as to where the problem is.

 

  11058   Mon Feb 23 15:04:10 2015 ZachUpdateGeneralElog restarted

The elog crashed while I was creating an entry to the Cryo log. I restarted it with the start-elog.csh script.

  1042   Mon Oct 13 11:32:50 2008 YoixhiUpdatePSLMOPA is in trouble now
Steve, Alberto, Yoichi

A quick update.
The MOPA output went down to zero on Sunday early morning (00:28 AM).
We found that the NPRO beam is mis-aligned on the power monitoring PD (126MON).
We don't know yet if it is also mis-aligned to the power amplifier (PA) because the mechanical shutter is not working (always closed).
Most likely the beam is not aligned to the PA.
A mystery is that although the beam is terribly (more than a half inch) missing the monitor PD, the beam still goes through two faradays.
Another mystery is that the NPRO output power is now increased to 600mW.

The power drop was a very fast phenomenon (less than 1/16 sec).
We are trying to figure out what happened.
The first step is to fix the mechanical shutter. We have a spare in our hand.
  1915   Mon Aug 17 02:05:49 2009 Yoichi,ranaUpdatePSLReference cavity reflection looks bad
Rana, Yoichi

It has been a well known fact that the reference cavity reflection beam looks ugly.

We measured the visibility of the RC by locking and unlocking it.
Comparing the reflected beam powers, we got the visibility of 0.46,
which is pretty bad.

The beam going into the RC looks fine (circular on a sensor card).
However, the beam reflected back from the RC is distorted into a
horizontal ellipse, even when the RC is not locked.

We took a picture of the reflected beam hitting a white paper with the
infrared camera (see the attachment). It looks like two overlapping
circles horizontally separated. Could it be a badly coated optics
producing a secondary reflection ?

We looked into the RC's front mirror with an inspection mirror, but we
could not identify any obstructing object.

Rana is now touching the RC alignment.

We plan to remove the periscope before the RC to have a better look
into the cavity for inspection.


Late breaking update:
- We also moved the Refcav reflection camera to look at the leakage through a reflection steering mirror so that there's less chance of distortion. There was previously a W1 window in there as a pickofff. Also changed the camera to autogain so that we can see something.

- Re-aligned onto the refl pd.

- Tweaked alignment into RC. Mainly in yaw. Transmission went from 5V to 7V. In your face, Aso!
  7213   Fri Aug 17 04:54:01 2012 Yoichi, KojiSummaryLSCPRMI Locking

 To taste the strangeness of the current 40m PRC, I locked the PRMI with the guide of Koji.

We first aligned MICH by mostly tweaking ITMX, assuming that ITMY is in a good place as the Y-arm locks. MICH lock was stable.

Then we restored the IFO to the PRM_SBres mode. With a bit of alignment work on PRM and gain tweaking, the PRMI locked.

Yes, the beam spots look UGLY !

Also the PRMI was not so stable. Especially, when the alignment fluctuates, the optical gain changes and the loop becomes temporarily unstable. We took POP_DC as the guide for the gain change and normalized the PRCL error signal with it. To do this smoothly, we first changed the input matrix to route the PRCL error signal, which is REFL33_I, so that the signal also goes to the MC filter bank. Then with dtt, we monitored the spectra of the PRCL_IN1 and MC_IN1. We tweaked the value of the element in the normalization matrix for the MC path until the two spectra look the same (at this moment, the normalizing factor for the PRCL path was still zero). During this process, we noticed that the MC path signal (normalized by POP_DC) is noisier at above 500Hz. This was because the POP_DC has a large noise at high frequencies. So we put a low pass filter (100Hz 2nd order Butterworth) to the POP_DC filter bank to reduce the noise. Then the two spectra looked almost the same. The correct normalization factor found in this way was 0.03. So we put this number in the normalization matrix for PRCL. It did not break the PRMI lock.

 

After the normalization is turned on, the PRMI lock became somewhat more stable. However, the POP_DC was still fluctuating a lot, especially when the alignment is good. So I made a boost filter: 5Hz pole Q=2, 15Hz zero Q=1.5. I also made this filter automatically triggered when the PRMI is locked. This made the PRMI lock acquisition quicker. However, still the POP_DC fluctuation is large. It seems that the alignment of PRC is really fluctuating a lot.

 The current UGF of PRMI is about 150Hz with the phase margin over 50deg.

 

 

 

 

  1233   Fri Jan 16 18:25:32 2009 Yoichi, Kakeru, RanaUpdateLSCArms were unstable
The single arm lock had been unstable for both arms in the past few days.

Symptoms:
When an arm was locked by itself, the transmitted power showed a lot of fluctuations (sharp drops).
The first attachment shows the arm power fluctuations in power spectrum and time series.
References are when the boost filters are off for the arm feedback.
You can see that when the boosts are off, the power fluctuates a lot.
Also it is obvious that X-arm is a lot worse than Y.


Diagnosis:
The second attachment is the comparison of the error signal spectra between boosts on and off.
(PD3_I is the error signal of X-arm, PD4_I is Y arm). References are boost on.
Since the arm power fluctuation was suppressed by the gain increase, it was suspected that the main
reason for the power fluctuation is not alignment fluctuation. Rather, it is length or frequency fluctuation.

Then I took spectra and coherences of PD3_I, PD4_I and MC_F with both arms locked independently.
You can see broadband coherence between PD3_I (Xarm) and MC_F (frequency noise). In contrast the coherence
between PD4_I and MC_F is smaller. This means X-arm is more susceptible to the frequency noise than Y.
What can make a simple Fabry-Perot cavity more susceptible to frequency noise ? An offset ?
So I canceled the X-arm offset at the X-arm filter bank. Bingo ! The arm power fluctuation of X-arm became as small as Y-arm
in the dataviewer.
But what is making this offset ?
After watching the dataviewer screen for a while, the arm power fluctuation became larger again. I had to re-adjust the artificial offset
to minimize the fluctuation. This made me think that the source of the offset must be something to do with alignment.
In this case, clipping of the beam at the PD was very suspicious.
So I checked the centering of the POX and POY PDs. As expected, POX was terribly off-centered.
POY was also not exactly at the center of the plateau of DC output.
After centering those PDs, the large offset in the arm loops went away.
Now the arm powers are stable without artificial offset in the loop filters.
The last attachment shows the comparison of arm power fluctuation before and after the PD centering.
(references are the measurements before the centering).
  1196   Fri Dec 19 14:35:58 2008 Yoichi AlbertoUpdateIOOMC WFS and IOO-POS QPD re-centering
For the past two days, the MC alignment has kept drifting.
This morning, the MC alignment was so bad that it wouldn't lock to the TEM00 mode.
We aligned the MC mirrors manually until the reflection looks like a nice bull's-eye (the WFSs were off at this moment).
Then we un-locked the MC and centered the beams on the WFS QPDs.
Since the QPDs were saturated with the full laser power falling on them, I reduced the PSL power by turning the HWP after the MOPA.
After this, we turned on the WFSs and everything looks normal now.
We will see the trend of the MC related channels to monitor the drift.

Although unlikely, it might be caused by the drift of the input beam to the MC.
We found that the IOO-POS QPD was mis-centered and saturating.
We replaced the BS picking up the beam for the QPD from 33% reflection to 10% one. The QPD was still saturated.
So we put the 33% BS in the beam path to the QPD to further reduce the power. The beam kicked by the 33% BS
is dumped to a black aluminum plate. We should use a better beam dump later.
Now the IOO-POS QPD should tell us some information about the beam pointing of the PSL, though it has no sensitivity
to the relative motion of the PSL table to the vacuum chambers.
  526   Mon Jun 9 17:32:14 2008 YoichiConfigurationPSLPMC transmittance
I checked the current PMC transmissivity at a low power.
The input laser power to the PMC was reduced to 75mW by rotating the HWP in front of the PBS.
In this configuration, the output power from the PMC was 50mW. So the transmittance is about 66%.
The reading of C1:PSL-PMC_PMCTRANSPD is now 0.1 whereas it was 2.7 before turning the power down.

I will check the transmittance at a higher power when I get the cable for the 35W calorie meter, which is missing now.
  527   Mon Jun 9 17:57:59 2008 YoichiConfigurationPSLPMC input power backed to the original
I rotated back the HWP before the PBS to restore the input laser power to the PMC.
Now the reading of PSL-PMC_PMCTRANSPD is 2.7.
  534   Fri Jun 13 11:17:25 2008 YoichiUpdatePSLPMC transmittance at high power
We received a new cable for the Scientech calorimeter. So I measured the transmittance of the PMC at higher power.

Summary:
Input power = 2.298W
Output power = 1.364W
Transmittance = 59%

Detail:
The input power to the PMC was measured between the two mode matching lenses by the calorimeter.
2.298W looks a bit too low. Actually, the calibrated monitor PD on the MEDM screen shows about 3W output from MOPA.
So we (me and Steve) measured the power right after the PBS after the periscope from MOPA with the HWP set to maximize the transmission of the PBS.
It was 2.77W. According to Steve's previous measurement, the first mirror of the periscope transmits about 200mW of the incoming light to the monitor PD. So the actual output of the MOPA is about 2.97W, which is consistent with the monitor PD reading.
The aperture of the EOM for the PMC control is glowing a lot. We suspect this is the main cause of the loss (from 2.77W to 2.298W).
We may want to re-align the EOM.

The output light from the PMC was picked off by a glass slide. The reflectance of the glass slide was measured first at a lower power (input 98mW, reflected power 1.58mW). Assuming that the reflectance is the same for the higher power, I turned up the input power to the PMC. This time, the picked off power was 22.45mW. This means the actual output power is 98/1.58*22.45=1364mW. The glass slide was kept at the same angle through out the measurement.
The measurement of the output power was done by the Ophir power meter. So calibration difference between the Ophir and the calorimeter may introduce some error.
  540   Wed Jun 18 18:20:10 2008 YoichiUpdatePSLInvestigation on the NPRO temperature stabilization glitches
As Rob pointed out in http://dziban.ligo.caltech.edu:40/40m/537 the MOPA NPRO has been showing some glitchiness in the LTEC loop.
Following Rana's suggestion, Steve and I opened the MOPA and directed a heat gun for a minute to the NPRO hoping that we can see something in the LTEC loop.
The first attachment shows the behavior of LTMP and LTECH along with DTMP and DTECH at the time of the heat gun attack.
T=0 is the time when Steve directed the heat gun to the NPRO. There is no response neither in LTMP nor LTECH.
DTMP and DTECH look like responding.
Around the center, there is a dip in LTMP. This might be caused by removing the heat gun. But we are not sure. This kind of small glitches can be found in LTMP everywhere (see the attachment 2).
It looks like the LTMP sensor is not working, or the LTECH loop is actually working but the LTECH reading is broken.
However, the scan of the slow actuator (temperature) shows the LTECH loop is actually working. So it is a bit confusing.
More investigation is necessary.
See the next entry by me.
  541   Wed Jun 18 18:26:19 2008 YoichiUpdatePSLFinding the optimal operation temperature for the NPRO by the slow act scan
Being suspicious of the temperature stabilization of the NPRO crystal, I ran the slow scan script written by Rana to find the suitable operation temperature.
The procedure is the same as the one explained in the entry below:
http://www.ldas-sw.ligo.caltech.edu/ilog/pub/ilog.cgi?group=40m&task=view&date_to_view=09/04/2006&anchor_to_scroll_to=2006:09:04:22:23:56-rana
The attached plots show the results. By looking at C1:PSL-126MOPA_126MON, I set the slow slider voltage to 0.
This time, it looks like the temperature control of the NPRO crystal is working fine.
Obviously, PMC picks up many higher order modes. I will try to mode match/align the PMC later.
  548   Fri Jun 20 02:20:33 2008 YoichiUpdate PMC transmittance degradation
The PMC transmitted light power has been degrading constantly for last two weeks (see the attachment 1).
I went down to 2.55V.
The output of the MOPA is constant during this period. More strangely, the reflected power from the PMC is also constant.
One possible explanation is the contamination of the PMC mirrors. But I don't know why it started two weeks ago.

I tweaked the alignment of PMC and was able to recover the transmitted power to above 2.7V (attachment 2).
I will keep eye on this issue.
  569   Wed Jun 25 18:03:21 2008 YoichiConfigurationPSLFSS Input Offset slider problem
While working on the PMC scanning, I noticed that the FSS input offset slider is doing nothing.
I traced the signal flow and checked the cables/boards.
The slider changes the output voltage from a VMIVME4116 DAC in the PSL rack. This output voltage is confirmed to be correct at the FLKM64 connector. The signal is connected to the FSS servo interface box (D040423) trough a ribbon cable. However, the output from the interface box is always -27V regardless of the slider position.
Therefore, either the interface box (D040423) or the ribbon cable has a problem.
I will debug the interface box using an extension card when no one is working on the interferometer.
  575   Thu Jun 26 18:24:28 2008 YoichiUpdatePSLFSS input ofset slider problem - fixed
I checked the FSS servo interface board and found that a LT1125CSW used to differentialize offset channel was broken (no virtual short).
So I replaced it. Now the slider is working.
The op-amp was hitting the rail. So it seems like we had been applying the maximum offset to the FSS input all the time.
The reason why the FSS loop still worked with the large offset is that the applied offset (~14V) is attenuated by a factor of 500 at the summing point.
  607   Mon Jun 30 18:36:01 2008 YoichiUpdatePSLMZ alignment again
John, Yoichi

We re-adjusted the MZ alignment. The reason behind this is to make sure that the MZ dark port is not falling at a strange fringe, where it is only dark at the dark port PD. It can happen when the two beams poorly overlap.
We tried both the minimization of the MZ dark PD and the maximization of the MZ transmission at the same time.
We also placed another PD in the MZ dark port at a different distance from the original dark PD and tried to minimize this too.
If the MZ dark port is at a strange fringe, one of the dark PD can be dark where the other one is still bright.
If both of the dark PD get dark, the overlap between the beams should be ok.
We tweaked only the two mirrors of the MZ after the EOMs (mainly the one with a PZT).

Right now, the MZ dark power is 0.432.
BTW, we should change the name of the MZ dark port on the medm screen (it is now MZ reflection, where it is not a reflection).
I will try to change it later.

We wanted to put the beam position on the IOO-QPD_POS_* back to the original (before John tweaked the MZ alignment earlier).
However, the trends of IOO-QPD_POS_* show a lot of fluctuation and jumps, of which we don't know the cause. So we could not find reasonable original values.
We suspect a circuit problem in IOO-QPD_POS, especially because the jumps are very strange.
We will do this investigation later too.
  610   Tue Jul 1 11:53:38 2008 YoichiUpdateComputersRFM network back
I took a tour of the FE machines and power cycled all of them.
After executing the software restart procedures of those computers, the RFM network got back to the normal state.
For some reason, the computers requiring startup.cmd (like c1lsc) halt after running this command. Actually the computer is running ok, but the command freezes. Basically, what it does is simply to load a kernel module. I don't know what is wrong.
Anyway, I just closed the terminal after running startup.cmd and it seems fine for now.
  611   Tue Jul 1 11:57:24 2008 YoichiConfigurationPSLMZ servo switch problem again
C1:PSL-MZ_BLANK switch (to turn on/off the servo) is not working again. The switch is always off regardless of the epics state.
I pushed the cables into the xycom card, but it did not fix the problem.
  639   Mon Jul 7 13:49:27 2008 YoichiHowToPSLMZ offset, gain tips
John, Yoichi

This morning John found that MZ servo is not working.
We were able to bring the MZ back by changing the output offset a bit. But we were not sure what was actually wrong.
So we pulled out the MZ board and checked several TPs to understand the behavior.
Here is the summary of what we learned this morning.

The MZ control board has the following stages:

[Mixer] -(error signal)-> [Sum amp for input offset] -(error + offset)-> [Variable Gain Amp] -> [Filter (x100 DC gain)] -(FB signal)-> [High Voltage Amp] -> output
(The HV amp also works as the sum amp for the output offset)

(1) We noticed that the Sum amp for the input offset has an output of -0.14V even when the offset input is 0V. This can be canceled by the input offset slider.
So for the moment, it is fine. But we might want to change the op-amp because the weird offset implies there might be something wrong with the chip.
The procedure to null the -0.14V offset is the following:
a) Enable Test 1 input on the MZ MEDM screen.
b) Move the input offset slider until the mixer monitor becomes 0V. Currently the input offset slider is at -7.5V to cancel the -0.14V offset.

(2) Because the gain of the Variable Gain Amp and the Filter combined is large, the Filter can be easily saturated if the output offset is not right.
This was the cause of the MZ problem this morning. The output offset slider was at a wrong position making the error signal slightly off centered from zero.
This residual DC error signal was amplified by the large gain chain and saturated the filter amp.
Our experience is that the output offset cannot go below -3V. We set it at 0V for now.
  641   Mon Jul 7 14:02:05 2008 YoichiUpdateComputersSVN conversion progress
So far /cvs/cds/caltech/medm, /cvs/cds/caltech/chans and /cvs/cds/caltech/scripts have been converted to svn working copies.
Now /cvs/cds/caltech/target is being converted.
  649   Tue Jul 8 21:46:38 2008 YoichiConfigurationPSLGC650M moved to the PMC transmission
I moved a GC650M, which was monitoring the light coming out of the PSL, to the transmission port of the PMC to see the transmitted mode shape.
It will stay there unless someone find other use of it.

Just FYI, you can see the picture from the control computers by the following procedure:

ssh -X mafalda
cd /cvs/cds/caltech/target/Prosilica/40mCode
./SampleViewer

Chose 02-2210A-06223 and click on the Live View icon.
  654   Thu Jul 10 13:47:12 2008 YoichiHowToComputerssvn access via https
Now you can access to the svn repository on nodus by https.
To perform a checkout, you can use the following command

svn co --username svn40m https://nodus.ligo.caltech.edu:30889/svn/trunk/chans

This will check out "chans" directory.
The password for svn40m is written in the usual place.
You can also access the URL by a web browser to see the repository in a very primitive way.
A nice web interface for browsing the repository is planed but not yet implemented.
  692   Thu Jul 17 20:13:34 2008 YoichiUpdatePSLPMC alignment/mode matching effort
I'm working to improve the mode matching of PMC.
Because I noticed that the beam was hitting the aperture of the EOM for PMC, I moved the EOM a little bit to maximize the transmission.
This did not change the alignment to the reference cavity but changed the alignment of the PMC a lot.
So I adjusted it back.
The alignment of the PMC can be easily optimized but the Hermite 02 mode still remains. This means the mode matching is bad.
Moving the lenses by a small amount (a few mm) did not change the height of 02 mode.
I'm planning to move the lenses by a large amount tomorrow. But it will destroy the alignment to the PMC.
So I installed two irises in the beam path after the lenses to remember the alignment roughly.
Right now the PMC transmission is slightly worse than before because the lens positions are not good.
  699   Fri Jul 18 19:41:09 2008 YoichiUpdatePSLPMC PZT investigation
I measured the HV coming to the PMC PZT by plugging it off from the PZT and hooking it up to a DVM.
The reading of DVM is pretty much consistent with the reading on EPICS. I got 287V on the DVM when the EPICS says 290V.

Then I used a T to monitor the same voltage while it is connected to the PZT. I attached a plot of the actual voltage measured by the DVM vs the EPICS reading.
It shows a hysteresis.
Also the actual voltage drops by more than a half when the PZT is connected. The output impedance of the HV amp is 64k (according to the schematic). If I believe this number, the impedance of the PZT should also be 64k. The current flowing the PZT is 1.6mA at 200V EPICS reading.
The impedance of the PZT directly measured by the DVM is 1.5M ohm, which is significantly different from the value expected above. I will check the actual output impedance of the HV amp later.
The capacitance of the PZT measured by the DVM is 300nF. I don't know if I can believe the DVM's ability to measure C.

I noticed that when a high voltage is applied, the actual voltage across the PZT shows a decay.
The second plot shows the step response of the actual voltage.
The voltage coming to the PZT was T-ed and reduced by a factor of 30 using a high impedance voltage divider to be recorded by an ADC.
The PMCTRANSPD channel is temporarily used to monitor this signal.
After the voltage applied to the PZT was increased abruptly (to ~230V), the actual voltage starts to exponentially decrease.
When the HV was reduced to ~30V, the actual voltage goes up. This behavior explains the weird exponential motion of the PZT feedback signal when the PMC is locked.
The cause of the actual voltage drop is not understood yet.
From the above measurements, we can almost certainly conclude that the problem of the PMC is in the PZT, not in the HV amp nor the read back.
  700   Fri Jul 18 19:43:55 2008 YoichiDAQComputersPSL fast channels cannot be read by dataviewer
At this moment only the PSL fast channels have trouble.
Rob restarted fb40m, c1IOVME, but no effect.
  703   Sat Jul 19 19:41:56 2008 YoichiAoGPSLThe author of the entry 702 is Yoichi not Rob
I made a mistake.
  717   Tue Jul 22 22:11:58 2008 YoichiUpdateLSCX-arm g factor measurement
Alberto, Yoichi

We measured the g factor of the X-arm by slightly shifting the 166MHz sideband frequency:

We first locked the X-arm to TEM00 mode. Then misaligned the ETMX in yaw a little bit until the transmitted power is a half of the normal value.
In this way, we can expect that TEM01 mode will be resonated in the arm if a sideband with a suitable frequency is introduced.
To add such a sideband, we used the 166MHz EOM. According to John's calculation (ELog entry 690), the TEM01 mode of the 166MHz upper sideband is only about 100kHz away from the resonance. So by changing the 166MHz modulation frequency, we should be able to see the 166MHz upper sideband resonating in the X-arm.
We used the 166MHz PD at the AS to find the resonance.
When we modulated the 166MHz RF frequency by +/- 100kHz, we could see spikes in the AS166_I signal.
Then we fine tuned the RF frequency slowly by hand to find the exact resonant frequency. At that time, the X-arm PDH servo was oscillating at ~480Hz.
So the resonance was determined by maximizing this signal in the AS166_I.
The 166MHz signal was originally at 165.977195 MHz. I found the resonance around 165.985MHz. It is surprisingly close to the original modulation frequency (only 7.3kHz away). This number yields the g factor of 0.626 and the transverse mode interval of 0.285*FSR. I used the arm length of 38.5750m in this calculation. Because of the 480Hz oscillation, it was difficult to precisely determine the resonant frequency. We will try this again tomorrow after mitigating the oscillation.
Although the resonance of the 166MHz upper sideband is located at a lower frequency in John's prediction, we found a resonance at a higher frequency.
This can be interpreted as the discrepancy between the actual g-factor and the designed g-factor.

To confirm what we saw was really an arm cavity resonance, we will try to do the same thing with the arm cavities all mis-aligned.
(We expect no signal in this configuration.)

Appendix: the expected signal from AS166 port when the 166MHz upper sideband passes by a resonance of the arm cavity.
Since the carrier is resonating in the cavity and kept there by the PDH feedback using 33MHz sideband, its phase is virtually fixed at the AS166 port. The lower sideband's phase also does not change much because it is off resonance. The upper sideband get a large phase change when approaching to the resonance. This effectively rotates the modulation angle of the 166MHz sidebands, which was orthogonal to the carrier when off resonance (i.e. phase modulation), to create 166MHz amplitude modulation. Because the sideband axis is rotated, the signal should appear both in I and Q phases.
  728   Wed Jul 23 22:34:07 2008 YoichiUpdateLSCArm cavity g-factor measurement
I tried the same thing as the X-arm to the Y-arm.
I'm puzzled. I found exactly the same behavior as the X-arm in the AS166 demodulated signals, whereas I expected different resonance frequency because of the arm length difference.

Here is more detailed account of the measurement today.

I locked the Y-arm and mis-aligned the end mirror in Yaw until the transmission power gets half.
Then I injected a 30Hz sinusoid into the error point of the Y-arm servo to shake the ETMY.
I observed AS166_I and AS166_Q as I changed the 166MHz frequency.

At 165.977MHz, both AS166_I and AS166_Q showed the 30Hz signal (15cnt p-p).
At 165.981MHz, Only I phase showed the 30Hz signal (40cnt p-p). No signal in Q.
At 165.984MHz, I and Q became the same amplitude again (20cnt p-p).
At 165.987MHz, Only Q phase showed the 30Hz signal (40cnt p-p). No signal in I.

Outside the above range, the signal decreases as the frequency go away. I think this is (at least partly) because the 166MHz sidebands no longer go through the MC at those frequencies.

I then locked the X-arm to the TEM01 mode. I saw exactly the same behavior as described above. This could be the resonance of TEM02 mode. I was expecting to see the resonance of TEM00 mode at the opposite side, but nothing there.

I unlocked the arm cavities and tried the same frequency scan of the 166MHz with one of the end mirrors shaken at 30Hz. I saw no signal at the AS166 port.
I also tried locking Y-arm and shaking the ETMX. No signal.
So it has to be something to do with the cavity resonance.

Since the MC transmission curve for 166MHz is folded in the measurement, it makes the interpretation of the results harder.
ELOG V3.1.3-