40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log, Page 285 of 344  Not logged in ELOG logo
ID Date Author Type Category Subjectdown
  14021   Tue Jun 26 17:54:59 2018 poojaUpdateCamerasDeveloping neural networks

Aim:  To find a model that trains the simulated data of Gaussian beam spot moving in a vertical direction by the application of a sinusoidal signal. The data also includes random uniform noise ranging from 0 to 10.

All the attachments are in the zip folder.

I simulated images 128*128 at 10 frames/sec by applying a sine wave of frequency 0.2Hz that moves the beam spot, added random uniform noise ranging from 0 to 10 & resized the image frame using opencv to 64*64. 1000 cycles of this data is taken as train & 300 cycles as test data for the following cases. Optimizer = Nadam (learning rate = 0.001), loss function used = mean squared error, batch size = 32,

Case 1:

Model topology:

                         256 (dropout = 0.1)  ->           256 (dropout = 0.1)   ->       1

Activation :             selu                                         selu

Number of epochs = 240.

Variation in loss value of train & test datasets is given in Attachment 1 of the attached zip folder & the applied signal as well as the output of neural network given in Attachments 2 & 3 (zoomed version of 2).

The model fits well but there is no training since test loss is lower than train loss value. I found in several sites that dropout of some of the nodes during training but retaining them during test could be the probable reason for this (https://stackoverflow.com/questions/48393438/validation-loss-when-using-dropout , http://forums.fast.ai/t/validation-loss-lower-than-training-loss/4581 ). So I removed dropout while training next time.

Case 2:

Model topology:

                         256 (dropout = 0.1)  ->           256 (dropout = 0.1)   ->       1

Activation :             selu                                         selu                          linear

Number of epochs = 200.

Variation in loss value of train & test datasets is given in Attachment 4 of the attached zip folder & the applied signal as well as the output of neural network given in Attachments 5 & 6 (zoomed version of 2).

But still no improvement.

Case 3:

I changed the optimizer to Adam and tried with the same model topology & hyperparameters as case 2 with no success (Attachments 7,8 & 9).

Finally I think this is because I'm training & testing on the same data. So I'm now training with the simulated video but moving it by a maximum of 2 pixels only and testing with a video of ETMY that we had captured earlier.

  13937   Sun Jun 10 15:04:33 2018 poojaUpdateCamerasDeveloping neural network

Aim: To develop a neural network in order to correlate the intensity fluctuations in the scattered light to the angular motion of the test mass. A block diagram of the technique employed is given in Attachment 1.

I have used Keras to implement supervised learning using neural network (NN). Initially I had developed a python code that converts a video (59 sec) of scattered light, after an excitation (sine wave of frequency 0.2 Hz) is applied to ETMX pitch, to image frames (of size 480*720)  and stores the 2D pixel values of 1791 images frames captured into an hdf5 file. This array of shape (1791,36500) is given as an input to the neural network. I have tried to implement regular NN only, not convolution or recurrent NN. I have used sequential model in Keras to do this. I have tried with various number of dense layers and varied the number of nodes in each layer. I got test accuracy of approximately 7% using the following network. There are two dense layers, first one with 750 nodes with a dropout of 0.1 ( 10% of the nodes not used) and second one with 500 nodes. To add nonlinearity to the network, both the layers are given an activation function of tanh. The output layer has 1 node and expects an output of shape (1791,1). This model has been compiled with a loss function of categorical crossentropy, optimizer = RMSprop. We have used these since they have been mostly used in the image analysis examples. Then the model is trained against the dataset of mirror motion. This has been obtained by sampling the cosine wave fit to the mirror motion so that the shapes of the input and output of NN are consistent. I have used a batch size ( number of samples per gradient update) = 32 and epochs (number of times entire dataset passes through NN) = 20. However, using this we got an accuracy of only 7.6%. 

I think that the above technique gives overfitting since dense layers use all the nodes during training apart from giving a dropout. Also, the beam spot moves in the video. So it may be necessary to use convolution NN to extract the information.

The video file can be accesses from this link https://drive.google.com/file/d/1VbXcPTfC9GH2ttZNWM7Lg0RqD7qiCZuA/view.

Gabriele told us that he had used the beam spot motion to train the neural network. Also he informed that GPUs are necessary for this. So we have to figure out a better way to train the network.  


gautam noon 11Jun: This link explains why the straight-up fully connected NN architecture is ill-suited for the kind of application we have in mind. Discussing with Gabriele, he informed us that training on a GPU machine with 1000 images took a few hours. I'm not sure what the CPU/GPU scaling is for this application, but given that he trained for 10000 epochs, and we see that training for 20 epochs on Optimus already takes ~30minutes, seems like a futile exercise to keep trying on CPU machines.

  13972   Fri Jun 15 09:51:55 2018 poojaUpdateCamerasDeveloping neural network

Aim : To develop a neural network on simulated data.

I developed a python code that generates a 64*64 image of a white Gaussian beam spot at the centre of black background. I gave a sine wave of frequency 0.2Hz that moves the spot vertically (i.e. in pitch). Then I simulated this video at 10 frames/sec for 10 seconds. Then I saved this data into an hdf5 file, reshaped it to a 1D array and gave as input to a neural network. Out of the 100 image frames, 75 were taken as training dataset and 25 as test data. I varied several hyperparameters like learning rate of the optimizer, number of layers, nodes, activation function etc. Finally, I was successful in reducing the mean squared error with the following network model:

  • Sequential model of 2 fully connected layers with 256 nodes each and a dropout of 0.1
  • loss function = mean squared error, optimizer = RMSprop (learning rate = 0.00001) and activation function that adds nonlinearity = relu
  • batch size = 32 and number of epochs = 1000

I have attached the plot of the output of neural network (NN) as well as sine signal applied to simulate the video and their residula error in Attachment 1. The plot of variation in mean squared error (in log scale) as number of epochs increases is given in Attachment 2.

I think this network worked easily since there is no noise in the input. Gautam suggested to try the working of this network on simulated data with a noisy background.

 

  14114   Sun Jul 29 23:15:34 2018 poojaUpdateCamerasDeveloping CNN

Aim: To develop a convolutional neural network that resolves mirror motion from video.

Input : Previous simulated video of beam spot motion in pitch by applying 4 sine  waves of frquencies 0.2, 0.4, 0.1, 0.3 Hz  and amplitude ratios to frame size to be 0.1, 0.04, 0.05, 0.08 where random uniform noise ranging 0.05 has been added to amplitudes and frequencies. This is divided into train (0.4), validation (0.1) and test (0.5).

Model topology:

  • Number of filters = 2
  • Kernel size = 2
  • Size of pooling windows = 2
  •                                        ----->         Dense layer of 4 nodes  ---->    Output layer of 1 node 

         Activation:                      selu                                                  linear

Batch size = 32, Number of epochs = 128, loss function = mean squared error

Optimizer: Nadam ( learning rate = 0.00001, beta_1 = 0.8, beta_2 = 0.85)

Plots of CNN output & applied signal given in Attachment 1. The variation in loss value with epochs given in Attachment 2.

This needs to be further analysed with increasing random uniform noise over the pixels and by training CNN on simulated data of varying ampltides and frequencies for sine waves.

  13109   Mon Jul 10 21:31:15 2017 KaustubhHowToComputer Scripts / ProgramsDetails on Cavity Scan Analysis

Summary:

The following elog describes the procedure followed for generating a sample simulation for a cavity scan, fitting an actual cavity scan and calculating the relevant paramaters using the cavity scan and fit data.

 

1. Cavity Scan Simulation:

  1. First, we define the sample cavity parameters, i.e., the reflectivitie,transmissivities of the mirrors, the RoCs of the mirrors and the absolute cavity length.
  2. We then define a frequency range using numpy.linspace function for which we want to take a scan.
  3. We then define a function that returns the tranmission power output of a Fabry-Perot cavity using the cavity equations as follows: P_{t} = \frac{t_{1}t_{2}}{1-r_{1}r_{2}\exp({\frac{4\pi Lf}{c}+(n+m+1)\phi_G)}} where Pt is the transmission power ratio of the output power to input power, t1,t2,r1,r2 are the transmissivities and reflectivities of the two mirrors, L is the absolute cavity length, f is the frequency of the input laser, c is the speed of light, \phi_G = \arccos{g_{1}g_{2}} is the gouy phase shift with g1,g2 being the g-factors for the two cavity mirrors(g=1-L/R). 'n' and 'm' correspond to the TEMnm higher order mode.
  4. We now obtain a cavity scan by giving the above defined function the cavity parameters and by adding the outputs for different higher order mode('n', 'm' values). Appropriate factors for the HOMs need to be chosen. The above function with appropriate coefficients can be used ti also add the modulated sidebands to the total transmission power.
  5. To this obtained total power we can add some random noise using numpy modules random.normal function. We need to normalise the data with respect to the max. power transmission ratio.
  6. We can now perform fitting on the above data using the procedure stated in the next section and then plot the two data sets using matplotlib module.
  7. A similar code to do the above is given here.

 

2. Fitting a Cavity Scan:

  1. The actual data for a cavity scan can be found in this elog entry or attached below in the zip folder.
  2. We read this data and separate the frequency data and the transmission data.
  3. Using the peakutils module's indexes function, we find the indices of the various peaks in the data set.
  4. These peaks are from the fundamental resonances, sideband resonances(both 11MHz and 55MHz) as well as a few HOMs.
  5. Each of these resonances follows the cavity equations and hence can be modelled as Lorentzian within small intervals around the peak frequencies. A detailed description of how this is possible is given here and is in the atached zip folder('Functionsused.pdf').
  6. We define a Lorentzian function which returns the fo\frac{a}{1+(\frac{\nu - \nu_0}{b})^{2}}llows:  where 'a' is the peak transmission value, 'b' is the 'linewidth' of the Lorentzian and \nu_0 is the peak frequency  about which the cavity equations behave like a lorentzian.
  7. We now, using the Lorentzian function, fit the various identified peaks using the curve_fit function of the scipy module. Remember to turn the 'absolute_sigma' parameter to 'True'.
  8. The parameters now obtained can be evaluated using the procedure given in the next section.
  9. The total transmission power is evaluated by feeding in the above obtained parameters back into the Lorentzian function and adding it for each peak.
  10. We can plot the actual data set and the data obtained using the fit of different peaks in a plot using matplotlib module. We can also plot the residuals for a better depiction of the fit quality.  
  11. The code to analyse the above mentioned cavity scan data is given here and attached below in the zip folder.

 

3. Calculating Physically Relevant Parameters:

  1. The data obtained from the fitting the peaks in the previous section now needs to be analysed in order to obtain some physically relevant information such as the FSR value, the TMS value, the modulation depths of the sidebands and perhaps even the linear caliberation of the frequency.
  2. First we need to identify the fundamental, TEM00 resonances among all the peaks. This we do by using the numpy.where function. We find the peaks with transmission values more than 0.9(or any suitable value).
  3. Using these indices we will now calculate the FSR and the Finesse of the peaks. A description of the correlation between the Fit Parameters and the FSR and Finesse is given here.
  4. We define a Linear fitting function for fitting the frequency values of the fundamental resonances against the ith fundamental resonance. The slope of this line gives us the value of FSR and the error in it.
  5. The Finesse can be calculated by fitting the linewidth with a constant function.
  6. The cavity length can be calculated using the FSR values as follows: L = \frac{c}{2\nu_{FSR}}.
  7. Now, the approximate positions of the sideband frequncies is given by 11*106%FSR and 55*106%FSR away from the fundamental, carrier resonances.
  8. The modulation depth, 'm', is given as \sqrt{\frac{P_{c}}{P_{s}}} = \frac{J_{0}(m)}{J_{1}(m)} where Pc is the carrier transmission power, Ps is the transmission power of the sideband and Jv is the Bessel Function of order 'v'.
  9. We define a function 'Bessel Ratio' using which we'll fit the transmission power ratio of the carrier to the sideband for the multiple sideband resonances.
  10. We also check for the Linearity in frequency data by plotting Fitting the frequencies corresponding to peaks in the actual data to ones obtained after fitting.
  11. After this we attempt to identify the other HOMs. For this we first determine a rough estimate for the value of TMS using the already known parameters of the mirrors,i.e., the RoC. We then look in small intervals (0.5 MHz) around frequencies where we would expect the HOMs to be, i.e., 1*TMS, 2*TMS, 3*TMS... away from the fundamental resonances. These positions are all modulo FSR.
  12. After identifying the HOMs, we take the difference from the fundamental resonance and then study these modulo the FSR.
  13. We perform a Linear Fit between these obtained values and (n+m).  As 'n','m' are degenerate, we can simply perform the fit against some variable 'k' and obtain the value of TMS as the slope of the linear fit.
  14. The code to do the above stated analysis is given here.

 

Most of the above info and some smaller details can be found in the markdown readme file in this git repo.

  7271   Fri Aug 24 14:46:08 2012 JenneSummaryGeneralDetailed alignment plan

Friday / pre-vent:

[done] Align the MC mirrors for the incident beam so that the mirrors can be the alignment reference [Koji]

[in progress] Center spots on MC mirrors [Jenne]

Put beam attenuator optics (PBS + waveplate) on PSL table, realign input beam to MC mirror centers

[In progress] See if we can design a set of nuts and bolts to use at bottom of tiptilt optic ring, to do small adjustments of pitch alignment [Steve]

After doors open:

Use CCD (Watek, with AGC on) to take images of everything we can think of, to see current status of clipping

Check that we get through the Faraday without clipping

Move PZT1 and MMT mirrors to get good spot positions on PR3, PR2.  Make sure we're clearing the Faraday's housing

Install dichroic optics, perhaps completely readjust pitch alignment of those tiptilts (we will measure the spares later, and call that good enough for our phase mapping).

Use some kind of oplev setup to check pitch alignment of PR2, PR3.

Tweak (if necessary) PR2 & PR3 pitch to go through center of PRM, BS, hit center of ITMY

Check that we're not clipping on the BS cage anywhere

Use CCD to take images with Sensoray of everything we can think of, to confirm we don't have clipping anywhere.  Want to see the edges of the beam on the targets, which would mean that the beam is hitting the center of the optic.  If necessary, we'll stay open an extra day to get good camera images everywhere, so we have a good record of what's going on inside.

Note:  While having good arm alignment would be good, we're willing to sacrifice some arm alignment to have good DRMI alignment, since we're re-venting and installing the new active tiptilts in another month or so. 

Things I'm leaving for Jamie-the-Vent-Czar to plan:

Order of door opening

Beam dump assembly and placement

 

  5530   Fri Sep 23 16:56:07 2011 MirkoUpdateLSCDesired MC modulation frequency measurement, tuning of modulation frequency

[ Mirko, Koji, Suresh ]

Looked into the modulation frequency that should pass the input MC. With a locked MC looked at the RF output of the PD in refl of the MC. Looked at the beat between 11MHz and 29.5MHz. Minimizing it by fine-tuning the 11MHz freq. ( which means maximizing the 11MHz transmission).

SB freq. [MHz]     Beat power [dBm]

11.065650          -75

11.065770          -80 (diving into spec. analyzer instrument noise)

11.066060          -80 (surfacing out of spec. analyzer instrument noise)

Set the freq. to the middle of the last two points: 11.065910MHz at 16:26.

ToDo: How big a problem is the AM?

  4441   Thu Mar 24 19:48:13 2011 Aidan, KiwamuUpdateGreen LockingDesigns for permanent electronics for ALS

Kiwamu and I looked at all the electronics that are currently in place for the green locking on the X-arm and have made a set of block diagrams of the rack mounted units that we should build to replace the existing ... "works of art" that sprawl around out there at the moment.

Main items

1. "ETM Green Oscillator/PDH support box". Not a great name but this would provide the local oscillator signal for the end PDH (with a controllable phase rotator) as well as the drive oscillator for the end laser PZT. Since we need to hit a frequency of 216.075kHz with a precision that Kiwamu needs to determine, we'd need to be able to tune the oscillator ... it needs to be a VCO. It'd be nice to be able to measure the output frequency so I've suggested dividing it down by N times to put it into the DAQ - maybe N = 2^7 = 128x to give a measured frequency of around 1.7kHz. Additionally this unit will sum the PDH control signal into the oscillation. This box would support the Universal PDH box that is currently at the X-end.

2. "Vertex X-arm beatnote box" - this basically takes the RF and DC signals from the beatnote PD and amplifies them. It provides a monitor for the RF signal and then converts the RF signal into a square wave in the comparator.

3. "Mixer Frequency Discriminator" - just the standard MFD setup stored in a box. For temperature stability reasons, we want to be careful about where we store this box and what it is made of. That's also the reason that this stage is separated from the X-arm beatnote box with it's high-power amps.

Other things

4. RS232 and EPICS control of the doubling ovens

5. Intensity stabilization of the End Laser

P.S. I used Google Diagrams for the pictures.

  5773   Mon Oct 31 21:46:32 2011 kiwamuUpdateLSCDependence of Recycling gain on incident beam pointing

I horizontally swept the translation of the incident beam in order to investigate a possible clipping in Power-recycled Michelson (PRMI).

The recycling gain of PRMI depended on the beam pointing but it did't improve the recycling gain.

I guess the amount of the entire shift I introduced was about +/- the beam diameter = +/- 5 mm or so.

Within the range of about +/- 5mm in the horizontal beam translation I didn't find any obvious sign of a clipping.

 

(Measurement)

 This is the procedure which I did:
  (1) Some amount of offsets were introduced on MC2 in both PIT and YAW such that the PZT1 won't rail (#5762).
      => Every time when I introduced the offset I realigned the zig-zag mirrors on the PSL table to maintain the high transmissivity of MC.
  (2) Fine tuning of the MC offsets so that the PZT1_X EPICS value becomes almost zero when the beam is aligned down to the Y arm.
     => 0.523 in C1:ASC_PZT1_X became a point where the coupling of the beam into the Y arm was maximized.
     => Last time the direction which we investigated was the positive side from this zero point (#5709) in PZT1_X.
  (3) Aligned MICH by steering BS.
  (4) Locked PRMI with carrier resonating and aligned PRM to maximized the power recycling gain which was obtained from POYDC.
  (5) Translated the beam pointing
     => First I shifted PZT1_X by a wanted amount.
    => Then I locked the Y arm and realigned PZT2_X by maximizing the Y arm transmission.
          This procedure should give us a pure translation on the incident beam.
  (6) Repeated the same procedure (3) through (5) in each PZT1 position.
 

(Results)

Here shows the measured recycling gain and the power reflectivity of PRMI as a function of the beam pointing.

beam_scan_PRCL.png

 Upper plot : measured recycling gains (Red) observed maximum values (Black) measured values on average.
 Lower plot : measured power reflectivity of PRMI (Blue) observed minimum values (Black) measured values on average.
 
 As shown in the plots the recycling gain could go up to 8 at some points.
As the PZT went away from 0 it decreased and eventually became about 3 in each side.
The reflectivity showed the minimum value of 0.4 when the PZT1 was at -1 in EPICS value.
One hypothesis to explain this plot can be that : we are just seeing the effect of the incident beam misalignment.

 

  1806   Wed Jul 29 13:15:35 2009 ClaraUpdatePEMDents = Bad??

I was in the lab last night accelerometerizing and noticed some dents on the tubes that stick out horizontally from the MC2 optical chamber (sorry, I don't know what they're called or what they do). One of them is pretty big... I don't know if this is a problem, but it probably isn't a good thing. Photos below:

big_dent1.png

big_dent2.png

small_dent.png

This last one is a little hard to see... I was having trouble getting a good angle on it, but it's there. Not quite as significant as the first one though. (The first two pictures are of the same dent.)

  1811   Wed Jul 29 19:46:04 2009 rana, albertoUpdatePEMDents = Bad??
It looks like the MC2 chamber and/or stack has been jarred and shifted. Please be careful and use much less force and speed around the MC2 chamber.

My guess is that the work with the accelerometers around there had made the MC2 angle and
position change last night. The reason that we don't see it in the OSEMs so much is that its
a change in the actual stack position and tilt.

To recover, we changed the MC2 alignment bias to get the beam through the Faraday. This did NOT get
the beam back onto the right place on the MC TRANS QPD. For tonight we decided to not recenter that
since Rob might not like this position. We did, however, zero out the MC WFS and the PSL POS/ANG.

If the interferometer locking is OK tonight, then we (Steve and whoever else is here at 7 AM)
should recenter IP POS and IP ANG and also fix up the PSL POS and PSL ANG QPDs. You can see
in the attached picture that there are two problems to fix:
1) The knobs (circled in red and blue) are wrapped in foil. Why???
2) The handedness of the mirror mount with the orange arrow is wrong. This should be unmounted and clocked
by 90 deg. Right now the beam is nearly clipping on the mount. Also, we need to change the channel names
on the PSL POS (or maybe its ANG). It has the horizontal/vertical channels misnamed.
  8494   Thu Apr 25 20:48:48 2013 KojiUpdateASSDen fixed the Yarm ASS scripts

I contacted Den about malfunctioning of the Yarm ASS.

He found the scripts were modified during the attempt to make it available for Xarm (cf. a related elog entry)
So far, he could manage to make the current scripts being modified to run.
A striptool file is still missing but this is what we can handle locally.

I thank Den for the remote caring of the issue despite the limited network bandwidth.

  8525   Fri May 3 01:24:25 2013 KojiUpdateASSDen fixed the Yarm ASS scripts

Output matrices are added to ASS. Currently ASS is based on the mirror bases.
I prefer to have the actuator bases as the coils are more stable than the sensors.

At this point, the output matrices are identity. So Den's scripts are still working.

Striptool settings were also fixed.

  10280   Mon Jul 28 10:42:43 2014 NichinUpdateElectronicsDemodulator board's characterization

 I used vector fitting to fit the transfer functions between RF input and PD RF MON of demodulator boards. These fittings can certainly do a lot better on LISO, but for the time being I will assume these to be good enough and change the main PDFR scripts to calibrate out this factor and get a decent reading of PD transimpedance. Then it will just be a matter of changing the transfer function parameters. A lot of work needs to be done on the PDFR interface and plot features.

Attached: The plots showing data and fits.

  9103   Wed Sep 4 17:22:09 2013 JenneUpdateLSCDemod phases for RFPDs

I checked the demod phases for AS55, POP110, and all the REFL PDs. 

AS55:  I locked MICH, and shook the ITMs (+1 for Y, -1 for X), and watched the AS55 I & Q spectra at 580Hz (notch in the servo was enabled).  I rotated the phase from -32.0 to +13.0 to get the signal entirely in the Q phase.

POP110:  I locked the PRMI (triggering on POP22), and maximized POP22.  I then rotated the phase of POP110 until the signal was maximally positive.  I forgot what the starting phase was, but it is now 84.  The POP11_I signal was entirely negative when I started, so the new phase is about 180 from the old phase.  I also checked by unlocking the cavity, and seeing that a large peak in POPDC corresponded to large negative dips in POP110_I and POP22_I.

REFL PDs:  I locked the PRMI, and shook the PRM (notches in the servos were enabled for both MICH and PRCL).  Maximized the peak in the I phase.  REFL11 was fine, REFL33 was fine.  REFL55 was changed from 120 to 45.  REFL165 was changed from 106 to 96.

I restored the SRM on the IFO_ALIGN screen, but the saved value was almost 2 full integers off in yaw from actual DRMI resonances.  It looks like it was saved when Rana and I were working late last week.  We must have accidentally saved it when it was misaligned, since hysteresis can't do that much.

I want to check the phases for POX and POY with arm locking, just in case.  Also need to set the AS110 phase (which is plugged into the AS11 channels - need to fix the channel names).

 

  15840   Wed Feb 24 12:11:08 2021 gautamUpdateGeneralDemod char part 3

I did the characterization discussed at the meeting today.

  1. RF signal at 100 Hz offset from the LO frequency was injected into the PD input on the demod boards.
  2. The digitized CDS channels were monitored. I chose to look at the C1:LSC-{PD}_I_OUT and C1:LSC-{PD}_Q_OUT channels. This undoes the effect of the analog whitening, but is before the digital phase rotation.
  3. Attachments #1 and Attachments #2 are for the case where the analog whitening is not engaged, white Attachments #3 and Attachments #4 are for when the whitening is engaged, and they look the same (as they should), which rules out any crazy mismatch between the analog filter and the digital dewhitening filter.
  4. I have absorbed the flat whitening gain applied to the various PDs in the cts/V calibration indicated on these plots. So the size of the ellipse is proportional to the conversion gain.

I think this test doesn't suggest anything funky in the analog demod/whitening/AA/digitization chain. We can repeat this process after the demod boards are repaired and use the angle of rotation of the ellipse to set the "D" parameter in the CDS phase rotator part, I didn't do it today.

  15839   Wed Feb 24 11:53:24 2021 gautamUpdateGeneralDemod char part 2

I measured the noise of the I/F outputs of all the LSC demodulators. I made the measurement in two conditions, one with the RF input to the demodulators terminated with 50 ohms to ground, and the other with the RFPD plugged in, but the PSL shutter closed (so the PD dark noise was the input to the demodulator). The LO input was driven at the nominal level for all measurements (2-3 dBm going in to the LO input, measured with the RF power meter, but I don't know what the level reaching the mixer is, because there is a complicated chain of ERA amplifiers and attenuators that determine what the level is). 

As in the previous elog, I have grouped the results into boards that do not (Attachment #1) and do (Attachment #2) have the low noise preamp installed. The top row is for the "Input terminated" measurements, while the bottom is with the RFPD plugged in, but dark. I think not a single board shows the "expected" noise performance for both I and Q channels. In the case where the preamp isn't installed and assuming the mixer is being driven with >17dBm LO, we expect the mixer to demodulate the Johnson noise of 50 ohms, which would be ~1nV/rtHz, and so with the SR785, we shouldn't measure anything in exceess of the instrument noise floor. With the low noise preamp installed, the expected output noise level is ~10nV/rtHz, which should just about be measurable (I didn't use any additional Low Noise front end preamp for these measurements). The AS55_I channel shows noise consistent with what was measured in 2017 after it was repaired, but the Q channel shows ~twice the noise. It seemed odd to me that the Q channels show consistently higher noise levels in general, but I confirmed that the SR785 channel 2 did not show elevated instrument noise at least when terminated with 50 ohms, so seems like a real thing.

While this is clearly not an ideal state of operation, I don't see how this can explain the odd PRMI sensing.

Quote:

For completeness, I will measure the input terminated I/F output noise levels later today. Note also that my characterization of the optical modulation profile did not reveal anything obviously wrong (to me at least). 

  15834   Tue Feb 23 00:10:05 2021 gautamUpdateGeneralDemod char part 1

I measured the conversion efficiencies for all the RFPD demod boards except the POP port ones. An RF source was used to drive the PD input on the demod board, one at a time, and the I/F outputs were monitored on a 300 MHz oscilloscope. The efficiency is measured as the percentage ratio V_IF / V_RF. 

I will upload the full report later, but basically, the numbers I measured today are within 10% of what I measured in 2017 when I previously did such a characterization. The orthogonality also seems fine. 

I believe I restored all the connections at 1Y2 correctly, and I can lock POX/POY and PRMI on 1f signals after my work. I will do the noise characterization tomorrow - but I think this test already rules out any funkiness with the demod setup (e.g. non orthogonality of the digitized "I" and "Q" signals). The whitening part of the analog chain remains untested.

Quote:

But I would still bet on demod chain funniness


Update 2/23 1215: I've broken up the results into the demod boards that do not (Attachment #1) and do (Attachment #2) have a D040179 preamp installed. Actually, the REFL11 AO path also has the preamp installed, but I forgot to capture the time domain data for those channels. The conversion efficiency inferred from the scope was ~5.23 V/V, which is in good agreement with what I measured a few years ago.

  • The scope traces were downloaded.
  • The resulting X/Y traces are fitted with ellipses to judge the gain imbalance and orthogonality.
  • The parameter phi is the rotation of the "bounding box" for the fitted ellipses - if the I and Q channels are exactly orthogonal, this should be either 0 or 90 degrees. There is significant deviation from these numbers for some of the demodulators, do we want to do something about this? Anyways, the REFL11 and AS55 boards, which are used for PRMI locking, report reasonable values. But REFL165 shows an ellipse with significant rotation. This is probably how the CDS phase rotator should be tuned, by fitting an ellipse to the digitized I/Q data and then making the bounding box rotation angle 0 by adjusting the "Measured Diff" parameter.
  • The gain imbalance seems okay across the board, better than 1dB.
  • The POX and POY traces are a bit weird, looks like there is some non-trivial amount of distortion from the expected pure sinusoid.
  • I measured the LO input levels going into each demod board - they all lie in the range 2-3dBm (measured with RF power meter), which is what is to be expected per the design doc. The exception the the 165 MHz LO line, which was 0.4 dBm. So this board probably needs some work. 
  • As I mentioned earlier, the conversion efficiencies are consistent with what I measured in 2017. I didn't break out the Eurocards using an extender and directly probe the LO levels at various points, but the fact that the conversion efficiencies have not degraded and the values are consistent with the insertion loss of various components in the chain make me believe the problem lies elsewhere. 

For completeness, I will measure the input terminated I/F output noise levels later today. Note also that my characterization of the optical modulation profile did not reveal anything obviously wrong (to me at least). 

  4736   Wed May 18 07:13:00 2011 SureshUpdateRF SystemDemod board measurements

I measured the amplitude and phase imbalances of the demod boards which have been modified.  This is just a basic health check.  We hope to use the script that Kiwamu is developing for a more accurate test.  The script can also use these measurements as a sanity check.  POP110  requires some further attention. 

Demod_Board_measurements.png

 

The RF distribution box outputs corresponding to the demod board (eg. AS55_LO --> AS55_demod) were used as LO sources.  The RF signal was generated with a Marconi and held a kHz away from the LO frequency.  The amplitude and phase unbalance were measured with SR785.  The RF Power meter was used to check the LO power in each case.

 

 

  4755   Fri May 20 05:41:22 2011 SureshUpdateRF SystemDemod board measurements

The POP110 board which had the large Amplitude and Phase unbalances was examined today.  It turned out that there was some stray solder which had connected the Sum port of the PSCQ-2-120 splitter to its body (ground).  After I removed that the amplitude unbalance was 0.3dB however the phase was 105deg.  The phase reduced to 90 deg only if the power on the splitter is around 19 dBm.  So removed the AT1 (10dB attenuator) and the phase unbalance dropped to 91 deg.  However this is not a sustainable solution as the ERA-5 max output is about 19.5 dBm.

As this is a side band power monitor (and not a length sensing RFPD), we can make do with a poorer phase.  I will therefore replace the AT1 and adjust the residual phase with cable delay lines.  

Quote:

I measured the amplitude and phase imbalances of the demod boards which have been modified.  This is just a basic health check.  We hope to use the script that Kiwamu is developing for a more accurate test.  The script can also use these measurements as a sanity check.  POP110  requires some further attention. 

Demod_Board_measurements.png

 

The RF distribution box outputs corresponding to the demod board (eg. AS55_LO --> AS55_demod) were used as LO sources.  The RF signal was generated with a Marconi and held a kHz away from the LO frequency.  The amplitude and phase unbalance were measured with SR785.  The RF Power meter was used to check the LO power in each case.

 

 

 

  11070   Wed Feb 25 20:00:39 2015 JenneUpdateLSCDelay line un-installed again - sensing matrix comparisons

I have measured the sensing matrix for the PRMI at the REFL photodiodes for both the nominal configuration and the 33MHz cancellation configuration.  The nominal configuration measurements do not compare well with those from November (http://nodus.ligo.caltech.edu:8080/40m/10701) which makes me unhappy no.  Both sets of nominal config reported below are from today, after tuning the demod phases and making sure the MICH and PRCL loops looked the same as yesterday (esp. overall gain).  The cancellation config data is from last night.

Note that the magnitude for each photodiode is referred to its own "PD counts".  Since the electronics are different for each PD, and that is not taken into account here, you cannot directly compare an element from one PD to an element from another PD.  What you can do (which is most of what we need right now) is compare all the different measurements for a single photodiode.

PRCL Sensing elements 33 MHz cancellation, REFL55 I&Q lock Nominal, REFL55I&Q lock Nominal, REFL33I&Q lock
  Mag [PD counts / m] Phase [deg] Mag [PD counts / m] Phase [deg] Mag[PD counts / m] Phase [deg]
REFL 11 3.0e13 0.05 4.2e13 0.1 2.9e13 0.1
REFL 33 1.1e11 3.4 1.8e12 0.25 1.2e12 0.3
REFL 55 1.8e10 2.1 1.4e11 4.8 9.3e10 4.1
REFL 165 7.0e9 15.5 6.3e11 0.5 4.2e11 0.9

 

MICH Sensing elements 33MHz cancellation, REFL55 I&Q lock Nominal, REFL55 I&Q lock Nominal, REFL33 I&Q lock
  Mag [PD counts / m] Phase [deg] Mag [PD counts / m] Phase [deg] Mag [PD counts / m] Phase [deg]
REFL 11 3.0e12 2.8 4.1e12 3.0 2.9e12 3.3
REFL 33 1.1e10 3.7 1.7e11 3.5 1.2e11 3.9
REFL 55 2.1e9 30.3 1.4e10 27.7 1.1e10 30.2
REFL165 6.4e8 24.4 6.3e10 21.0 4.5e10 23.0

So, what I'm apparently seeing is that the magnitudes of the sensing matrix signals that are made using 55MHz (i.e. everything but REFL11) change when we go into the cancellation configuration, but the phases of the sensing elements do not change significantly.  Also, I am apparently seeing that REFL11 and REFL33 only have about 3 degrees of separation between the MICH and PRCL signals no matter what configuration is used.  This doesn't make a lot of sense, since we know that we can lock robustly on REFL33I&Q (it's been sitting there happily as I write this elog), so it seems crazy that we could actually be so degenerate.  Also, at the bottom of the elog that I wrote in November 2014, I show a sensing matrix where both REFL11 and REFL33 have about 45 degrees of separation between the MICH and PRCL signals.

I don't think I'm doing anything too crazy here, particularly with the phase.  For a given PD and given DoF, I find the magnitude of the peaks of the I and Q signals, and just do atan2(Q-signal, I-signal)*180/pi, and those are the numbers that go in the phase columns above. 


Before taking my measurements, I tuned up the demod phases for the PRMI-only case.  I think REFL11 may have previously been tuned for CARM when the arms were held with ALS, but I don't really recall.  Anyhow, now all 4 REFL PDs are tuned for PRMI-only.

This was done while the PRMI was locked with REFL 55 I&Q. 

EDIT, 26Feb2015: Last night I mixed up the REFL11 and REFL33 new demod phases.  Bold are the corrected version.  Also, note that REFL33 was formerly tuned for PRCL in PRFPMI, which may be why it changed by ~10 degrees.

  Old demod phase [deg] New demod phase [deg]
REFL11 20 131.7 +/- 0.1   74.7 +/- 0.1
REFL33 142.2 74.7 +/- 0.1    131.7 +/- 0.1 
REFL55 19

15.3 +/- 0.3

REFL165 -172 -170.0 +/- 0.2

Here's the recipe for locking REFL 55 I&Q in the nominal modulation configuration.  It's the same as the REFL33 I&Q lock that I was using today, except that for the REFL33 version, the matrix elements are both unity.

PRMI REFL55 I&Q, nominal configuration MICH PRCL

Input Matrix

36*REFL55Q 12*REFL55I
Gain 5.0 -0.028
DoF trigger POP22I; 50 up, 0.5 down POP22I; 50 up, 0.5 down
FMs FM 4, 5 on FM 4, 5 on
FM trigger FM 2, 3, 6, 9; 50 up, 0.5 down, 5 second wait FM 1, 2, 6, 8, 9; 50 up, 0.5 down, 1 second wait
Normalization none none
Output matrix -1*ITMX, +1*ITMY 1*PRM

 

  11045   Tue Feb 17 19:49:51 2015 KojiUpdateLSCDelay line un-installed again

The modulation setting was reverted.
Demod phase for REFL11/33/55/165 and AS55 were reverted to the previous numbers too.

  11069   Wed Feb 25 14:51:13 2015 JenneUpdateLSCDelay line un-installed again

And now I've removed the delay line, and am in the process of reverting the demod phases, etc.

  11064   Wed Feb 25 04:59:47 2015 JenneUpdateLSCDelay line re-installed, measurements round 2

[Jenne, EricQ, Rana]

We spent this evening measuring and thinking about our 3f signals, and the effect of the modulation cancellation. 

I reinstalled the delay line box, and reduced the modulation depth of the 55MHz signal, so that we are in the state of modulation cancellation - there should be almost no power at 33MHz injected into the vacuum.  I carefully tuned the demod phase of the REFL 11, 33 and 55 MHz PDs, and locked the PRMI on REFL55 I&Q.  The signal in REFL 165 was very tiny, so as best as I could tell, the demod phase that Koji found last week was correct. 

Here is a little record of what the demodulation phases should be, for the "nominal" and "cancellation" configurations, so that we don't have to continually use the time machine.

  "Nominal" configuration 3f modulation cancellation
REFL 11 20 deg 76.0 deg (tuned to nearest 0.5 deg)
REFL 33 142.2 deg 120.3 deg (tuned to nearest 0.1 deg)
REFL 55 19 deg 173.0 deg (tuned to nearest 0.5 deg)
REFL 165 -172 deg 18 deg (same number as last week)
AS 55 -166.1 deg -111.1 deg (same number as last week)

Also, here is the locking recipe for REFL55 I&Q in the cancellation configuration:

PRMI, 3f cancellation, REFL55 I&Q MICH PRCL
Input Matrix acquire with -2*REFL55Q, then put matrix to -15*REFL55Q -12*REFL55I
Gain 3.0 -0.1
DoF trigger POP22I; 50 up, 0.5 down POP22I; 50 up, 0.5 down
FMs FM 4, 5 on FM 4, 5 on
FM trigger FM 1, 2, 3, 6, 9; 50 up, 0.5 down, 5 second wait FM 1, 2, 6, 9; 50 up, 0.5 down, 1 second wait
Normalization none none
Output matrix -1*ITMX, +1*ITMY 1*PRM

With this setup, we measured the sensing matrix.  MICH had an excitation at 370.123 Hz with 8,000 counts to -ITMX+ITMY (this is about 0.3nm for each ITM), and PRCL had an excitation at 404.123 Hz with 50 counts to PRM.  For tonight, here is a PDF of the peaks in DTT.  The data is saved in /users/Templates/LSC_error_spectra/SensMat_PRMI_24Feb2015.xml. 

SensMat_PRMI_24Feb2015.pdf


Rana has proposed to us an idea for why the REFL 33 signal should be dominated by the 55*22 contribution, rather than -11*22.  Eric is in the process of checking this out with a Mist model to see if it makes sense. Here's the gist:

Our Schnupp asymmetry is small (3.9cm, IIRC), so the transmission of the 11MHz signal out the dark port is small.  This means that the finesse of the PRC for 11MHz isn't so huge.  On the other hand, since 55MHz is a higher frequency, the transmission out the dark port is larger and is a closer match to the PRM transmission, so the finesse of the PRC for 55MHz is higher. 

Since the magnitudes of the fields at the reflection port are not changing significantly, our PDH signals are being created by the relative phase of something which is anti-resonant (ex. carrier or 22MHz for sideband lock) vs. something which is resonant (ex. 11MHz or 55MHz).  Since the finesse of the 55MHz signal is larger, the accumulated phase change is greater, so we expect a larger slope to our PDH signals that involve 55MHz as compared to those that use 11MHz.

If we are comparing the contributions between -11*22 and 55*22, they both include the anti-resonant 22MHz. So, the difference in the signal strengths comes directly from the difference in phase accumulation due to the variation in the dark port transmission.

EricQ had a thought, and while I have enough awake brain cells to report the thought, we're going to have to revisit it when more of our brains are awake.  In either case, the transmission through the dark port is small compared to the transmission of the ITMs, so why don't the ITMs dominate the finesse calculation, and thus are the 11MHz and 55MHz really getting that much of a difference in finesse?  To be checked out.

  11066   Wed Feb 25 12:16:27 2015 KojiUpdateLSCDelay line re-installed, measurements round 2

WHAT? WHAT? WHAT? It's obviously opposite.

If the reflectivity of the front mirror is fixed (=PRM reflectivity), the finesse increases when the reflectivity of the end
mirror (=Compond mirror reflectivity) increases. i.e. 11MHz has higher finesse, 55MHz has lower finesse.

{\cal F} = \frac{\pi \sqrt{r_{\rm PRM} r_{\rm COM}}}{1-r_{\rm PRM} r_{\rm COM}}

If the reflectivity of the front mirror is fixed, the amplitude gain of the cavity is higher when the reflectivity of the end mirror increases. i.e. 11MHz has higher gain, 55MHz has lower gain

g_{\rm PRM} = \frac{t_{\rm PRM}}{1-r_{\rm PRM} r_{\rm COM}}
 

Quote:

Our Schnupp asymmetry is small (3.9cm, IIRC), so the transmission of the 11MHz signal out the dark port is small.  This means that the finesse of the PRC for 11MHz isn't so huge.  On the other hand, since 55MHz is a higher frequency, the transmission out the dark port is larger and is a closer match to the PRM transmission, so the finesse of the PRC for 55MHz is higher. 

 

  11067   Wed Feb 25 14:18:28 2015 ranaUpdateLSCDelay line re-installed, measurements round 2

The Schnupp asymmetry is definitely not an important parameter (no need for Koji to explode). It only serves to give us a bigger Q phase signal slope, but is not significant for the I phase signals.

The main anomaly is that the REFL33 optical gain (W/m) seems to have been reduced so much by the phase and amplitude adjustment of the 55 MHz modulation signal. One guess is that the true 3f signal is being made not by the (2*f1 - (-f1)) beat, but by some higher order beat. In addition to the usual RF fields produced by the 2 modulations, we must consider that the sidebands on sidebands produce intermodulation fields just after the EOM and so the fields with which we interrogate the PRMI are more complicated.

Jenne's Optickle calculation today should show us what the sensing matrix contribution is from each pair of fields, so that we can have a sensing matrix signal budget.

 

  11062   Tue Feb 24 23:39:16 2015 JenneUpdateLSCDelay line re-installed again

About 5-10 minutes ago I just put in the modulation cancellation setup according to the recipe in http://nodus.ligo.caltech.edu:8080/40m/11032

  11059   Mon Feb 23 21:57:13 2015 KojiUpdateLSCDelay line installed again (experiment, round 1)

Last Wednesday we tried PRMI 3f modulation cancellation. Under the 3f modulation cancellation, the PRMI could not be locked
with REFL signals, and the PRCL signal was just barely sufficient to lock PRCL with help of AS55Q MICH.

- The PRCL signal level in REFL33 was reduced by factor of 20 compared with the conventional modulation setting.
=> The 3f modulation cancellation does not chage the level of 11/22MHz sidebands, it is expected that REFL33 signal
has no significant change of the signal level. But it does.  If we change the relative phase between the modulations
at 11 and 55MHz, the signal level is recovered by factor of 5. Therefore something related to 55MHz modulation
(55MHz x 22MHz, or 44MHz x 11MHz) was contributing more than -11MHzx22MHz.

- Under the 3f demodulation cancellation, MICH signal in the REFL ports were extremely weak and there was
no hope to use it for any feedback control.

- WIth the PRMI locking by REFL33->PRCL and AS55Q->MICH, the sensing matrix was measured. All of the REFL
ports however, showed extremely degenerate sensing matrix between MICH and PRCL.

This was enough confusing to us, and we didn't draw any useful information from these. Here are some ideas to
investigate what is happening in out optical and electrical system.

- One approach is to use as simple optical setup as possible to inspect our sensing systems. For example,
we may want to try PRX/PRY/XARM/YARM cavities to check the functions of the REFL diodes and qualitatively characterize
the sensing chain (Optical gain [W/m], noise level, demodulation phase) so that we can compare these with
an interferometer seinsing model.

- Another approach is to change the mdulation setting more freely and empirically try to find the characteristic
of our optical/electrical systems. e.g. change the relative modulation phase and/or 55MHz attenuation, and try to understand
how 11-22, 11-44, 22-55, 0-33 pairs are contributing the signal.

  11044   Tue Feb 17 16:44:04 2015 KojiUpdateLSCDelay line installed again

For tonight's experiment, I re-installed the delay line cable and changed the attenuation to 10dB for the 55MHz modulation.

I quickly locked the PLL and checked that the modulation is the ratio of the field strength between the worst (19ns) and best
case (28ns) is 31dB, that is ~35 times reduction.

  11235   Wed Apr 22 11:48:30 2015 manasaSummaryGeneralDelay line frequency discriminator for FOL error signal

Since the Frequency counters have not been a reliable error signal for FOL PID loop, we will put together an analog delay line frequency dicriminator as an alternative method to obtain the beat frequency.

The configuration will be similar to what was done in elog 4254 in the first place.

For a delay line frequency dicriminator, the output at the mixer is proportional to cos(\theta_{b}) where \theta_{b} = 2 \pi f_{b}L/v

L - cable length asymmetry, fb - beat frequency and v - velocity of light in the cable.

The linear output signal canbe obtained for  0< \theta_{b}<\pi

For our purpose in FOL, if we would like to measure beat frequency over a bandwidth of 200MHz, this would correspond to a cable length difference of 0.5 m (assuming the speed of light in the coaxial cable is ~ 2x108m/s.

  11236   Wed Apr 22 14:56:18 2015 manasaSummaryGeneralDelay line frequency discriminator for FOL error signal

[Koji, Manasa]

Since the bandwidth of the fiber PD is ~ 1GHz, we could design the frequency discriminator to have a wider bandwidth (~ 500MHz). The output from the frequency discriminator could then be used to define the range setting of the frequency counter for readout or may be even error signal to the PID loop.

A test run for the analog DFD with cable length difference of 27cm gave a linear output signal with zero-crossing at ~206MHz.

Detailed schematic of the setup and plot (voltage vs frequency) will be updated shortly.

  11270   Mon May 4 10:21:09 2015 manasaSummaryGeneralDelay line frequency discriminator for FOL error signal

Attached is the schematic of the analog DFD and the plot showing the zero-crossing for a delay line length of 27cm. The bandwidth for the linear output signal obtained roughly matches what is expected from the length difference (370MHz) .

We could use a smaller cable to further increase our bandwidth. I propose we use this analog DFD to determine the range at which the frequency counter needs to be set and then use the frequency counter readout as the error signal for FOL.

 

  11272   Mon May 4 12:42:34 2015 manasaSummaryGeneralDelay line frequency discriminator for FOL error signal

Koji suggested that I make a cosine fit for the curve instead of a linear fit.

I fit the data to V(f) = A + B cos(2\pi f_{b}L/v) 
where L - cable length asymmetry (27 cm) , fb - beat frequency and v - velocity of light in the cable (2*10m/s)

The plot with the cosine fit is attached. 

Fit coefficients (with 95% confidence bounds):
       A =      0.4177  (0.3763, 0.4591)
       B =       2.941  (2.89, 2.992)

  11300   Mon May 18 14:46:20 2015 manasaSummaryGeneralDelay line frequency discriminator for FOL error signal

Measuring the voltage noise and frequency response of the Analog Delay-line Frequency Discriminator (DFD)

The schematic and an actual photo of the setup is shown below. The setup was checked to be physically sturdy with no loose connections or moving parts.

The voltage noise at the output of the DFD was measured using an SR785 signal analyzer while simultaneously monitoring the signal on an oscilloscope.

The noise at the output of the DFD was measured for no RF input and at several RF input frequencies including the zero crossing frequency and the optimum operating frequency of the DFD (20MHz).

The plot below show the voltage noise for different RF inputs to the DFD. It can be seen that the noise level is slightly lower at the zero crossing frequency where the amplitude noise is eliminated by the DFD.

I also did measurements to obtain the frequency response of the setup as the cable length difference has changed from the prior setup. The cable length difference is 21cm and the obtained linear signal at the output of the DFD extends over ~ 380MHz which is good enough for our purposes in FOL. A cosine fit to the data was done as before. //edit- Manasa: The gain of SR560 was set to 20 to obtain the data shown below//

Fit Coefficients (with 95% confidence bounds):
       a =     -0.8763  (-1.076, -0.6763)
       b =       3.771  (3.441, 4.102)

Data and matlab scripts are zipped and attached.

  15927   Wed Mar 17 00:05:26 2021 gautamUpdateLSCDelay line BIO remote control

While Koji is working on the REFL 11 demod board, I took the opportunity to investigate the non-remote-controllability of the delay line in 1Y2, since the TTs have already been disturbed. Here is what I found today.

  1. First, I brought over the spare delay line from the rack Chiara sits in over to 1Y2. 
    • Connected a Marconi to the input, monitored a -3dB pickoff and the delay line output simultaneously on a 300MHz scope.
    • With the front panel selector set to "Internal", verified that local (i.e. toggling front panel switches) switchability seems to work 👍 
    • Set the front panel switch to "External", and connected the D25 cable from the BIO card in 1Y3 to the back panel of the delay line unit - found that I could not remotely change the delay 😒 
    • I thought it'd be too much of a coincidence if both delay lines have the same failure mode for the remote switching part only, so I decided to investigate further up the signal chain.
  2. BIO switching - the CDS BIO bit status MEDM screen seems to respond, indicating that the bits are getting set correctly in software at least. I don't know of any other software indicator for this functionality further down the signal processing chain. So it would seem the BIO card is not actually switching.
  3. The Contec DO cards don't actually source the voltage - they just provide a path for current to flow (or isolate said path). I checked that pin 12 of the rear panel D25 connector is at +5 V DC relative to ground as indicated in the schematic (see P1 connector - this connector isn't a Dsub, it is IDE24, so the mapping to the Dsub pins isn't one-to-one, but pin 23 on the former corresponds to pin 12 on the latter), suggesting that the pull up resistors have the necessary voltage applied to them.
  4. Made a little LED tester breakout board, and saw no swtiching when I toggled the status of some random bits.
  5. Noted that the bench power supply powering this setup (hacky arrangement from 2015 that never got unhacked) shows a current draw of 0A.
    • I am not sure what the quiescent draw of these boards is - the datasheet says "Power consumption: 3.3VDC, 450mA", but the recommended supply voltage is "12-24V DC (+/-10%)" not 3.3VDC, so not sure what to make of that.
    • To try and get some insight, I took one of the new Contec-32L-PE cards we got from near Jon's CDS test stand (I've labelled the one I took lest there be some fault with it in the future), and connected it to a bench supply (pin 18 = +15V DC, pin1 = GND). But in this condition, the bench supply reports 0A current draw.
  6. Ruled out the wrong cable being plugged in - I traced the cable over the cable tray, and seems like it is in fact connecting the BIO card in the c1lsc expansion chassis to the delay line.

So it would seem something is not quite right with this BIO card. The c1lsc expansion chassis, in which this card sits, is notoriously finicky, and this delay line isn't very high priority, so I am deferring more invasive investigation to the next time the system crashes.

* I forgot we have the nice PCB Contec tester board with LEDs - the only downside is that this board has D37 connectors on both ends whereas the delay line wants a D25, necessitating some custom ribbon cable action. But maybe there is a way to use this.

As part of this work, I was in various sensitive areas (1Y3, chiara rack, FE test stand etc) but as far as I can tell, all systems are nominal.

  15917   Fri Mar 12 19:44:31 2021 gautamUpdateLSCDelay line

I may want to use the delay line phase shifter in 1Y2 to allow remote actuation of the REFL11 demod phase (for the AO path, not the low bandwidth one). I had this working last Feb, but today, I am unable to remotely change the delay. @Koji, it would be great if you could fix this the next time you are in the lab - I bet it's a busted latch IC or some such thing. I did the non-invasive tests - cable is connected, control bits are changing (at least according to the CDS BIO indicators) and the switch controlling remote/local switching is set correctly. The local switching works just fine.

In the meantime, I will keep trying - I am unconvinced we really need the delay line.

  6625   Tue May 8 16:43:15 2012 JenneUpdateCDSDegenerate channels, potentially a big mess

Rana theorized that we're having problems with the MC error signal in the OAF model (separate elog by Den to follow) because we've named a channel "C1:IOO-MC_F", and such a channel already used to exist.  So, Rana and I went out to do some brief cable tracing.

MC Servo Board has 3 outputs that are interesting:  "DAQ OUT" which is a 4-pin LEMO, "SERVO OUT" which is a 2-pin LEMO, and "OUT1", which is a BNC->2pin LEMO right now.

DAQ OUT should have the actal MC_F signal, which goes through to the laser's PZT.  This is the signal that we want to be using for the OAF model.

SERVO OUT should be a copy of this actual MC_F signal going to the laser's PZT.  This is also acceptable for use with the OAF model.

OUT1 is a monitor of the slow(er) MC_L signal, which used to be fed back to the MC2 suspension.  We want to keep this naming convention, in case we ever decide to go back and feed back to the suspensions for freq. stabilization.

Right now, OUT1 is going to the first channel of ADC0 on c1ioo.  SERVOout is going to the 7th channel on ADC0.  DAQout is going to the ~12th channel of ADC1 on c1ioo.  OUT1 and SERVOout both go to the 2-pin LEMO whitening board, which goes to some new aLIGO-style ADC breakout boards with ribbon cables, which then goes to ADC0.  DAQout goes to the 4pin LEMO ADC breakout, (J7 connector) which then directly goes to ADC1 on c1ioo.

So, to sum up, OUT1 should be "adc0_0" in the simulink model, SERVOout should be "adc0_6" on the simulink model, and DAQout should be "adc1_12" (or something....I always get mixed up with the channel counting on 4pin ADC breakout / AA boards). 

In the current simulink setup, OUT1 (adc0_0) is given the channel name C1:IOO-MC_F, and is fed to the OAF model.  We need to change it to C1:IOO-MC_L to be consistent with the old regime.

In the current simulink setup, SERVOout (adc0_6) is given the channel name C1:IOO-MC_SERVO.  It should be called C1:IOO-MC_F, and should go to the OAF model.

In the current simulink setup,DAQout (~adc1_12) doesn't go anywhere.  It's completely not in the system.  Since the cable in the back of this AA / ADC breakout board box goes directly to the c1ioo I/O chassis, I don't think we have a degenerate MC_F naming situation.  We've incorrectly labeled MC_L as MC_F, but we don't currently have 2 signals both called MC_F.

Okay, that doesn't explain precisely why we see funny business with the OAF model's version of MCL, but I think it goes in the direction of ruling out a degenerate MC_F name.

Problem:  If you look at the screen cap, both simulink models are running on the same computer (c1ioo), so when they both refer to ADC0, they're really referring to the same physical card.  Both of these models have adc0_6 defined, but they're defined as completely different things.  Since we can trace / see the cable going from the MC Servo Board to the whitening card, I think the MC_SERVO definition is correct.  Which means that this Green_PH_ADC is not really what it claims to be.  I'm not sure what this channel is used for, but I think we should be very cautious and look into this before doing any more green locking.  It would be dumb to fail because we're using the wrong signals.

 

  7759   Wed Nov 28 23:18:35 2012 CharlesUpdatePEMDecreased RMS in Seismometers

The attached plots display RMS noise from various accelerometers and seismometers over the past 90 days. One can see how after the reinstallation of the seismometers in November, RMS from the GUR1Z and GUR1X channels decreases by a factor of about 100 from data in August. Additionally, the RMS over the course of the last 90 days has notably decreased in all instruments. In many cases, the RMS is only the result of inherent electronics noise, rather than from a signal.

  7762   Thu Nov 29 02:43:48 2012 AyakaUpdatePEMDecreased RMS in Seismometers

Quote:

The attached plots display RMS noise from various accelerometers and seismometers over the past 90 days. One can see how after the reinstallation of the seismometers in November, RMS from the GUR1Z and GUR1X channels decreases by a factor of about 100 from data in August. Additionally, the RMS over the course of the last 90 days has notably decreased in all instruments. In many cases, the RMS is only the result of inherent electronics noise, rather than from a signal.

The Image is replaced

[Den, Ayaka]

We found that seismometer was working and the calibration in the filter banks should have been wrong.
We turned off the all FM2 filter in RMS filter banks.

We also installed STS seismometer. It is under the BS. Now we have spectrum of three seismometers.
GUR1Xfilterbank.pngseismometers1129.pdf


 

RA: the above plot is kind of unreadable and useless. Please replace with something legible and put in some words about why there is a wrong filter, what exactly it is, etc., etc. etc. And why would you leave in a filter which is not supposed to be on? We might as well leave a few secretly broken chairs in the control room...

  7763   Thu Nov 29 09:58:06 2012 DenUpdatePEMDecreased RMS in Seismometers

Quote:

[Den, Ayaka]

We found that seismometer was working and the calibration in the filter banks should have been wrong.
We turned off the all FM2 filter in RMS filter banks.
 

We also installed STS seismometer. It is under the BS. Now we have spectrum of three seismometers;
 

RA: the above plot is kind of unreadable and useless. Please replace with something legible and put in some words about why there is a wrong filter, what exactly it is, etc., etc. etc. And why would you leave in a filter which is not supposed to be on? We might as well leave a few secretly broken chairs in the control room...

 First of all, STS-2 is in the end of X arm, GUR2 is under BS, GUR1 is in the end of Y arm.

BLRMS were small because we applied calibration from counts to um/s two times. In the past we had calibration in the RMS BP filter bank (vel2vel = FM2). Now we have calibration in the seismometer input filter bank so we can save calibrated _OUT channels.

  12435   Tue Aug 23 22:58:16 2016 KojiUpdateElectronicsDecoupling capacitor 101

What I suggested was:
- For most cases, power decoupling capacitors for the regulators should be ~100nF "high-K ceramic capacitors" + 47uF~100uF "electrolytic capacitors".
- For opamps, 100nF high-K ceramic should be fine, but you should consult with datasheets.
- Usually, you don't need to use tantalum capacitors for this purpose unless specified.
- Don't use film capacitors for power decoupling.

79XXs are less stable compared to 78XXs, and tend to become unstable depending on the load capacitance.
One should consult with the datasheet of each chip in order to know the proper capacitors values.
But also, you may need to tweak the capacitor value when necessary. Above recipe works most of the case.

  12437   Wed Aug 24 14:44:33 2016 PrafulUpdateElectronicsDecoupling capacitor 101

Do these look good for the ceramic capacitors? We're running low.

http://www.mouser.com/ProductDetail/Vishay-BC-Components/K104K15X7RF53L2/?qs=sGAEpiMZZMuMW9TJLBQkXmrXPxxCV7CRo6C15yUYAos%3d

Quote:

What I suggested was:
- For most cases, power decoupling capacitors for the regulators should be ~100nF "high-K ceramic capacitors" + 47uF~100uF "electrolytic capacitors".
- For opamps, 100nF high-K ceramic should be fine, but you should consult with datasheets.
- Usually, you don't need to use tantalum capacitors for this purpose unless specified.
- Don't use film capacitors for power decoupling.

79XXs are less stable compared to 78XXs, and tend to become unstable depending on the load capacitance.
One should consult with the datasheet of each chip in order to know the proper capacitors values.
But also, you may need to tweak the capacitor value when necessary. Above recipe works most of the case.

 

  12438   Wed Aug 24 19:37:55 2016 KojiUpdateElectronicsDecoupling capacitor 101

Yes

Interesting articles how they should only be used for power decoupling and not in the signal path.

http://www.edn.com/design/analog/4416466/Signal-distortion-from-high-K-ceramic-capacitors

http://www.edn.com/design/analog/4426318/More-about-understanding-the-distortion-mechanism-of-high-K-MLCCs

  12703   Wed Jan 11 19:20:23 2017 Max IsiUpdateSummary PagesDecember outage

The summary pages were not successfully generated for a long period of time at the end of 2016 due to syntax errors in the PEM and Weather configuration files.

These errors caused the INI parser to crash and brought down the whole gwsumm system. It seems that changes in the configuration of the Condor daemon at the CIT clusters may have made our infrastructure less robust against these kinds of problems (which would explain why there wasn't a better error message/alert), but this requires further investigation.

In any case, the solution was as simple as correcting the typos in the config side (on the nodus side) and restarting the cron jobs (on the cluster side, by doing `condor_rm 40m && condor_submit DetectorChar/condor/gw_daily_summary.sub`). Producing pages for the missing days will take some time (how to do so for a particular day is explained in the wiki https://wiki-40m.ligo.caltech.edu/DailySummaryHelp).

RXA: later, Max sent us this secret note:

However, I realize it might not be clear from the page which are the key steps. These are just running:

1) ./DetectorChar/bin/gw_daily_summary --day YYYYMMDD --file-tag some_custom_tag To create pages for day YYYYMMDD (the file-tag option is not strictly necessary but will prevent conflict with other instances of the code running simultaneously).

2) sync those days back to nodus by doing, eg: ./DetectorChar/bin/pushnodus 20160701 20160702

This must all be done from the cluster using the 40m shared account.
  12709   Thu Jan 12 23:22:34 2017 ranaUpdateSummary PagesDecember outage

Pages still not working: PEM and MEDM blank.

  • Committed existing MEDM grabbing scripts to SVN. Ran the cron job on megatron by hand. It grabs PNG files, but somehow its not getting into the summary pages.
  • Changed the MEDM grabbing scripts to use '/usr/bin/env'.
  • GW summary log files were numbering in the many thousands, so I moved everything over 320 days old into the OLD/ sub-directory using 'find . -type f -mtime +320 -exec mv {} OLD/ \;' (the semi-colon is needed)
  • Did apt-get upgrade on Megatron.
  • pinged Max
  • Stared at GWsumm docs to see if there's a clue about what (if anything) is wrong with the .ini file.
  12713   Fri Jan 13 14:33:00 2017 MAX (not Rana)UpdateSummary PagesDecember outage

PEM config file was also lacking a section named "summary", which is necessary for all parent tabs; this has now been solved. I have deactivated the MEDM pages because Praful's screencap script seemed to be broken (I should have logged this, I apologize).

Quote:

Pages still not working: PEM and MEDM blank.

  • Committed existing MEDM grabbing scripts to SVN. Ran the cron job on megatron by hand. It grabs PNG files, but somehow its not getting into the summary pages.
  • Changed the MEDM grabbing scripts to use '/usr/bin/env'.
  • GW summary log files were numbering in the many thousands, so I moved everything over 320 days old into the OLD/ sub-directory using 'find . -type f -mtime +320 -exec mv {} OLD/ \;' (the semi-colon is needed)
  • Did apt-get upgrade on Megatron.
  • pinged Max
  • Stared at GWsumm docs to see if there's a clue about what (if anything) is wrong with the .ini file.

 

  10771   Tue Dec 9 16:07:16 2014 manasaSummaryGeneralDec 9 - FC module and fiber chassis

Quote:

Quote:

Attached is the timeline for Frequency Offset Locking related activities. All activities will be done mostly in morning and early afternoon hours.

Elaborate to do list:

1. The FC module should be mounted on the IOO rack. Domenica has to be powered up appropriately to the rack power supply.

2. The fiber chassis needs to be built. This will hold all the fiber components and will sit inside the PSL enclosure.
Fiber connectors and fiber couplers need to be installed in the chassis. Attached is the cartoon sketch of layout in the chassis.

3. User guide for FC module (work in progress)

1. FC module has been mounted on the IOO rack. The module gets it AC supply from the powerstrip already installed on the back side of the rack.

FCmodule.png

2. The fiber chassis has not been put together completely. We have still not received the front and back panels for the chassis; which is keeping me on hold. Diego is almost done with his housekeeping on Domenica. He will post an elog with all the details.

3. User guide for FC module (work in progress)

  10767   Tue Dec 9 00:30:27 2014 manasaSummaryGeneralDec 9 - Elaborate to do list

Quote:

Attached is the timeline for Frequency Offset Locking related activities. All activities will be done mostly in morning and early afternoon hours.

Elaborate to do list:

1. The FC module should be mounted on the IOO rack. Domenica has to be powered up appropriately to the rack power supply.

2. The fiber chassis needs to be built. This will hold all the fiber components and will sit inside the PSL enclosure.
Fiber connectors and fiber couplers need to be installed in the chassis. Attached is the cartoon sketch of layout in the chassis.

3. User guide for FC module (work in progress)

  10765   Mon Dec 8 15:54:39 2014 manasaSummaryGeneralDec 8 - Check Frequency Counter module

Quote:

Attached is the timeline for Frequency Offset Locking related activities. All activities will be done mostly in morning and early afternoon hours.

[Diego, Manasa]

We looked into the configuration and settings that the frequency counters (FC) and Domenica (the R pi to which the FCs talk to) were left at . After poking around for a few hours, we were able to readout the FC output and see it on StripTool as well.

We have made a list of modifications that should be done on Domenica and to the readout scripts to make the FC module automated and user-friendly.

I will prepare a user manual that will go on the wiki once these changes are made.

 

  10766   Mon Dec 8 20:53:51 2014 diegoSummaryGeneralDec 8 - Check Frequency Counter module

Quote:

Quote:

Attached is the timeline for Frequency Offset Locking related activities. All activities will be done mostly in morning and early afternoon hours.

[Diego, Manasa]

We looked into the configuration and settings that the frequency counters (FC) and Domenica (the R pi to which the FCs talk to) were left at . After poking around for a few hours, we were able to readout the FC output and see it on StripTool as well.

We have made a list of modifications that should be done on Domenica and to the readout scripts to make the FC module automated and user-friendly.

I will prepare a user manual that will go on the wiki once these changes are made.

 

 OUTDATED: see elog 10779

 

I started working on the scripts/FOL directory (I did a backup before tampering around!):

  • I still need to make some serious polishing in the folder, and into the Raspberry Pi itself, in order to have a clean and understandable environment;
  • as of now, I created an single armFC.c program, which takes as arguments the device (/dev/hidraw0 for the X arm, and /dev/hidraw1 for the Y arm) and the value to write into the frequency counter (0x3 for initialization and 0x2 for actual use); hence, no more need for recompilation!
  • I improved the codetorun.py script (and gave the fellow a proper name, epics_channels.py) which handles the initialization AND the availability of the channels;
  • On the Raspberry Pi, I created two init scripts, /etc/init.d/epics_server.sh and /etc/init.d/epics_channels.sh, which start at the end of the boot process with default runlevels; the former starts the softIOc process (epics itself), while the latter executes the constantly running epics_channels.py script; as they are services, they can be started/stopped with the usual sudo /etc/init.d/NAME start|stop|restart

 

As a result, as soon as the Raspberry Pi completes its boot process, the two beatnote channels are immediately available.

 

ELOG V3.1.3-