40m QIL Cryo_Lab CTN SUS_Lab TCS_Lab OMC_Lab CRIME_Lab FEA ENG_Labs OptContFac Mariner WBEEShop
  40m Log, Page 12 of 341  Not logged in ELOG logo
ID Date Author Typedown Category Subject
  1184   Sun Dec 7 16:12:53 2008 ranaUpdateDAQbooted awg
because it was red
  1187   Mon Dec 8 11:54:27 2008 YoichiUpdateGeneralIFO mirrors aligned
This morning, I re-aligned the IFO mirrors to see if they were badly moved by the earthquake.
The both arms locked just by the restoring scripts, but the transmission was about 0.7. So I aligned them
with the dithering scripts.
To lock the PRMI, I had to manually tweak the PRM alignment. After running the dithering script, the SPOB
went up to 1200.
I also had to tweak the SRM to get the DRMI locked. After the dithering script, the SPOB was 4200 and REFL166Q
was 3000.
  1188   Mon Dec 8 17:50:21 2008 YoichiUpdateSUSITMY drift
The suspension drift monitor shows that the ITMY alignment was shifted after the earthquake.
Looks like only the UL sensor had a step at the earthquake (see the attachment 1).
So it is probably an electronics problem.
I pushed in the cable between the rack and the ITMY satellite amplifier, but no change observed.
Actually, the ITMY-UL sensor looks like it has been dead before the earthquake.
The second attachment shows a long-term trend of the UL sensor.
The sensor output had been around zero since Nov. 17th.
When I disabled the output of the UL sensor, the sus-drift-mon fields turned green.
So I think the drift-mon's reference values are wrong, and currently the ITMY is in a good alignment.

I also attached the free-swing measurements of the ITMY taken on Aug. 18th and today.
There is no notable change in the resonant frequencies.
Attachment 1: ITMY-OSEMs.png
ITMY-OSEMs.png
Attachment 2: ITMY-UL.png
ITMY-UL.png
Attachment 3: ITMY-08-18.pdf
ITMY-08-18.pdf
Attachment 4: ITMY-12-08.pdf
ITMY-12-08.pdf
  1190   Fri Dec 12 22:51:23 2008 YoichiUpdatePSLReference cavity ring down measurement again
Bob made new HV-cables with HV compatible coaxes. The coax cable is rated for 2kV, which was as high as Bob
could found. I used it with 3kV hoping it was ok.
I also put a series resistor to the pockels cell to tame down the ripples I saw in elog:1136.

Despite those efforts, I still observed large ringings.
I tried several resistor values (2.5k, 1k, 330ohm), and found that 330ohm gives a slightly better result.
(When the resistance is larger, the edge of the PBS Refl. becomes dull).
Since the shape of the ringing does not change at all even when the pulse voltage is lowered to less than 1kV,
I'm now suspicious of the DEI pulser.

Anyway, I estimated the cavity pole using the MATLAB's system identification toolbox again.
This time, I locked the reference cavity using only the PZT feedback, which makes the UGF about a few kHz.
So, within the time scale shown in the plot below, the servo does not have enough time to respond, thus the laser
frequency stays tuned with the cavity. This was necessary to avoid non-linear behavior of the transmitted power
caused by the servo disturbing the laser frequency. With this treatment, I was able to approximate the response of
the cavity with a simple linear model (one pole low-pass filter).

MATLAB estimated the cavity pole to be 47.5kHz.
The blue curve in the plot is the measured RC transmitted power.
The incident power to the cavity can be inferred from the inverse of the red curve (the PBS reflection power).
The brown curve is the response of the first order low-pass filter with fc=47.5kHz to the input power variation.
The blue and brown curves match well for the first 10usec. Even after that the phases match well.
So the estimated 47.5kHz is probably a reasonable number. I don't know yet how to estimate the error of this measurement.

According to http://www.ligo.caltech.edu/~ajw/PSLFRC.png the designed transmission of the reference cavity mirrors is 300ppm (i.e.
the round trip loss (RTL) is 600ppm).
This number yields fc=35kHz. In the same picture, it was stated that fc=38.74kHz (I guess this is a measured number at some point).
The current fc=47.5kHz means, the RTL has increased by 200ppm from the design and 150ppm from the time fc=38.74kHz was measured.
Attachment 1: RC-Ringdown.png
RC-Ringdown.png
  1191   Tue Dec 16 19:06:01 2008 YoichiUpdatePSLReference cavity ring down repeated many times
Today, I repeated the reference cavity ring down measurement many times to see how much the results vary.

I repeated the ring down for 20 times and the first attachment shows the comparison of the measured and estimated cavity transmission power.
The blue curve is the measured one, and the red curve is the estimated one. There are only 10 plots because I made a mistake when transferring data
from the oscilloscope to the PC, and one measurement data was lost.

The second attachment shows the histogram of the histogram of the estimated cavity pole frequencies.
I admit that there are not enough samples to treat it statistically.
Anyway, the mean and the standard deviation of the estimated frequencies are 47.6kHz and 2.4kHz.
Assuming a Gaussian distribution and zero systematic error, both of which are bold assumptions though, the result is 47.6(+/-0.6)kHz.

Now I removed the Pockels Cell from the RC input beam path.
I maximized the transmission by tweaking the steering mirrors and rotating the HWP.
Since the transmission PD was saturated without an ND filter on it, I reduced the VCO RF power slider to 2.85.
Accordingly, I changed the nominal common gain of the FSS servo to 10.5dB.
Attachment 1: RC_Ringdown_Estimates.png
RC_Ringdown_Estimates.png
Attachment 2: Cavity_Pole_Histogram.png
Cavity_Pole_Histogram.png
  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.
  1197   Fri Dec 19 16:38:09 2008 steveUpdateLSCall optlevs centered
All optlevs were centered after full alignment.

Qpd sums are:
ETMX 12,229 counts
ITMX 9,932
ETMY 12,043
ITMY 4,362
BS 1,880
PRM 1,423
SRM 11,641
  1199   Sun Dec 21 13:00:06 2008 steveUpdateall down cond.vac and laser back on
There was a power outage sometimes early Saturday.

All things are down.

The Maglev was started and reached normal operating condition.
V1 was opened at P1 3.8 mTorr and cc1 is back to 3e-6 Torr now

The MOPA was turned on.
Ion pump HV on to FSS cavity.
  1200   Sun Dec 21 14:18:04 2008 YoichiUpdateComputersRFM network bypass box's power supply is dead
I restarted the front-end computers by power cycling them one-by-one.
After issuing startup commands, most of them started normally at least by looking
at the output from telnet/ssh.
However, the status monitors of the FE computers on the EPICS screen are still red.
I noticed that all the LEDs on the VMIC 5594 RFM network bypass box are off.
According to the labels, fb40m, c0daqctrl, c0dcu are connected to the box.
This means (I believe) c1dcuepics cannot access the RFM network. So we have no control over
the FE computers through EPICS.

I pushed the reset button on the box, power cycled it, but nothing changed.
I checked the fuse and it was OK. Then I found that the power supply was dead.
It is a small AC adapter supplying +5VDC with a 5-pin DIN like connector.
We have to find a replacement.
  1201   Mon Dec 22 13:48:22 2008 YoichiUpdateComputersRFM network bypass box's power supply is dead
As a temporary fix, I cut the cable of the power supply and connected it to the Sorensen power supply +5V on the rack.
Now, the RFM bypass box is powered up, but some LEDs are red, which looks like a bad sign.
I restarted all the FE computers, but this time I got errors during the execution of the startup commands in the VxWorks machines.
The errors are "General Protection Fault" or "Invalid Opcode".
The linux machines do not show errors but still the status lights in EPICS are red.
We need Alex's help. He did not answer the phone, so Alberto left a voice mail.
  1202   Tue Dec 23 10:35:40 2008 YoichiUpdateComputersRFM network breakdown mostly fixed
Rana, Rolf, Alberto, Yoichi

The source of the problem was the RFM bypass box, as expected.
Rana pointed out that the long cable I used to bring the 5V from the Sorensen to the box
may cause a large voltage drop considering that the box is sucking ~3A.
So we connected the cable to another power supply (5V/5A linear power supply).
Then the LEDs on the bypass box turned green from red, and everything started to work.

A weired thing is that when I connected the cable to the wrong terminals of the power supply which
have lower current supply capabilities, the supply voltage dropped to 3V, but still the LEDs on the bypass box
turned green. This means the bypass box can live with 3V.
I noticed that there is a long cable from the Sorensen to the cross connect on the side of the rack, where I
connected my cable to the bypass box. This long cable had somewhat large resistance (1 or 2 Ohms) and dropped
the supply voltage to less than 3V ?
Anyway, the bypass box is now on a temporary power supply. Alberto was assigned a task to find a replacement power
supply.

There are two remaining problems.
c1susvme1 fails to start often claiming a DMA error on a Pentek. After several attempts, you can start the machine,
but after a while (1 hour ?) it fails again.
op340m is not responding to ssh login. It responds to ping.
We hooked up a monitor and keyboard (USB because the machine does not have a PS/2 port) to it and rebooted.
At the boot, it briefly displays a message "No keyboard, try TTYa", but after that no display signal.
Steve found me a serial cable. I will try to login to the machine using the serial port.

  1203   Wed Dec 24 10:33:24 2008 YoichiUpdateComputersSeveral fixes. Test point problem remains.
Yesterday, I fixed several remaining problems from the power failure.

I found a LEMO cable connecting the timing board to the Penteks was lose on the c1susvme1 crate.
After I pushed it in, the DMA error has not occured on c1susvme1.

I logged into op340m using a Null Modem Cable.
The computer was failing to boot because there were un-recoverable disk errors by the automatic fsck.
I run fsck manually and corrected some errors. After that, op340m booted normally and now it is working fine.
Here is the serial communication parameters I used to communicate with op340m:
>kermit      (I used kermit command for serial communication.)
>set modem type none
>set line /dev/ttyS0     (ttyS0 should be the device name of your serial port)
>set speed 9600
>set parity none
>set stop-bits 1
>set flow-control none
>connect

After fixing op340m, the MC locked.
Then I reset the HV amps. for the steering PZTs.
Somehow, the PZT1 PIT did not work. But after moving the slider back and forth several times, it started to work.

I reset the mechanical shutters around the lab.

I went ahead to align the mirrors. The X-arm locked but the alignment script did not improve
the arm power.
I found that test points are not available. (diag said test point management not available).
Looks like test point manager is not running. Called Rolf, but could not reach him.
I'm not even sure on which machine, the tp manager is supposed to be run.
Is it c0daqawg ?
  1204   Wed Dec 24 12:46:54 2008 YoichiUpdateComputersTest points are back
Rob told me how to restart the test point manager.
It runs on fb40m and actually there is an instruction on how to do that in the Wiki.
http://lhocds.ligo-wa.caltech.edu:8000/40m/Computer_Restart_Procedures#fb40m

I couldn't find the page because when I put a keyword in the search box on the upper right
corner of the Wiki page and hit "enter", it only searches for titles. To do a full text
search, you have to click on the "Text" button.

Anyway, now the test points are back.
  1205   Mon Dec 29 18:01:07 2008 AidanUpdateAuxiliary lockingUpdated 40m Upgrade Document T080074-00-R

Added a paragraph to the 40m Upgrade document describing the fiber stabilization and frequency doubling proposed for auxiliary locking.

Also added a complete diagram of the fiber stabilization and a draft sketch of the frequency doubling.

Uploaded to https://nodus.ligo.caltech.edu:30889/svn/trunk/docs/upgrade08/ via svn.
  1206   Mon Dec 29 21:38:57 2008 YoichiUpdateComputersSnapshots of MEDM screens
I wrote scripts to take snapshots of MEDM screens in the background.
These scripts work even on a computer without a physical display attached.
You don't need to have X running.
So now the scripts run on nodus every 5 minutes from cron.
The screen shots are saved in /cvs/cds/caltech/statScreen/images/

There is a wiki page for the scripts.
http://lhocds.ligo-wa.caltech.edu:8000/40m/captureScreen.sh

Someone has to make a nice web page summarizing the captured images.
  1210   Thu Jan 1 00:55:39 2009 YoichiUpdateASCAlignment scripts for Linux
A Happy New Year.

The dither alignment scripts did not run on linux machines because tdscntr and ezcademod do not run
on linux. Tobin wrote a perl version of tdscntr and I modified it for 40m some time ago.
Today, I wrote a perl version of ezcademod. The script is called ditherServo.pl and resides in /cvs/cds/caltech/scripts/general/.
It is not meant to be a drop-in replacement, so the command line syntax is different. Usage is explained in the comment of the script.

Using those two scripts, I wrote linux versions of the alignment scripts.
Now when you call, for example, alignX script, it calls alignX.linux or alignX.solaris depending on the OS of
your machine. alignX.solaris is the original script using the compiled ezcademod.
In principle, ezcademod is faster than my ditherServo.pl because my script suffers from the overhead of
calling tdsdmd on each iteration of the servo. But in practice ditherServo.pl is not that bad. At least, as far as
the alignment is concerned, the performances of the both commands are comparable in terms of the final arm power and the convergence.

Now the alignXXX commands from the IFO Configure MEDM screen work for X-arm, Y-arm, PRM and DRM. I did not write a script for Michelson, since
it is optional.
I confirmed that "Align Full IFO" works correctly.
  1214   Fri Jan 2 18:49:54 2009 YoichiUpdateLSCLSC modulation frequencies adjusted
I noticed that the IFO did not lock in the MICH configuration.
This was because AS166Q signal was too small.
The demodulation phase seemed not right, i.e. the I-phase signal was larger than Q.
I suspected that the 166MHz modulation frequency was not exactly on the MC FSR, since I just
recovered the number written on the Marconi after the power failure.
I measured the optimal frequency by the method explained in elog:752.
It was 165981500Hz, which is pretty close to the number Rob measured in elog:952, but significantly different from
the label on the Marconi.
I set the frequencies of all the MARCONIs accordingly and updated the labels.

After this, the AS166 demodulation phase was still not good enough (the Q and I signals were about the same).
So I rotated the phase by 45deg. In principle, this should set the demod-phase right for DARM too. Is it correct, Rob ?
I also adjusted the PD offsets. After those adjustments, MICH locks stably with a slightly increased gain (20 as compared to 10 before).
  1219   Fri Jan 9 09:14:17 2009 steveUpdatePEMifo is recovered after eq
There is no obvious damage from last night earth quake.
All sus dampings were turned on, MC locked and the arms locked right on
Attachment 1: eq4.5sb.jpg
eq4.5sb.jpg
  1220   Fri Jan 9 16:52:18 2009 steveUpdateVACtp3 forline pump replaced
Alberto took 40m vacuum 101 class as we replaced the drypump at the annulus pump line.

He is still not authorized to use the monster 2 3/8" open end wrench that is 36" long.

The fore line pressure dropped to 20 mTorr from 1 Torr as the pumps were swapped.
Bob needs to be given credit for replacing the tip seal on this Varian SH-100 drypump
The ss-hose felt dry at the tp3 exhaust end but it was some what "teflon coated-placticky-
-almost oily" at intake end of the dry pump.
We'll have to replace this metal hose next time.

This is a reminder that the 40m vacuum operation is fully manual.
It requires two people to switch a vacuum valve and one of them has to be experienced.
  1221   Fri Jan 9 17:30:10 2009 KakeruUpdateComputersSnapshots of MEDM screens
I wrote a web page which shows snapshots of MEDM screens generated by Yoich's script (e-log #1206).
https://nodus.ligo.caltech.edu:30889/medm/screenshot.html
This page refreshes itself every 5 minutes automatically.

The .html file is generated by /cvs/cds/caltech/statScreen/bin/genHtml.pl
This script generates the .html file contains snapshots listed on /cvs/cds/caltech/statScreen/etc/medmScreens.txt every 5 minutes with cron.
When you wont to display other screens, please edit this .txt file and wait 5 minutes!


To make thumbnails, I wrote /cvs/cds/caltech/statScreen/bin/genThumbnail.pl
This script reads /cvs/cds/caltech/statScreen/etc/medmScreens.txt, too.
(Sometimes, it makes thumbnails with larger storage...)


Quote:
I wrote scripts to take snapshots of MEDM screens in the background.
These scripts work even on a computer without a physical display attached.
You don't need to have X running.
So now the scripts run on nodus every 5 minutes from cron.
The screen shots are saved in /cvs/cds/caltech/statScreen/images/

There is a wiki page for the scripts.
http://lhocds.ligo-wa.caltech.edu:8000/40m/captureScreen.sh

Someone has to make a nice web page summarizing the captured images.
  1222   Mon Jan 12 10:57:38 2009 robUpdateGeneralsome stuff

The AS beam was not hitting the AS166 diode, so I aligned the last little steering mirror and adjusted the phase for MICH locking.

I turned on the HV supplies for the OMC.

Then I realigned the beam onto the AS166 diode, since the steering mirrors came on when I turned on the HV supplies.

It took awhile to find the alignment of the beam into the OMC. Once that was done, the output beam alignment was set, so I aligned onto the AS166 diode a third time.

The bottom two Sorensens in the OMC voltage supply don't look right. They have stickers that say +-24V, but each is sitting at 17.5V and showing no current draw. What's going on here?
  1223   Mon Jan 12 18:53:03 2009 YoichiUpdateLSCAS CCD centering and ASDD demod phase
After Rob's AS beam work, I centered the beam on the AS CCD.
I also optimized the ASDD demod-phase for the MICH signal.
Rob suggested to me that whenever we restart or change the frequency of the DD Marconis, we have to re-optimize the demod-phase
because the initial phase of the Marconi is random. We had the power failure, so it was time to do so.
I confirmed that MICH hand-off from REFL33Q to AS133DDQ is ok.
I will do the same thing for the PRCL, SRCL hand-offs.
  1225   Tue Jan 13 18:59:09 2009 KakeruUpdateLSCAS CCD centering and ASDD demod phase
I tuned the demod-phase for PRCL and SRCL hand-off, but it have not been optimized enoughly.
I continue this work tomorrow.


Quote:
After Rob's AS beam work, I centered the beam on the AS CCD.
I also optimized the ASDD demod-phase for the MICH signal.
Rob suggested to me that whenever we restart or change the frequency of the DD Marconis, we have to re-optimize the demod-phase
because the initial phase of the Marconi is random. We had the power failure, so it was time to do so.
I confirmed that MICH hand-off from REFL33Q to AS133DDQ is ok.
I will do the same thing for the PRCL, SRCL hand-offs.
  1229   Thu Jan 15 09:19:32 2009 steveUpdateIOOMC locking
MC2 sus damping was found tripped at the morning the second time this week.

Damping was restored, ISS gain lowered to avoid saturation, MZ manually locked
and MC locking was back.
  1231   Fri Jan 16 11:28:54 2009 YoichiUpdateComputersLab. laptop needs wireless lan driver update
One of the lab. laptops (belladonna) cannot connect to the network now.
I guess this was caused by someone clicked the update icon and unknowingly updated the kernel, which resulted in the wireless lan driver malfunctioning.
It was using a Windows driver through ndiswrapper.
Someone has to fix it.
  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).
Attachment 1: TRXY.pdf
TRXY.pdf
Attachment 2: ErrorSignals.pdf
ErrorSignals.pdf
Attachment 3: coherenceBetweenArms.pdf
coherenceBetweenArms.pdf
Attachment 4: ArmPowersAfterPDwasCentered.pdf
ArmPowersAfterPDwasCentered.pdf
  1234   Fri Jan 16 18:29:08 2009 YoichiUpdateSUSOplevs QPDs centered
Kakeru centered ITMX and BS optical levers with the help of Jenne on the walkie-talkie.
  1237   Mon Jan 19 13:58:53 2009 YoichiUpdateASCBetter ditherServo.pl
Nick Smith (@LHO) tested the ditherServo.pl at Hanford.
He added options to specify exit conditions to the script. Now you can make the script exit when
a condition, such as ArmPower > 1.0, is satisfied, or let it wait until a certain condition is satisfied.

I also modified the script to use ezcastep instead of tdswrite for feedback actuation.
The script now runs ezcastep in the background while the next iteration of the tdsdmd is performed.
Instead of kicking mirrors with a big thrust each time by a single tdswrite command, ezcastep gently moves the mirrors with fine steps.
I also implemented this "background ezcastep" technique in Tobin's tdscntr.pl.

The alignment scripts run smoother now.
  1239   Mon Jan 19 18:21:41 2009 ranaUpdateComputersloadLIGOData a GUI for mDV
The tool is very nice; I looked at the seismic trend for 16 days (attached).
However, it gives some kind of error when trying to get Hanford or Livingston data.
Attachment 1: a.png
a.png
  1240   Tue Jan 20 15:28:42 2009 YoichiUpdateComputersloadLIGOData a GUI for mDV

Quote:
The tool is very nice; I looked at the seismic trend for 16 days (attached).
However, it gives some kind of error when trying to get Hanford or Livingston data.


I fixed it.
You have to click "Load channels" button when you select a new site.
I plotted one minute of MC_F signals from H1, H2, L1 and 40m.
Looks like L1 MC was swinging a lot.
Attachment 1: MC_F.png
MC_F.png
  1241   Wed Jan 21 16:18:17 2009 KakeruUpdateLSCAS CCD centering and ASDD demod phase
I tuned the DD demod-pahse for SRM.
It was tuned as the error singnal is to be 0 when the cavity is locked.

The problem is that the good phase changes if MICH and PRM are handed to DD or not.
This may be a result of the demod-phase of these two signals are tuned to be maximise the error signal, not to be 0-offset.

I will tune these two demod-phases, and write a script to tune.


Quote:
I tuned the demod-phase for PRCL and SRCL hand-off, but it have not been optimized enoughly.
I continue this work tomorrow.


Quote:
After Rob's AS beam work, I centered the beam on the AS CCD.
I also optimized the ASDD demod-phase for the MICH signal.
Rob suggested to me that whenever we restart or change the frequency of the DD Marconis, we have to re-optimize the demod-phase
because the initial phase of the Marconi is random. We had the power failure, so it was time to do so.
I confirmed that MICH hand-off from REFL33Q to AS133DDQ is ok.
I will do the same thing for the PRCL, SRCL hand-offs.
  1242   Wed Jan 21 22:53:08 2009 ranaUpdateLSCAS CCD centering and ASDD demod phase
Just my opinion, but I think all we want out of the DD signals is something to control the DRM
and not be sensitive to the carrier and the CARM offset. So if the handoff can be done so that
the lock point is unchanged from single demod then everything is fine.

A second order concern is how the 133 & 199 MHz signals are mixed in order to minimize the
matrix cross-coupling and the SNR of the diagonal elements.
  1245   Thu Jan 22 12:08:59 2009 peteUpdateoplevsoplev calibration
Following the procedure described in Royal Reinecke's 2006 SURF report, I've calibrated the ETMY yaw oplev DOF. The idea is to sweep the mirror tilt, measuring the transmitted cavity power and the oplev error signal. The cavity power can be related to the mirror tilt in radians following D. Anderson APPLIED OPTICS, Vol. 23, No. 17, 1984.

I've made a simple matlab script which spits out the final number; it calls Royal's perl script to do the sweep. I get 420 microrad/ct for ETMY yaw. In 2006 Royal got 250 microrad/ct. Could something have changed this much, or is one of us wrong? I'll double check my procedure and do the other arm cavity oplevs, and describe it in detail when I have more confidence in it.

Kakeru and I plan to extend this to handle the PRM, SRM, and BS. One script to rule them all.
  1248   Fri Jan 23 10:00:21 2009 steveUpdatePSLPMC transmission is down
The PMC transmission is going down.
I have not relocked the PMC yet.
Attachment 1: pmc4d.jpg
pmc4d.jpg
  1249   Fri Jan 23 12:48:12 2009 KakeruUpdateoplevsarm cavity oplev calibration
I calibrated optlevs of x and y arm cavity, indipendently from Peter's work.

ITMX pit: 77 microrad/ct
ITMX yaw: 73 microrad/ct
ETMX pit: 280 microrad/ct
ETMX yaw: 263 microrad/ct

ITMY pit: 120 microrad/ct
ITMY yaw: 93 microrad/ct
ETMY pit: 280 microrad/ct
ETMY yaw: 270 microrad/ct

This result is similar to Royal's one (within 30% difference except for ETMX pit), but different from Peter's in ETMY.

The attached figure is the data and fitted curve of ITMX pit.
I took this data for 8s, with 4 Hz excitation.
Attachment 1: ITMX_pitch.png
ITMX_pitch.png
  1250   Fri Jan 23 14:00:02 2009 YoichiUpdatePSLPMC transmission is down

Quote:
The PMC transmission is going down.
I have not relocked the PMC yet.


I tweaked the alignment to the PMC.
The transmission got back to 2.65. But it is still not as good as it was 3 days ago (more than 3).

It is interesting that the PMC transmission is inversely proportional to the NPRO output.
My theory is that the increased NPRO power changed the heat distribution inside the power amplifier.
Thus the output mode shape changed and the coupling into the PMC got worse.
MOPA output shows a peak around Jan-21, whereas the NPRO power was still climbing up.
This could also be caused by the thermal lensing decreasing the amplification efficiency.
Attachment 1: LaserPower.png
LaserPower.png
  1251   Fri Jan 23 16:33:27 2009 peteUpdateoplevsx-arm oplev calibrations
ITMXpit 71 microrad/ct
ITMXyaw 77 microrad/ct
ETMXpit 430 microrad/ct
ETMXyaw 430 microrad/ct

As with y-arm, my ITM measurements agree with Kakeru and Royal, but my ETM measurements are not quite a factor of 2 higher. Kakeru and I are investigatin.
  1254   Wed Jan 28 12:42:51 2009 YoichiUpdatePSLMOPA dying
Yoichi, Jenne, Peter

As most of you know, the MOPA output power has been declining rapidly since Jan 21. (See the attachment 1)
There was also an increase in the NPRO power observed in LMON, which is an internal power monitor of the NPRO.
Similar trend can be seen in 126MON, which picks up some scattered light from the NPRO but there may be some contributions from the PA output.

The drop in the AMPMON, LMON and CURMON (NPRO current) from the middle of Jan 26 to the end of Jan27 was caused by me.
I tried to decrease the NPRO current to put the NPRO power back to the level when the MOPA output was higher. But it did not bring back the MOPA power.
So I put back the current after an hour. This caused the sharp power drop on Jan26.
By mistake, I did not fully recover the current at that time and left it like that for a day. This accounts for the long power drop period continued until Jan27.

Shortly after I tweaked the current, the MOPA output power started to fluctuate a lot. This drives the ISS crazy.
To see if this was caused by the NPRO or power amplifier,
we decided to fix the 126MON to monitor the real NPRO power.
We opened the MOPA box and installed a mirror to direct a picked off NPRO beam to the outside of the box through an unused hole.
We set up a lens and a PD outside of the MOPA box to receive this beam. The output from the PD is connected to the 126MON cable.
So 126MON is now serving as the real monitor of the NPRO power. It has not yet been calibrated.

The second attachment shows a short time series of the MOPA power and NPRO power. When the beam is blocked, the 126MON goes to -22.
So the RIN of the NPRO is less than 1%, whereas the MOPA power fluctuates about 5%. There is also no clear correlation between the power fluctuation of the MOPA and the NPRO. So probably the MOPA power fluctuation is not caused by NPRO.

At this moment, all the feedback signals (current shunt, slow and fast actuators) are physically disconnected from MOPA box so that we can see the behavior of MOPA itself.
Attachment 1: Recent10Days.png
Recent10Days.png
Attachment 2: 126_MOPA.png
126_MOPA.png
  1255   Wed Jan 28 12:51:32 2009 YoichiUpdateComputersMegatron is dying
For the past three days, Megatron has been making a huge noise. Sounds like a fan is failing.
There is an LED with "!" sign on the front panel. It is now orange. Looks like some kind of warning.
We can login to the machine. "top" shows the CPU load is almost zero.
Shall we try rebooting it ?
  1256   Wed Jan 28 19:08:50 2009 YoichiUpdatePSLLaser is back (sort of)
Yoichi, Peter, Jenne

Summary:
We found that the chiller water is not going to the NPRO base. It was hot whereas it was cold when I touched it a few months ago.
I twisted the needle valve on the water line to the NPRO base. Then we heard gargling noise in the pipe and the water started to flow.
The laser power is now climbing up slowly. The noisiness of the MOPA output is reduced.

I will post more detailed entry explaining my theory of what actually happened later.
Attachment 1: Improving.png
Improving.png
  1257   Thu Jan 29 13:52:34 2009 YoichiUpdatePSLLaser is back (sort of)
Here is what I think has happened to the laser.

After the chiller line to the NPRO base clogged, the FSS slow slider went down to keep the laser frequency constant.
It is evident in the attachment 1 that the behavior of the slow slider and the DTEC (diode temp. stabilization feedback signal) are almost the same except for the direction. This means the slow servo was fighting against the increased heat caused by the lack of the cooling from the bottom.
DTEC was doing the same thing to keep the diode temperature constant.

Even though the slow actuator (a Peltier on the crystal) worked hard to keep the laser frequency constant, one can imagine that there was a large temperature gradient in the crystal and the mode shape may have changed.

Probably this made the coupling of the NPRO beam to the PA worse. It may also have put the NPRO in a mode hopping region, which could be the cause of the noisiness.

Right now, the MOPA power is 2.7W.
The FSS, PMC, MZ are locked. At first, the PMC locked on a sideband. I had to twiddle the phase flip button of the PMC servo to lock the PMC. Probably this is another sticky channel, which needs to be tweaked after a reboot of c1psl. I added a code to do this in /cvs/cds/caltech/scripts/Admin/slider_twiddle.

Currently the ISS is unstable. Kakeru and I are now taking OPLTF of the servo.
Looks like the phase margin at the lower UGF is too small.
Attachment 1: SlowDC.pdf
SlowDC.pdf
  1259   Thu Jan 29 17:24:41 2009 KakeruUpdateoplevsarm cavity oplev calibration
I calibrated optlevs again. My previous work has a lot of mistakes, so ignore it.

ITMX pit: 195 microrad/ct
ITMX yaw: 185 microrad/ct
ETMX pit: 303 microrad/ct
ETMX yaw: 296 microrad/ct

ITMY pit: 192 microrad/ct
ITMY yaw: 141 microrad/ct
ETMY pit: 294 microrad/ct
ETMY yaw: 301 microrad/ct

(For ITMY, the data is low quality)

My calcuration and Peter's(based on Royal's report) is different in two point.
i) Royal uses some geometrical factor to calibrate ITM.
ii) Royal fits data to exp(-a^2/(2*w0^2)), and I fit data to exp(-a^2/w0^2).

When I calculate with modification of these differences, my result became almost same value of Peter's one.
Now we are discussing which equation is correct.


But we must do some laser works before it...
  1260   Thu Jan 29 18:10:13 2009 YoichiUpdatePSLISS Bad
Kakeru, Yoichi

As we noted before, the ISS is unstable. You can see the laser power oscillation around 3Hz.
We took the open-loop transfer function of the ISS around the lower UGF.
The phase margin is almost non-existent.
It was measured with the ISS gain slider at 2dB (usually it was set to 7dB).
So if we increase it by 3dB, it is guaranteed to be unstable.

The higher UGF has also a small phase margin (about 12deg.).
With the ISS gain slider at 2dB, the upper UGF is too low, i.e. the UGF is located at the beginning of the 1/f region.
So we if we make the lower UGF stable by lowering the gain, the upper UGF becomes unstable.

We took out the ISS box from the PSL table.
Kakeru and Peter are now trying to modify the filter circuit to give more phase margin at the lower UGF.
Attachment 1: OPLTF1.png
OPLTF1.png
  1262   Fri Jan 30 19:38:57 2009 KakeruUpdatePSLISS Bad
Kakeru, Peter

We try to improve ISS bord, but there isn't circuit diagram with correct parameters.
We are to measure transfar function and guess each parameter before we desogn new circuit parameters.
  1267   Mon Feb 2 19:23:53 2009 YoichiUpdateGeneralNew optical layout plan
The attached is a plan of the optical layout in the central part for the upgrade.
I included, the folded recycling cavities, oplevs for the core optics, POX, POY, POB and video views.
I have not worked out how to handle the beams outside the chambers. It should not be that difficult.
I also did not include beam dumps for unwanted beams.

I used pink for main beams, brown for picked off beams, red for oplevs.

Comments, suggestions are welcome.
Attachment 1: 40mUpgradeOpticalLayoutPlan01.pdf.zip
  1269   Tue Feb 3 19:24:14 2009 YoichiUpdateGeneralNew optics layout wiki page
I uploaded a slightly updated version of the new optics layout on the 40m wiki.
http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_08/Optical_Layout

I also uploaded the Mathematica notebook I used to calculate various parameters of the new recycling cavities, including the lengths, asymmetry, ROCs, PRM reflectivity and TT-mirror loss margin etc.
It would be nice if someone could check if the calculation is reasonable.
There is a PDF version of the document for non-Mathematica users.
  1270   Tue Feb 3 23:44:44 2009 Kakeru, Peter, YoichiUpdatePSLISS unstability

We found that one OP-amp used in ISS servo oscillated in 10 MHz, 100mV.

Moreover, we found another OP-amp had big noise.

We guess that these oscilation or noise cause saturation in high frequency, and they effect to lower frequency to cause 

 Attached files are open loop transfar function of ISS.

The blue points are open loop TF, and the green line is product of TF of ISS servo filter and TF of current shunt TF of servo filter.

This two must be same in principle, but They have difference f<2Hz and f>5kHz.

Attachment 1: TFgain.png
TFgain.png
Attachment 2: TFphase.png
TFphase.png
  1271   Wed Feb 4 17:45:39 2009 YoichiUpdateGeneralMode matching of the upgraded IFO
I did mode matching calculations for the new optical layout.
For the input mode matching, we have to change the focal length of the second mirror from 687mm to 315mm and the distance between the two MMT mirrors from 137mm to 149.2mm.
For the mode matching to the OMC, we only have to change the distance between the OMMT mirrors from 384mm to 391mm. No need to change the mirrors.

Details of the calculations can be found in
http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_08/Optical_Layout?action=AttachFile&do=get&target=NewRecyclingCavities.zip
(Mathematica notebook)
or if you prefer PDF, here
http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_08/Optical_Layout?action=AttachFile&do=get&target=NewRecCav.pdf
  1272   Wed Feb 4 19:22:57 2009 YoichiUpdateGeneralDo we need off-axis parabolic mirrors ?
I also estimated the mode matching degradation caused by the astigmatism.
Since the incident angles to the mode matching mirrors are not 0, the effective focal lengths in the incident plane and the perpendicular plane are different.
This effect leads to astigmatism of the beam.
When there is astigmatism, the maximum achievable mode matching rate becomes less than 100%.
According to my calculation, the mode matching cannot be better than 94% for the input beam.
For the output mode matching, we can theoretically achieve more than 99% even with the astigmatism.
The difference comes from the fact that the OMMT is longer, thus the incident angle is smaller.

If we don't like this 94%, we have to use off-axis parabolic mirrors, or modify the IMMT to a longer one.
I prefer to make it longer. Just 5" elongation will increase the mode matching rate to 99.4%.
We have a room for this 5" elongation.

Again, the details of the calculation are added to the Mathematica notebook below.


Quote:
I did mode matching calculations for the new optical layout.
For the input mode matching, we have to change the focal length of the second mirror from 687mm to 315mm and the distance between the two MMT mirrors from 137mm to 149.2mm.
For the mode matching to the OMC, we only have to change the distance between the OMMT mirrors from 384mm to 391mm. No need to change the mirrors.

Details of the calculations can be found in
http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_08/Optical_Layout?action=AttachFile&do=get&target=NewRecyclingCavities.zip
(Mathematica notebook)
or if you prefer PDF, here
http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_08/Optical_Layout?action=AttachFile&do=get&target=NewRecCav.pdf
  1274   Thu Feb 5 10:42:33 2009 YoichiUpdateGeneralDo we need off-axis parabolic mirrors ? No way !
I made a mistake in estimating the astigmatism problem.
If we use the current MMT1 as it is, this one is already an off-axis parabolic (OAP) mirror.
In this case, the astigmatism of this mirror is very small (if we use it with the correct angle). I did not include this effect in the previous calculation.
It turned out that the maximum achievable mode matching becomes far smaller (only 77%) if we use the OAP for MMT1 and a spherical mirror for MMT2.
This is not acceptable.
The reason behind this is that when we use spherical mirrors for both MMT mirrors, the astigmatism caused by the MMT1 is somewhat canceled by the astigmatism of MMT2. We don't get this cancellation if we mix OAP and spherical mirrors.

We should either (1) change MMT1 to a spherical mirror and keep the length of the input MMT as it is, or (2) change MMT1 to a spherical mirror and elongate the length of the input MMT.
In the case of (1) the maximum achievable mode matching is 94%. The focal length of MMT2 should be 315.6mm.
If we do (2), the mode matching rate can be as high as 99.8%. The focal lengths are MMT1 = -301.3mm, MMT2=558mm. The distance between the mirrors is 262mm.
We have enough space to do this elongation. But we have to mechanically modify the MMT mount.
I prefer (2).

As usual, the document on the Wiki was updated to include the above calculations.
ELOG V3.1.3-