[Rana, Jenne]

There is some craziness going on with the delay in the PEM path for the OAF.  We plot the difference between the C1:PEM-SEIS_GUR1_X and C1:ASS-TOP_PEM_10.  These are physically the same channel, plugged into the PEM ADCU, and then the signal is used as a regular PEM channel, and is also sent to the ASS computer and used there for the OAF system.  As you can see in the blue trace on the bottom plot, there is a huge amount of delay, and it's very noisy.  We also plot the _GUR2_X / ASS-TOP_PEM_2 pair (red), and it has a similar amount of delay, but it is not nearly as fuzzy and noisy.  For comparison, we plot the SUS-MC2_MCL (which is identical to IOO-MC_L) and ASS-TOP_ERR_MCL pair (green), and they don't have any big overall delay problems, so it's not totally a problem with the signals getting to the ASS computer.

This problem was present during/after all of the following attempts to fix it:

* The sample rate on the ASS computer is 2048.  The PEM channels were being acquired the ADCU at 512.  We changed the ADCU sampling rate to 2048 to match.

* We soft rebooted the ASS computer, in case it was a timing problem.

* Doing a "sudo shutdown -r now" while logged in as controls.

We might also try resetting/power cycling c0dcu in the morning.  Alex has been emailed to help us try to figure this out.

In other news, the time delay that we measure from the plot gives us 180degrees in ~210Hz.  This corresponds to a little more than 2msec of delay, with the C1:ASS version lagging behind the C1:PEM version.  (2 samples at 840Hz) Converting to the 2048 sampling rate, we have a delay of 4.8, so 5 front-end cycles.  Since Rana measured this morning that the delay indicated by the transfer function is 10 cycles, and this delay shows that the ASS lags the actual seismometer signal by 5 cycles, we should subtract this 5 from the 10 from the transfer function, giving us a final sample-and-hold delay of 5.  Coincidentally(?), 5 is the delay that was found in the C1:ASS-TOP screen, after it's one year of dormancy.  The point of the delay feature in the code is to help match the delay in the two signal paths: the PEM path and the output path of the filter.  Since the output has a lag of 10, and the PEM path has a lag of 5, to make them match, we artificially put in a delay of 5.

At around 9:45 the RFM/FB network alarm went off, and I found c1asc, c1lsc, and c1iovme not responding. 

I went out to hard restart them, and also c1susvme1 and c1susvme2 after Jenne suggested that.

c1lsc seemed to have a promising come back initially, but not really.  I was able to ssh in and run the start command.  The green light under c1asc on the RFMNETWORK status page lit, but the reset and CPU usage information is still white, as if its not connected.   If I try to load an LSC channel, say like PD5_DC monitor, as a testpoint in DTT it works fine, but the 16 Hz monitor version for EPICs is dead.  The fact that we were able to ssh into it means the network is working at least somewhat.

I had to reboot c1asc multiple times (3 times total), waiting a full minute on the last power cycle, before being able to telnet in.  Once I was able to get in, I restarted the startup.cmd, which did set the DAQ-STATUS to green for c1asc, but its having the same lack of communication as c1lsc with EPICs.

c1iovme was rebooted, was able to telnet in, and started the startup.cmd.  The status light went green, but still no epics updates.

The crate containing c1susvme1 and c1susvme2 was power cycled.  We were able to ssh into c1susvme1 and restart it, and it came back fully.  Status light, cpu load and channels working.  However I c1susvme2 was still having problems, so I power cycled the crate again.  This time c1susvme2 came back, status light lit green, and its channels started updating.

At this point, lacking any better ideas, I'm going to do a full reboot, cycling c1dcuepics and proceeding through the restart procedures.

Power cycling c1dcuepics seems to have fixed the EPICs channel problems, and c1lsc, c1asc, and c1iovme are talking again.

I burt restored c1iscepics and c1Iosepics from the snapshot at 6 am this morning.

However, c1susvme1 never came back after the last power cycle of its crate that it shared with c1susvme2.  I connected a monitor and keyboard per the reboot instructions.  I hit ctrl-x, and it proceeded to boot, however, it displays that there's a media error, PXE-E61, suggests testing the cable, and only offers an option to reboot.  From a cursory inspection of the front, the cables seem to look okay.  Also, this machine had eventually come back after the first power cycle and I'm pretty sure no cables were moved in between.

 I had a go at trying to bring c1susvme1 back online.  The first few times I hit the physical reset button, I saw the same error that Joe mentioned, about needing to check some cables.  I tried one round of rebooting c1sosvme, c1susvme2 and c1susvme1, with no success.  After a few iterations of jiggle cables/reset button/ctrl-x on c1susvme1, it came back.  I ran the startup.cmd script, and re-enabled the suspensions, and Mode Cleaner is now locked.  So, all systems are back online, and I'm crossing my fingers and toes that they stay that way, at least for a little while.

Last night (Oct 07), I ran armLoss script in order to obtain the latest numbers for the arm cavity loss.
Here is the summary

Parameters:

TE=10ppm, LE=L_RT/2, RE=1-TE-LE
tE=Sqrt(TE), rE=Sqrt(RE)

TF=0.005, LF=L_RT/2, RF=1-TF-LF
tF=Sqrt(TF), rF=Sqrt(RF)

rcav = -rF +(tF^2 rE)/(1-rF rE)
R_cav = rcav^2

F = pi Sqrt(rF rE)/(1-rF rE)

I looked at the data of the day before yesterday (Oct 06) to know how much is the recycling gain.

X arm: (TRX_PRecycled) / (TRX_PRMmisaligned) * T_PRM = 83.1/0.943*0.07 = 6.17
Y arm: (TRX_PRecycled) / (TRX_PRMmisaligned) * T_PRM = 99.2/1.017*0.07 = 6.83

==> G_PR = 6.5 +/- 0.5     (oh...this estimation is so bad...)

From the recycling gain and the arm cavity reflectance, one can get the loss in the recycling cavity.

G_PR = T_PRM  / (1-Sqrt(R_PRM * (1-L_PRC)*R_cav))^2

==> loss in the recycling cavity L_PRC: 0.009+/-0.009
       (About 1% loss is likely in the recycling cavity)

I changed the input channels of the DMF recently so that it now uses 3 Guralp channels in addition to the 3 ACC and 1 Ranger.

op440m:seisblrms>diff seisBLRMS-datachans.txt~ seisBLRMS-datachans.txt
4,7c4,7
< C1:PEM-ACC_MC2_X
< C1:PEM-ACC_MC2_Y
< C1:PEM-ACC_MC2_Z
< C1:PEM-SEIS_MC1_Y
---
> C1:PEM-SEIS_GUR1_X
> C1:PEM-SEIS_GUR1_Y
> C1:PEM-SEIS_GUR1_Z
> C1:PEM-SEIS_RANGER_Y
op440m:seisblrms>pwd
/cvs/cds/caltech/apps/DMF/compiled_matlab/seisblrms

The seisBLRMS channels still have the wrong names of IX and EX, but I have chosen to keep them like this so that we have a long trend. When looking at the historical seisBLRMS trend, we just have to remember that all of the sensors have been around the MC since last summer.

Does anyone know if this master file is the real thing that's in use now? Are we really using a file called tpchn_C1_new.par? If anyone sees Alex, please get to the bottom of this.

allegra:daq>pwd
/cvs/cds/caltech/chans/daq
allegra:daq>more master
/cvs/cds/caltech/chans/daq/C1ADCU_PEM.ini
#/cvs/cds/caltech/chans/daq/C1ADCU_SUS.ini
/cvs/cds/caltech/chans/daq/C1LSC.ini
/cvs/cds/caltech/chans/daq/C1ASC.ini
/cvs/cds/caltech/chans/daq/C1SOS.ini
/cvs/cds/caltech/chans/daq/C1SUS_EX.ini
/cvs/cds/caltech/chans/daq/C1SUS_EY.ini
/cvs/cds/caltech/chans/daq/C1SUS1.ini
/cvs/cds/caltech/chans/daq/C1SUS2.ini
#/cvs/cds/caltech/chans/daq/C1SUS4.ini
/cvs/cds/caltech/chans/daq/C1IOOF.ini
/cvs/cds/caltech/chans/daq/C1IOO.ini
/cvs/cds/caltech/chans/daq/C0GDS.ini
/cvs/cds/caltech/chans/daq/C0EDCU.ini
/cvs/cds/caltech/chans/daq/C1OMC.ini
/cvs/cds/caltech/chans/daq/C1ASS.ini
/cvs/cds/gds/param/tpchn_C1_new.par
/cvs/cds/gds/param/tpchn_C2.par
/cvs/cds/gds/param/tpchn_C3.par

The c1ass computer, which is now used for the OAF system, has many remnants from the days when it was actually used as an ASS.  These PIT and YAW filter banks and other things were taking up a lot of unnecessary space, so I deleted them in the ass.mdl file.  These files are all backed up, so we can always revert back to an older version when we want some Alignment Stabilization again someday.  I then did a make ass, following the instructions on the 40m Wiki -> Computers and Scripts -> Simulink to Front-End Code page.  Rana moved some things around, most notably all of the things (like the ASS screens) which were only in ...../users/alex/.... are now in ....../caltech/cds/advLigo/..... .  This required a few restarts of the c1ass machine (after a couple different versions of the simulink diagram....one to make sure we knew how to do it, and then again actually deleting the unused portions).

The big lesson of the night was that there are 2 signal paths for the PEM channels.  As is shown in Figure 3 in the mevans document, the PEM channels get the matching filters when they go to the adaptation algorithm, but when they go to the FIR filter, they do not get the matching filters. This is implemented by taking the output of the giant PEM matrix, and having a duplicate of each of the channels "selected for adaptation", one which gets filtered through the PEM_N_ADPT banks, and one which goes straight (in code-land) to the FIR filter.  So, it seems like all the filters which we had been including in the input side of the matrix for matching purposes need to be put in the output side.  One of the AA32 filters needs to stay in the input side, for actual anti imaging of the PEM channels, then we put the AA32 and AI32 which are for matching the ERR_EMPH and CORR filter banks up in the PEM_N_ADAPT banks.  Rana and I made these filters, and they are now turned on appropriately with the OAF down script (so that all the filters are ready and waiting for the OAF to be turned on).

A little success with getting the 3Hz peak reduced, but not a lot beyond that.  Tomorrow I'll put the accelerometers back where they used to be to see if they help out at all.

"Yes. This master file is used."

I used the XARM as a reference to measure the frequency noise after the MC.  It's huge around 4kHz--hundreds of times larger than the frequency noise the MC servo is actually squashing.  This presents a real problem for our noise performance.

An elog search reveals that this noise has been present (although not calibrated till now) for years.  We're not sure what's causing it, but suspicion falls on the piezojena input PZTs. 

I didn't bother too much about it before because we previously had enough common mode servo oomph to squash it below other DARM noises, and I didn't worry too much about stuff at 4kHz..  Now that we have a weaker FSS and thus much weaker CM servo, we can't squash it, and the most interesting feature of our IFO is at 4kHz. 

I'll measure the actual voltage noise going to the PZTs.  I remember doing this before and concluding it was ok, but can't find an elog entry.  So this time maybe I'll  do it right.

Last night while noise hunting, Rana found that the MC2 trans beam has been left for an unknown length of time just hitting the side of the black enclosure box.  Today I put a brand-spankin'-new razor dump on the MC2 table, to dump the beam.

[Sanjit, Jenne]

The set of 6 accelerometers which were semi-randomly placed underneath the MC2 tank are now back into 2 separate sets of 3 - one set at MC2 and one set at MC1.  The channel names once again reflect reality, i.e. MC1_Y is actually under the MC1 tank, and aligned with the y direction.  Also, the Guralp under MC1 was moved a little bit to the left, because Sanjit wanted to put the accelerometers where the seismometer had been. 

Some of these channels are not like the others.

 Last week, Alex recompiled the c1susvme2 code without the decimation filters for the OUT16 channels, so these channels are now as aliased as the rest of them.  This appears to have helped with the timing issues: although it's not completely cured it is much better.  Attached is a five day trend.

Last night, 2+ hour lock, probably broken by me driving too hard (DARM_EXC).

Steve, Kiwamu, and Koji

We went through the PSL table to make sure any strong beam did not hit the wall.

We found that the reflection of Stephanie's OSA returned its path down to the beamsplitter.
This BS reflect that beam to the wall. That was fixed.

The surprising was that the relatively strong beam (~1mW?) went through the steering mirror
just before the PMC. We put thorlabs razor blades. I am still thinking what this indicates...
because the beam had been blocked if it was such from long time before.

During the work we found some stray optics such as a cube BS, a flipper mirror, and so on.
We can see them in the photo as those enclosed with yellow circles.
One of the beams was obtained from the reflection of the ND filter (...almost illeagal), and 
was even hittting a metal fixture for the BS cube.

If someone uses these components for useful purposes,
please let me(Koji) know. Otherwise, they are removed next week.

The other thing we found was the bright scatter from the EOM for the PMC.
As this scatter is so blight, I am going to align it.

--Apologies for the late post--

I was at the PSL table taking an inventory of the components for a while after Koji, Steve, and Kiwamu were there. I set the HEPAs back to 20% when I left (assuming that they were turned up when the compartment was opened).

 These components are from when Rana and I used the StochMon PD to do the RFAM tuning, documented in elog 1926.  This was a very handy measurement, but I'm not sure if whether or not we need it often enough to keep the optics there.

OK! I saw the optics are redundant and some of the components are not in a right place.
I will talk with you when you are back such that we can keep the usefulness of the setup.

You can remove the RFAM measuring setup. Once we upgrade, we will no longer have a MZ or the related problems.

I am at the PSL table taking what is hopefully the last set of pictures for the diagram. Woohoo.

 I'm out, HEPAs are back at 20%.

I have gotten the hang of the procedure for measuring phase noise on the AOMs. 

Koji suggested I right up a short guide (wiki page?) on how to do this. 

I will finish up here, then go measure the AOMs at the other lab (may have to be tomorrow, after laser safety), and then write up the instructions.

The IFO can now be locked during the daytime.  Well, it's locked now.

This is huge.    Five hours of lock only interrupted by intentional break from transfer function abuse.

Thermal lensing formula:

from (T090018 by A. Abramovici (which references another doc).

In the above equation:

w        1/e^2 beam radius

k        thermal conductivity (not the wave vector) = 1.3 W / m/ K

alpha    absorption coefficient (~10 ppm/cm for our glass)

NP       power in the glass (alpha*NP = absorbed power)

dn/dT    index of refraction change per deg  (12 ppm/K)

d        mirror thickness (25 mm for all of our SOS)

I'm attaching a plot showing the focal length as a function of recycling cavity power for both our current MOS and future SOS designs.

I've assumed a 10 ppm/cm absorption here. It may actually be less for our current ITMs which are made of Heraeus low absorption glass - our new ITMs are Corning 7980-A (measured to have an absorption of 13 ppm/cm ala the iLIGO COC FDD). I expect that our thermal lens focal length will always be longer than 1 km and so I guess this isn't an issue.

Bottom trace is proportional to the OMC PZT voltage - top trace is the transmitted light through the OMC. Interferometer is locked (DARM- RF) with arm powers = 80 / 100. The peaks marked by the cursors are the +(- ?) 166 MHz sidebands.

In addition to the main mirrors (PRM, SRM) we will also have fold mirrors (called PR1, PR2, SR1, SR2). I am curious to see if we can get away with just using commercial optics; I think that the CVI Y1S coatings may do the trick.

The above plots show the reflectivities v. wavelength. I've asked the sales rep to give us specs on the reflectivity v. angle. I bet that we can guess what the answer will be from these plots.

Awesome 5 hrs of locking  Rob!

I'm at the PSL table taking inventory of the elements I don't have down yet.

 OK, I'm out--hopefully for good. HEPAs back at 20%.

While measuring the Piezo Jena noise tonight we noticed that the LSC timing is setup strangely.

Instead of using the Fiber Optic Sander Liu Timing board, we are just using a long 4-pin LEMO cable which comes from somewhere in the cable tray. This is apparent in the rack pictures (1X3) that Kiwamu has recently posted in the Electronics Wiki. I think all of our front ends are supposed to use the fiber card for this. I will ask Jay and Alex what the deal is here - seems like to me that this can be a cause for timing noise on the LSC.

We should be able to diagnose timing noise between the OMC and the LSC by putting in a signal in the OMC and looking at the signal on the LSC side. Should be a matlab script that we can run whenever we are suspicious of this. This is an excellent task for a new visiting grad student to help learn how to debug the digital control system.

Pointing stability of 4 days. Initial pointing does not go through suspended optics. It is launched  right after the Piezo Jena steering mirrors in the BS chamber.

IP-ANG on epics screen is  C1:ASC-IBQPD_X and Y in dataviewer  were recentered. This beam is clipping a bit in ETMX chamber  pick off mirror.

IP-POS pick  off is in the BS chamber and it's qpd on the BS_ISCT This beam is also clipping just a little bit. This is easy to fix. We'll have to remove an iris from the BS optical levers table.

note: arms were not locked when I recentered

Some questions came arise to me:

A. How the green injection system should be? How the handing off between 532 and 1064 should be?

This is not new, though. It would be worth reminding.

B. Do we still need PMC if we use 2W innolight?

Innolight has low intensity noise at the detection freq. Also the spacial mode is clean.

C. Do we still need frequency prestabilization by RC?

Is the stabilization of the laser freq by the MC not enough?
What is the relationship with the green?

Could be this.

http://ilog.ligo-la.caltech.edu/ilog/pub/ilog.cgi?group=detector&task=view&date_to_view=10/02/2009&anchor_to_scroll_to=2009:10:02:13:33:49-waldman

I pushed the "closed loop" button on PZT2 YAW around 3:40 pm today, then roughly recentered it using the DC Offset knob on the PiezoJena controller and the IP ANG QPD readbacks.  There was a large DC shift.    We'll watch and see how much it drifts in this state.

This afternoon the elog crashed. We just restarted it.

 Here's the trend.

The transient at ~22:40 is Rob switching to 'Open Loop' on the Piezo Jena PZTs. I don't see any qualitative change in the drift after this event.

At 05:55 UTC, I removed an iris that was blocking the IP POS beam (the sum goes up from 2 to 6.5) without disturbing the mirrors who's oplev beam are on that table. Steve has conceded one sugar Napoleon after betting against my ninja-like iris skills.

We should recenter the beam on IP POS now that its unclipped - I'll let it sit this way overnight just to get more drift data.

Inspecting the AS table to make an inventory of the photodiodes in use around the interferometer, I found a mysterious photodetector hiding behind PD1 (AS166).

It turned out the detector was an old type of QPD from the Squeezing Experiment a few years ago.

We removed the box and the cable to which it was connected from the table. We stored it in the optics cabinet along the X arm.

I wanted to see how long our IP POS beam has been badly clipped - turns out its since April 1, 2007.

Steve's April Fool's joke is chronicled then. The attached trend shows that the drop in IP POS is coincident with that event.

In trying to align IPPOS, I noticed that someone has placed a ND2.0 filter (factor of 100 attenuation) in front of it. This is kind of a waste - I have removed IPPOS to fix its resistors and avoid this bad optic. Also the beam coming onto the table is too big for the 1" diameter optics being used; we need to replace it with a 2" diamter optic (Y1-2037-45P).

IP ANG dropped by a factor of 2 back in early August of '08.

We need this guy on the investigation:

Its back in and re-centered. Our next move on IPPOS should be to replace its steering mirror with something bigger and more rigid.

Electronics changes:

20K -> 3.65 K  (R6, R20, R42, R31) (unused)

20K -> 3.65 K  (R7, R21, R32, R43, R11, R24, R35, R46)

If you look in the schematic (D990272), you see that its an AD797 transimpedance stage with a couple of LT1125 stages set to give some switchable gain. It looks like some of these

switches are on and some are not, but I am not sure where it would be controlled from. I've attached a snapshot of one quadrant of the schematic below.

The schematic shows the switches in the so-called 'normally closed' configuration. This is what the switches do with zero volts applied to the control inputs. As the schematic also shows,

just disconnecting the 'switch' inputs cause the switch's control inputs to go high (normally open configuration, i.e. pins 2-3 connected, pin4 open). For the record, the default positions of the IPPOS switches are:

switch1   high

switch2   low

switch3   low

switch4   high

** EDIT (Nov 2, 2009): I forgot to attach the before and after images; here they are:

Today I made a measurement to research the LSC timitng issue as mentioned on Oct.16th.

*method

I put the triangular-wave into the OMC side (OMC-LSC_DRIVER_EXT) by AWG,

then looked at the transferred same signal at the LSC side (LSC_DARM_IN1) by using tdsdata.

I have compared these two signals in time domain to check whether they are the same or not.

In the ideal case it is expected that they are exactly the same.

*preliminary result

The measured data are shown in attached fig.1 and 2.

In the fig.1 it looks like they are the same signal.

However in fig.2 which is just magnified plot of fig.1, it shows a time-delay apparently between them.

The delay time is roughly ~50 micro sec.

The surprising is that the LSC signal is going beyond the OMC signal, although the OMC signal drives the LSC !!

We can say it is "negative delay"...

Anyway we can guess that the time stamp or something is wrong.

*next plan

Tomorrow I'm going to measure the transfer-function between them to see the delay more clearly.

( And I would like to fix the delay. )

I tried to compare the IP_POS time series with the IPANG and MC_TRANS but was foiled at first:

1) The IPANG scan rate was set to 0.5 second, so it doesn't resolve the pendulum motions well. Fixed in the .db file.

2) Someone had used a Windows/DOS editor to edit the .db file and it was filled with "^M" characters. I have removed them all using this command:   tr -d "\r" <ETMXaux.db > new.db

3) The MC_TRANS P/Y channels were on the MC Lock screen but had never been added to the DAQ. Remember, if there's a useful readback on an EPICS screen. its not necessarily in the frames unless you add it to the C0EDCU file. I have done that now and restarted the fb daqd. Channels now exist.

4) Changed the PREC of the IPPOS channels to 3 from 2.

5) changed the sign for the IBQPD (aka IPANG) so that bigger signal is positive on the EPICS screen.

You yourself told me that tdsdata uses some downconversion from 32k to 16k!

So, how does the downconversion appears in the measurement?
How does the difference of the sampling rate appears in the measurement?
If you like to understand the delay, you have to dig into the downconversion
issue until you get the EXACT mechanism including the filter coefficients.

AND, is the transfer function the matter now?

As far as the LSC and OMC have some firm relationship, whichever this is phase delay or advance or any kind of filering,
this will not introduce any noise. If so, this is just OK.

In my understanding, the additional noise caused by the clock jitter is the essential problem.
So, did you observe any noise from the data?

Of course I know there is a downconversion in OMC signal from 32k to 16k.

But I was just wondering if the delay comes from only downconversion.

And I can not find any significant noise in both signals because I use the triangular, which cause the higer harmonics and can hide the timing noise in frequency domain.

So I'm going to make the same measurement by using sinusoidal instead of triangular, then can see the noise in frequency domain.

Kiwamu, Alex and Zach are practicing mandatory IR-safety scan at the 40m-PSL

40m specific safety indoctrination were completed.

 Alex came in a week ago Friday to help figure this timing problem out, and some progress was made, although there's more to be done. 

Here are the (meager) notes that I took while he was working:

we can rename the tpchn_C1_new back to tpchn_C1, but the _new one works right now, so why change it?

need to find dcuDma.c source code...this is (?) what sends the PEM channels over to ASS.  Found:  source code is dcu.c, th
en the binary is dcuDma.o  Trying to recompile/remake dcuDma to make everything (maybe) good again.

Possibility: maybe having so many channels written to the RFM takes too long? shouldn't be  a problem, but maybe it is.  I
n the startup.cmd (or similar?) change the number of ISC modules to 1, instead of 2, since we only have one physical board
 to plug BNCs into, even though we have 2 isc boards.  c0dcu1 rebooted fine with the one isc board.  now can't get ass tes
tpoints to try the DTT timing measurement again.  rebooting fb40m to see if that helps.  fb40m is back, but we still don't
 have ASS testpoints.  Alex had to leave suddenly, so maybe more later.

Also, next possibility is that c0dcu and c1ass are not synched together properly....we should look at the timing of the AS
S machine.
 

After these adventures, the noisy trace in the timing delay (in the plot in elog 2066) has become quiet, as shown below (The blue trace, which was noisy in 2066 is now hiding behind the red trace).  However, the overall timing delay problem still exists, and we don't quite understand it.  Alex and I are meeting tomorrow morning at the 40m to try and suss this out.  Our first plan of attack is to look at the ASS code, to see if it puts any weird delays in.