I've been meaning to do this analysis ever since putting in the new laser at the X-end, and finally got down to getting all the required measurements. Here is a summary of my results, in the style of the preceeding elogs in this thread. I dither aligned the arms and maximized the green transmission DC levels, and also the alignment on the PSL table to maximize the beat note amplitude (both near and far field alignment was done), before taking these measurements. I measured the beat amplitude in a few ways, and have reported all of them below...

             XARM   YARM 
o BBPD DC output (mV), all measured with Fluke DMM
 V_DARK:     +1.0    +3.0
 V_PSL:      +8.0    +14.0
 V_ARM:      +175.0  +11.0

o BBPD DC photocurrent (uA)
I_DC = V_DC / R_DC ... R_DC: DC transimpedance (2kOhm)

 I_PSL:       3.5    5.5
 I_ARM:      87.0    4.0

o Expected beat note amplitude
I_beat_full = I1 + I2 + 2 sqrt(e I1 I2) cos(w t) ... e: mode overlap (in power)

I_beat_RF = 2 sqrt(e I1 I2)

V_RF = 2 R sqrt(e I1 I2) ... R: RF transimpedance (2kOhm)

P_RF = V_RF^2/2/50 [Watt]
     = 10 log10(V_RF^2/2/50*1000) [dBm]
     = 10 log10(e I1 I2) + 82.0412 [dBm]
     = 10 log10(e) +10 log10(I1 I2) + 82.0412 [dBm]

for e=1, the expected RF power at the PDs [dBm]
 P_RF:      -13.1  -24.5

o Measured beat note power (measured with oscilloscope, 50 ohm input impedance)      
 P_RF:      -17.8dBm (81.4mVpp)  -29.8dBm (20.5mVpp)   (38.3MHz and 34.4MHz)  
    e:        34                    30  [%]                          
o Measured beat note power (measured with Agilent RF spectrum analyzer)       
 P_RF:      -19.2  -33.5  [dBm] (33.2MHz and 40.9MHz)  
    e:       25     13    [%]                          

I also measured the various green powers with the Ophir power meter: 

o Green light power (uW) [measured just before PD, does not consider reflection off the PD]
 P_PSL:       16.3    27.2
 P_ARM:       380     19.1

Measured beat note power at the RF analyzer in the control room
 P_CR:      -36    -40.5    [dBm] (at the time of measurement with oscilloscope)
Expected    -17    - 9    [dBm] (TO BE UPDATED)

Expected Power: (TO BE UPDATED)
Pin + External Amp Gain (25dB for X, Y from ZHL-3A-S)
    - Isolation trans (1dB)
    + GAV81 amp (10dB)
    - Coupler (10.5dB)

The expected numbers for the control room analyzer in red have to be updated. 

The main difference seems to be that the PSL power on the Y broadband PD has gone down by about 50% from what it used to be. In either measurement, it looks like the mode matching is only 25-30%, which is pretty abysmal. I will investigate the situation further - I have been wanting to fiddle around with the PSL green path in any case so as to facilitate having an IR beat even when the PSL green shutter is closed, I will try and optimize the mode matching as well... I should point out that at this point, the poor mode-matching on the PSL table isn't limiting the ALS noise performance as we are able to lock reliably...

I don't think the loss of 25 ppm is outrageous. Its just surprisingly good. The SIS model predicted numbers more like 1 ppm / mirror taking into account just the phase map and not the coating defects.

However, we should take into account the lossed in the DRMI to be more accurate: AR coating reflectivities, scatter loss on those surfaces, as well as possible clipping around BS or some other optics.

https://chat.ligo.org/ligo/channels/40m

Quad rods and  ionizer kit: consisting of repeller cage, anode grid, focus plate and  filament were replaced....... under repair # RGA200/12 ECA 100416-12967

The electronic ECU is not connected. It is beeing pumped at IFO ITcc 9.7E-6 Torr vacNormal

Salton See is shaking again.

I poked around a bit thinking about what we need for a single AS WFS. 

New things that we would need:

• The WFS head itself: aLIGO ASC RF Photodetector (WFS) - E1400103 (Which will need some modification for 55MHz)
• Interface board for DC power and DC signals: aLIGO WFS Interface Chassis - D1101906
• Quad Demod chassis: Quad IQ Demodulator Chassis -  D0902796 [Variant B (WFS)] (Needs LO, but we have unused outputs in the LSC LO distributor)

Things we have:

• Many Eurocard-style Whitening chassis, such as we use for all of the LSC PDs. 
• Enough ADC (c1ioo has two ADCs, but doesn't even use 32 channels, so we can steal one, and throw it into c1lsc)

We'd have 12 new signals to acquire: 4 quadrants x DC, I, Q. In principle the DC part could go into a slow channel, but we have the ADC space to do it fast, and it'll be easier than mucking around with c1iscaux or whatever.

Open question: What to do about AA? A quick search didn't turn up any eurocard AA chassis like the ones we use for the LSC PDs. However, given the digital AA that happens from 64kHz->16kHz in the IOP, we've talked about disabling/bypassing the analog AA for the LSC signals. Maybe we can do the same for the QPD signals? Or, modify the post-demod audioband amplifer in the demod chassis to have some simple, not too agressive lowpass.

installing the BLRMS 2k blocks turned out to be quite non-trivial due to a whole host of CDS issues that had to be debugged, but i've restored everything to a good state now, and the channels are being logged. detailed entry with all the changes to follow.

Building:         Campus Wide         

Date:             Thursday 11/03/16 at Approx. 6:20 a.m.   

Notification:     Unplanned City Wide Power Glitch Affecting Campus   

*This is to notify you that the Caltech Campus experienced a campus wide power glitch at approx. 6:20 a.m. this morning.

The city was contacted and they do not expect any further interruptions related to this event.

The vacuum was not effected. ITM sus damping restored. IFO room air conditions on.

PSL Innolight and ETMY Lightwave lasers turned on

I did the following:

• Hard reboots for fb, megatron, and all the frontends, in that order
• Checked time on all FEs, ran sudo ntpdate -b -s -u pool.ntp.org where necessary
• Restarted all realtime models
• Restarted monit on all FEs
• Reset Marconi to nominal settings, fCarrier=11.066209MHz, +13dBm amplitude
• In the control room, restarted the projector and set up the usual StripTool traces
• Realigned PMC
• Slow machines did not need any touchups - interestingly, ITMX did not get stuck during this power glitch!

There was a regular beat coming from the speakers. After muting all the channels on the mixer and pulling the 3.5mm cable out, the sound persisted. It now looks like the mixer is broken

     ProFX8v2

A number of changes were made to C1PEM and some library parts. Recall that the motivation was to add BLRMS channels for all our suspension coils and shadow sensor PDs, which we are first testing out on the IMC mirrors.

Here is the summary:

BLRMS_2k library block

• The name of the custom C code block in this library part was named 'BLRMSFILTER' which conflicted with the name of the function call in the C code it is linked to, which lead to compilation errors
• Even though the part was found in /opt/rtcds/userapps/release/cds/c1/models and not in the common repository, just to be safe, I made a copy of the part called BLRMS_2k_40m which lives in the above directory. I also made a copy of the code it calls in /opt/rtcds/userapps/release/cds/c1/src

C1PEM model + filter channels

• Adding the updated BLRMS_2k_40m library part still resulted in some compilation errors - specifically, it was telling me to check for missing links around the ADC parts
• Eric suggested that the error messages might not be faithfully reporting what the problem is - true enough, the problem lay in the fact that c1pem wasn't updated to follow the namespace convention that we now use in all the RT models - the compiler was getting confused by the fact that the BLRMS stuff was in a namespace block called 'SUS', but the rest of the PEM stuff wasn't in such a block
• I revamped c1pem to add namespace blocks called PEM and DAF, and put the appropriate stuff in the blocks, after which there were no more compilation errors
• However, this namespace convention messed up the names of the filter modules and associated channels - this was resolved with Eric's help (find and replace did the job, this is a familiar problem that we had encountered not too long ago when C1IOO was similarly revamped...)
• There was one last twist in that the model would compile and install, but just would not start. I tried the usual voodo of restarting all the models, and even did a soft reboot of c1sus, to no avail. Looking at dmesg, I tracked the problem down to a burt restore issue - the solution was to press the little 'BURT' button next to c1pem on the CDS overview MEDM screen as soon as it appeared while restarting the model

All the channels seem to exist, and FB seems to not be overloaded judging by the performance overnight up till the power outage. I will continue to monitor this...

GV Edit 3 Nov 2016 7pm:

I had meant to check the suitability of the filters used - there is a detailed account of the filters implemented in BLRMSFILTER.c here, and I quickly looked at the file on hand to make sure the BP filters made sense (see Attachment #1). These the BP filters are 8th order elliptical filters and the lowpass filters are16th order elliptical filters scaled for the appropriate frequency band, which are somewhat different from what we use on the seismometer BLRMS channels, where the filters are order 4, but I don't think we are significantly overloaded on the computational aspect, and the lowpass filters have sufficiently steep roll-off, these should be okay...

The projector failed just now with a pretty loud 'pop' sound - I've never been present when the lamp goes out, so I don't know if this is usual. I have left the power cable unplugged for now...

Replacement is ordered Nov 4

It seems that the EX and EY BLRMS banks were missing the BP and LP filters for the 0.03-0.1 and 0.1-0.3 bands. I've copied over the filters from the BS seismometer.

However, if it looks like the integrated C code BLRMS block works out well, we could replace the seismometers' filter module heavy BLRMS blocks and cut down on the PEM model bloat.

[ericq, lydia]

Background: 

We believe the optimal OSEM damping would use an input matrix diagonalized to the free swing modes of the optic, and an output matrix which drives the coils appropriately to damp these free swing modes. As was discovered, a free swinging optic does not necessarily have eigenmodes that match up perfectly with pitch and yaw, however in the current state the "TO_COIL"  output matrix that determines the drive signals in response to the diagonlized sensor output also controls the drive signals for the oplevs, LSC/ASC, and alignment biases. So attempts to diagonalize the output matrix to agree with the input matrix have resulted in problems elsewhere. (See previous elog). So, we want to expand the "TO_COIL" matrices to treat the OSEM sensor inputs separately from the others.

Today:

• We modified the ETMX suspension model (c1scx) to use a modified copy of the sus_single_control block (sus_single_control_mod) that has 3 additional input columns. These are for the sensing modes determined by the input matrix, and are labeled "MODAL POS", "MODAL PIT", and "MODAL YAW." 
  • The regular POS, PIT, and YAW columns no longer include the diagonalized OSEM sensor signals for ETMX.
  • The suspension screen now out of date; it doesn't show the new columns under Output Filters and the summed values displayed for each damping loop do not incluse the OSEM damping.
• The new matrix can be acessed at /opt/rtcds/caltech/c1/medm/c1scx/C1SUS_ETMX_TO_COIL.adl (see Attachment 1). For now, it has the naive values in the new columns so the damping behavior is the same. 
• In trying to get a properly generated MEDM screen for the larger matrix, we discovered that the Simulink block for TO_COIL specifies in its description a custom template for the medm autogeneration. We made a new verion of that template with extra columns and new labels, which can be reused for the other suspensions. These templates are in /opt/rtcds/userapps/release/sus/c1/medm/templates, the new one is SUS_TO_COIL_MTRX_EXTRA.adl
• I will be setting the new column values to ones that represent the diagonlized free swing modes given by the input matrix. Hopefully this will improve OSEM damping without getting in the way of anything else. If this works well, the other SUS models can be changed the same way. 

Here are the channels we are planning to switch over from c1auxex to Acromag, and their current pin numbers on the existing VME boards. 

Analog inputs: 

C1:SUS-ETMX_UL_AIOut    #C0 S0
C1:SUS-ETMX_LL_AIOut    #C0 S1
C1:SUS-ETMX_UR_AIOut    #C0 S2
C1:SUS-ETMX_LR_AIOut    #C0 S3
C1:SUS-ETMX_Side_AIOut    #C0 S4
C1:SUS-ETMX_OL_SEG1    #C0 S5
C1:SUS-ETMX_OL_SEG2    #C0 S6
C1:SUS-ETMX_OL_SEG3    #C0 S7
C1:SUS-ETMX_OL_SEG4    #C0 S8
C1:SUS-ETMX_OL_X    #C0 S9
C1:SUS-ETMX_OL_Y    #C0 S10
C1:SUS-ETMX_OL_S    #C0 S11
C1:SUS-ETMX_ULPD    #C0 S12
C1:SUS-ETMX_LLPD    #C0 S13
C1:SUS-ETMX_URPD    #C0 S14
C1:SUS-ETMX_LRPD    #C0 S15
C1:SUS-ETMX_SPD    #C0 S16
C1:SUS-ETMX_ULV    #C0 S17
C1:SUS-ETMX_LLV    #C0 S18
C1:SUS-ETMX_URV    #C0 S19
C1:SUS-ETMX_LRV    #C0 S20
C1:SUS-ETMX_SideV    #C0 S21
C1:SUS-ETMX_ULPD_MEAN    #C0 S12
C1:SUS-ETMX_LLPD_MEAN    #C0 S13
C1:SUS-ETMX_SDPD_MEAN    #C0 S16

Analog Outputs:

C1:ASC-QPDX_S1WhiteGain    #C0 S0
C1:ASC-QPDX_S2WhiteGain    #C0 S1
C1:ASC-QPDX_S3WhiteGain    #C0 S2
C1:ASC-QPDX_S4WhiteGain    #C0 S3
C1:SUS-ETMX_ULBiasAdj    #C0 S4
C1:SUS-ETMX_LLBiasAdj    #C0 S5
C1:SUS-ETMX_URBiasAdj    #C0 S6
C1:SUS-ETMX_LRBiasAdj    #C0 S7
C1:LSC-EX_GREENLASER_TEMP    #C0 S0 This appears to have the same pin as another channel-- is it not being used? 

Binary Outputs:

C1:SUS-ETMX_UL_ENABLE    #C0 S0
C1:SUS-ETMX_LL_ENABLE    #C0 S1
C1:SUS-ETMX_UR_ENABLE    #C0 S2
C1:SUS-ETMX_LR_ENABLE    #C0 S3
C1:SUS-ETMX_SD_ENABLE    #C0 S4
C1:ASC-QPDX_GainSwitch1    #C0 S7
C1:ASC-QPDX_GainSwitch2    #C0 S8
C1:ASC-QPDX_GainSwitch3    #C0 S9
C1:ASC-QPDX_GainSwitch4    #C0 S10
C1:AUX-GREEN_X_Shutter2    #C0 S15

We don't need to record any of the AIOut channels, the OL channels (since we record them fast), or the _MEAN channels (I think they must be CALC records or just bogus).

I just realized that Gautam set this test up and turned damping off......He will explane the details

Short summary of my Sat. Box. debugging activities over the last few days. Recall that the SRM Sat. Box has been plugged into the PRM suspension for a while now, while the SRM has just been hanging out with no electrical connections to its OSEMs.

As Steve mentioned, I had plugged in Ben's extremely useful tester box (I have added these to the 40m Electronics document sub-tree on the DCC) into the PRM Sat. Box and connected it to the CDS system over the weekend for observation. The problematic channel is LR.  Judging by Steve's 2 day summary plots, LR looks fine. There is some unexplained behavior in the UR channel - but this is different from the glitchy behaviour we have seen in the LR channel in the past. Moreover, subsequent debugging activities did not suggest anything obviously wrong with this channel. So no changes were made to UR. I then pulled out the PRM sat.box for further diagnostics, and also, for comparison, the SRM sat. box which has been hooked up to the PRM suspension as we know this has been working without any issues. 

Tracing out the voltages through the LED current driver circuit for the individual channels, and comparing the performance between PRM and SRM sat. boxes, I narrowed the problem down to a fault in either the LT1125CSW Quad Op-Amp IC or the LM6321M current driver IC in the LR channel. Specifically, I suspected the output of U3A (see Attachment #1) to be saturated, while all the other channels were fine. Looking at the spectrum at various points in the circuit with an SR785, I could not find significant difference between channels, or indeed, between the PRM/SRM boxes (up to 100kHz). So I decided to swap out both these ICs. Just replacing the OpAmp IC did not have any effect on the performance. But after swapping out the current buffer as well, the outputs of U3A and U11 matched those of the other channels. It is not clear to me what the mode of failure was, or if the problem is really fixed. I also checked to make sure that it was indeed the ICs that had failed, and not the various resistors/capacitors in the signal path. I have plugged in the PRM sat. box + tester box setup back into our CDS data acquisition for observation over a couple of days, but hopefully this does the job... I will update further details over the coming days.

I have restored control to PRM suspensions via the working SRM sat. box. The PRM Sat. Box and tester box are sitting near the BS/PRM chamber in the same configuration as Steve posted in his earlier elog for further diagnostics...

GV Edit 2230 hrs 7Nov2016: The signs from the last 6 hours has been good - see the attached minute trend plot. Usually, the glitches tend to show up in this sort of time frame. I am not quite ready to call the problem solved just yet, but I have restored the connections to the SRM suspension (the PRM and SRM Sat. Boxes are still switched). I've also briefly checked the SRM alignment, and am able to lock the DRMI, but the lock doesn't hold for more than a few seconds. I am leaving further investigations for tomorrow, let's see how the Sat. Box does overnight.

I've been trying to understand the green beat setup on the PSL table to see if I can explain the abysmal mode-matching of the arm and PSL green beams on the broadband beat PDs. My investigations suggest that the mode-matching is very sensitive to the position of one of the lenses in the arm green path. I will upload a sktech of the PSL beat setup along with some photos, but here is the quick summary.

1. I first mapped the various optical components and distances between them on the PSL table, both for the arm green path and the PSL green path
2. Next, setting the PSL green waist at the center of the doubling oven and the arm green waist at the ITMs (in vacuum distances for the arm green backed out of CAD drawing), I used a la mode to trace the Gaussian beam profile for our present configuration. The main aim here was to see what sort of mode matching we can achieve theoretically, assuming perfect alignment onto the BBPDs. The simulation is simplified, the various beam splitters and other transmissive optics are treated as having 0 width
3. It is pretty difficult to accurately measure path lengths to mm accuracy, so to validate my measurement, I measured the beam widths of the arm and PSL green beams at a few locations, and compared them to what my simulation told me to expect. The measurements were taken with a beam profiler I borrowed from Andrew Wade, and both the arm and PSL green beams have smooth Gaussian intensity profiles for the TEM00 mode (as they should!). I will upload some plots shortly. The agreement is pretty good, to within 10%, although geometric constraints on the PSL table limited the number of measurements I could take (I didn't want to disturb any optics at this point)
4. I then played around with the position of a fast (100mm EFL) lens in the arm green path, to which the mode matching efficiency on the BBPD is most sensitive, and found that in a +/- 1cm range, the mode matching efficiency changes dramatically

Results:

Attachments #1 and 2: Simulated and measured beam profiles for the PSL and arm green beams. The origin is chosen such that both beams have travelled to the same coordinate when they arrive at the BBPD. The agreement between simulation and measurement is pretty good, suggesting that I have modelled the system reasonably well. The solid black line indicates the (approximate) location of the BBPD

Attachment #3: Mode matching efficiency as a function of shift of the above-mentioned fast lens. Currently, after my best efforts to align the arm and PSL green beams in the near and far fields before sending them to the BBPD results in a mode matching efficiency of ~30% - the corresponding coordinate in the simulation is not 0 because my length measurements are evidently not precise to the mm level. But clearly the mode matching efficiency is strongly sensitive to the position of this lens. Nevertheless, I believe that the conclusion that shifting the position of this lens by just 2.5mm from its optimal position degrades the theoretical maximum mode matching efficiency from >95% to 50% remains valid. I propose that we align the beams onto the BBPD in the near and far fields, and then shift this lens which is conveniently mounted on a translational stage, by a few mm to maximize the beat amplitude from the BBPDs. 

Unrelated to this work: I also wish to shift the position of the PSL green shutter. Currently, it is located before the doubling oven. But the IR pickoff for the IR beat setup currently is located after the doubling oven, so when the PSL green shutter is closed, we don't have an IR beat. I wish to relocate the shutter to a position such that it being open or closed does not affect the IR beat setup. Eventually, we want to implement some kind of PID control to make the end laser frequencies track the PSL frequency continuously using the frequency counter setup, for which we need this change...

[Lydia, Johannes]

We're waiting on the last couple electrical components to arrive that are needed to complete the acromag chassis, but it is essentially operational. Right now it is connected to the PSL Mephisto's diagnostics port, for which only a single XT1221 A/D unit is needed. We assigned the IP address 192.168.113.121 to it. For the time being I'm running a tmux session on megatron (named "acromag") that grabs and broadcasts the epics channels, with Lydia's original channel definitions. Since the chassis is 4U tall, there's not really any place in the rack for it, so we might want to move it to the X-end before we start shuffling rack components around. Once we finalize its location we can proceed with adding the channels to the frames.

For the eventual gradual replacement of the slow machines, we need to put some thought into the connectors we want in the chassis. If we want to replicate the VME crate connectors we probably need to make our own PCB boards for them, as there don't seem to be panel-mount screw terminal blocks readily available for DIN 41612 connectors. Furthermore, if we want to add whitening/AA filters, the chassis may actually be large enough to accomodate them, and arranging things on the inside is quite flexible. There are a few things to be considered when moving forward, for example how many XT units we can practically fit in the chassis (space availability, heat generation, and power requirements) and thus how many channels/connectors we can support with each.

Steve: 1X3 has plenty of room

This is where the mammal came though. It is reach able from room 108 CES

Looks like the PRM Sat. Box is now okay, no evidence of the kind of glitchy behaviour we are used to seeing in any of the 5 channels.

We set up the chassis in 1X7 today. Steve is ordering a longer 25 pin cable to reach. Until then the PSL diagnostic channels will not be usable.

This is long overdue, but our burt files for SDF now live in the LIGO userapps SVN, as they should.

The canonical files live in places like /opt/rtcds/userapps/release/cds/c1/burtfiles/c1x01_safe.snap and are symlinked to files like /opt/rtcds/caltech/c1/target/c1x01/c1x01epics/burt/safe.snap

I tried to realize an improvement in the mode matching onto the BBPDs by moving the lens mentioned in the previous elog in this thread. My best efforts today yielded X and Y beats at amplitudes -15.9dBm (@37MHz) and -25.9dBm (@25MHz) respectively. The procedure I followed was roughly:

1. Do the near-field far-field alignment of the arm and PSL green beams
2. Steer beam onto BBPD, center as best as possible using the usual technique of walking the beam across the photodiode
3. Hook up the output of the scope to the Agilent network analyzer. Tweak the arm and PSL green alignments to maximize the beat amplitude. Then move the lens to maximize the beat amplitude.

As per my earlier power budget, these numbers translate to a mode matching efficiency of ~53% for the X arm beat and ~58% for the Y arm beat, which is a far cry from the numbers promised by the a la mode simulation (~90% at the optimal point, I could not achieve this for either arm scanning the lens through a maximum of the beat amplitude). Looks like this is the best we can do without putting in any extra lenses. Still a marginal improvement from the previous state though...

I've been noticing over the last couple of days that the EPICS freezes are occurring more frequently again. Attached is an instance of StripTool traces flatlining. Not sure what has changed recently in terms of the network to cause the return of this problem... Also, they don't occur coincidentally on multiple workstations, but they do pop up in both pianosa and rossa.

Not sure if it is related, but we have had multiple slow machine crashes today as well. Specifically, I had to power cycle C1PSL, C1SUSAUX, C1AUX, C1AUXEX, C1IOOL0 at some point today

Now that we have all Satellite boxes working again, I've been working on trying to recover the DRMI 1f locking over the last couple of days, in preparation for getting back to DRFPMI locking. Given that the AS light levels have changed, I had to change the whitening gains on the AS55 and AS110 channels to take this into account. I found that I also had to tune a number of demod phases to get the lock going. I had some success with the locks tonight, but noticed that the lock would be lost when the MICH/SRCL boosts were triggered ON - when I turned off the triggering for these, the lock would hold for ~1min, but I couldn't get a loop shape measurement in tonight.

As an aside, we have noticed in the last couple of months glitchy behaviour in the ITMY UL shadow sensor PD output - qualitatively, these were similar to what was seen in the PRM sat. box, and since I was able to get that working again, I did a similar analysis on the ITMY sat. box today with the help of Ben's tester box. However, I found nothing obviously wrong, as I did for the PRM sat. box. Looking back at the trend, the glitchy behaviour seems to have stopped some days ago, the UL channel has been well behaved over the last week. Not sure what has changed, but we should keep an eye on this...

I took data of the ETMX SUSPOS, SUSPIT and SUSYAW channels while driving each of the 4 face coils. I manually turned off all the damping except the side. 

Excitation: I used white noise bandpassed from 0.4 to 5 Hz in order to examine the responses around the resonance frequencies. To avoid ringing things up too much, I started with a very weak drive signal and gradually increased it until it seemed to have an effect on the mirror motion by looking at the oplev signals/sensor RMS values on the SUS screen; it's possible I'll need to do it again with a stronger signal if there's not enough coherence in the data. 

Finding the matrix: The plan is to estimate the transfer function of the coil drive signal with the sensed degrees of freedom (specified by the already diagonalized input matrix). This transfer function can be averaged around the resonance peak for each dof to find the elements of the matrix that converts signals to dof responses, (the "response matrix", which is the inverse of the output matrix). Each column of the response matrix gets normalized so that the degrees of freedom influence the drive signals in the right ratio. 

Other notes: 

• I had some trouble getting the awg python library to work: I had to manually edit a CDLL statement to use the absolute path of an .so file. I wasn't sure what environment variable to set to make it look in the right folder automatically.
• The awg ArbitraryLoop object seems to be affected by cds getdata calls (The EXC signal stopes early and then stop() hangs) so I ended up doing the excitation and data reading in 2 separate scripts. 
• Reminder that the watchdogs must be on "Normal" for the EXC signal to make it to the coils, so the damping must be turned off manually with the watchdogs still on if you want to drive without damping. 

I replaced the suspected faulty DIMM earlier today (actually I replaced a pair of them as per the Sun Fire X4600 manual). I did things in the following sequence, which was the recommended set of steps according to the maintenance manual and also the set of graphics on the top panel of the unit:

1. Checked that Optimus was shut down
2. Removed the power cables from the back to cut the standby power. Two of the fan units near the front of the chassis were displaying fault lights, perhaps this has been the case since the most recent power outage after which I did not reboot Optimus
3. Took off the top cover, removed CPU 6 (labelled "G" in the unit). The manual recommends finding faulty DIMMs by looking for an LED that is supposed to indicate the location of the bad card, but I couldn't find any such LEDs in the unit we have, perhaps this is an addition to the newer modules?
4. Replaced the topmost (w.r.t the orientation the CPU normally sits inside the chassis) DIMM card with one of the new ones Steve ordered
5. Put everything back together, powered Optimus up again. Reboot went smoothly, fan unit fault lights which I mentioned earlier did not light up on the reboot so that doesn't look like an issue.

I then checked for memory errors using edac-utils, and over the last couple of hours, found no errors (corrected or otherwise, see Praful's earlier elog for the error messages that we were getting prior to the DIMM swap)- I guess we will need to monitor this for a while more before we can say that the issue has been resolved.

Looking at dmesg after the reboot, I noticed the following error messages (not related to the memory issue I think):

[   19.375865] k10temp 0000:00:18.3: unreliable CPU thermal sensor; monitoring disabled
[   19.375996] k10temp 0000:00:19.3: unreliable CPU thermal sensor; monitoring disabled
[   19.376234] k10temp 0000:00:1a.3: unreliable CPU thermal sensor; monitoring disabled
[   19.376362] k10temp 0000:00:1b.3: unreliable CPU thermal sensor; monitoring disabled
[   19.376673] k10temp 0000:00:1c.3: unreliable CPU thermal sensor; monitoring disabled
[   19.376816] k10temp 0000:00:1d.3: unreliable CPU thermal sensor; monitoring disabled
[   19.376960] k10temp 0000:00:1e.3: unreliable CPU thermal sensor; monitoring disabled
[   19.377152] k10temp 0000:00:1f.3: unreliable CPU thermal sensor; monitoring disabled

I wonder if this could explain why the fans on Optimus often go into overdrive and make a racket? For the moment, the fan volume seems normal, comparable to the other SunFire X4600s we have running like megatron and FB...

Looking back at elog 12528, the uncertainty in the armloss number from the individual quantities in the equation for can be written as:

Making some generous assumption about the individual uncertainties and filling in typical values we get in our measurements, results in the following uncertainty budget:

In my recent round of measurements I had a 2.5% uncertainty in the ASDC reading, which completely dominates the armloss assessment.

The most recent numbers are 57 ppm for the YARM and 21 ppm for the XARM, but both with an uncertainty of near 150 ppm, so while these numbers fit well with Gautam's estimate of the average armloss via PRG, it's not really a confirmation.

I set the whitening gain in ASDC to 24 dB and ran LSC offsets, and now I'm getting a relative uncertainty in measured reflected power of .22%, which would be sufficient for ~25ppm accuracy according to the above formula. I'm going to start a series of measurements tonight when I leave, should be done in ~2 hours (10 pm) the latest.

If anybody wants to do some night work: I misaligned ITMY by a lot to get its reflection off ASDC. Approximate values are saved as a restore point. Also the whitening gain on ASDC will have to be rolled back (was at 0dB) and LSC offsets adjusted.

I did apt-get update and then apt-get upgrade on optimus. All systems are nominal.

PRM and SRM sat. boxes have been switched for some time now - but the PRM sat. box has one channel with a different transimpedance gain, and the damping loops for the PRM and SRM were not systematically adjusted to take this into account (I just tweaked the gain for the PRM and SRM side damping loops till the optic damped). Since both sat. boxes are nominally functioning now, I saw no reason to maintain this switched configuration so I swapped the boxes back, and restored the damping settings to their values from March 29 2016, well before either of this summer's vents. In addition, I want to collect some data to analyze the sat. box noise performance so I am leaving the SRM sat. box connected to the DAQ, but with the tester box connected to where the vacuum feedthroughs would normally go (so SRM has no actuation right now). I will collect a few hours of data and revert later tonight for locking activities....

I powered up the existing ace100gm GigE cam with the PoE injector and tried to interface with it as described in elog 4163. After a few initial problems with IP assignment and interfacing I connected it to one of the gigabit hubs and installed the most recent pre-compiled software suite on /opt/pylon5 on optimus, after which I was able to find it with the configuration software. I named it "c1gige_bas100-1" and gave it the static IP address 192.168.113.151.

Afterwards the image acquisition worked without problems.

It may be a good idea to leave the gigecam interfacing up to a dedicated machine. I was thinking I could use Mafalda for this, and also for developing the code for framegrabbing and imager settings, but found that it was dead, burnt at the stake so to say. I guess it wasn't running anything critical, since it wasn't even connected to the network and smelled like burnt electronics. I'll get a replacement desktop for it.

I had a mistake in my script that reported the wrong error after averaging several datapoints, and because I hadn't looked at the individual numbers I didn't catch it so far. Thanks to Gautam it is no more.

The updated numbers are (with fresh, more trustworthy data):

XARM: 21 +/ 35 ppm
YARM: 69 +/- 45 ppm

This looks much better. I'm planning to take more data with the AS110 PD rather than AS55 when I get the chance, increase the averaging time, and also sigma filter the datapoints. That should get us to a good spot and cut down the uncertainty even further.

After much trial and error with whitening gains, demod phases and overall loop gains, I was finally able to lock the DRMI on both 1f and 3f signals! I went through things in the following order tonight:

1. Lock the arms, dither align
2. Lock the PRMI on carrier and dither align the PRM to get good alignment
3. Tried to lock the DRMI on 1f signals - this took a while. I realized the reason I had little to no success with this over the last few days was because I did not turn off the automatic unwhitening filter triggering on the demod screens. I had to tweak the SRM alignment while looking at the AS camera, and also adjust the demod phases for AS55 (MICH is on AS55Q) and REFL55 (SRCL is on REFL55I). Once I was able to get locks of a few seconds, I used the UGF servos to set the overall loop gain for MICH, PRCL and SRCL, after which I was able to revert the filter triggering to the usual settings
4. Once I adjusted the overall gains and demod phases, the DRMI locks were very stable - I left a lock alone for ~20mins, and then took loop shape measurements for all 3 loops
5. Then I decided to try transfering to 3f signals - I first averaged the IN1s to the 'B' channels for the 3 vertex DOFs using cds avg while locked on the 1f signals. I then set a ramp time of 5 seconds and turned the gain of the 'A' channels to 0 and 'B' channels to 1. The transition wasn't smooth in that the lock was broken but was reacquired in a couple of seconds.
6. The lock on 3f signals was also pretty stable, the current one has been going for >10 minutes and even when it loses lock, it is able to reacquire in a few seconds

I have noted all the settings I used tonight, I will post them tomorrow. I was planning to try a DRFPMI lock if I was successful with the DRMI earlier tonight, but I'm calling it a night for now. But I think the DRMI locking is now back to a reliable level, and we can push ahead with the full IFO lock...

It remains to update the auto-configure scripts to restore the optimized settings from tonight, I am leaving this to tomorrow as well...

Updated 16 Nov 2016 1130am

Settings used were as follows:

Nice job.

I am currently looking at the acoustic noise around both arms to see if there are any frequencies from machinery around the lab that stand out and to see what we can remove/change.

• Attachment 1 is a picture of the microphone and suspension system (bungee cords) that hangs from the cable trays to isolate it from vibrations.
• To record data, I used both the microphone (attachment 1) attach it its preamp connected to a spectrum analyzer in order get a graph of power spectral density, recording from 0-10k Hz and 10-100kHz. I started recording data at the furthest end of the x arm and worked towards the center taking measurements every couple of feet (ten rungs on the cable tray). 
• The second attachment is the first 5 psd I got from the furthest end of the x arm going 10 rungs on the cable tray closer each measurement.
• Going forward, I am going to take more measurements with greater resolution at the lower frequencies from 0-200 and stepping up from there by factors of 2.

Indeed. I suggest discussing with Joe B. I believe we should use a dedicated cam network to get the camera signals from the ends and corner all into one machine. Do not use the main CDS FE network for this since it might produce weird collissions. How about make a diagram, post it to elog, and send link to Joe?

As a starting point, I was looking at some of the old elogs and tried turning on the MCL feedback path with the existing control filters today. I tried various combinations of MCL Feedback and FF on and off, and looked at the MCL error signal (which I believe comes from the analog MC servo board?) spectrum for each case. We had used this earlier this year when EricQ and I were debugging the EX laser frequency noise to stabilize the low frequency excursions of the PSL frequency. The low frequency suppression can be seen in Attachment #1, there looks to be some excess MCL noise around 16Hz when the servo is turned on. But the MC transmission (and hence the arm transmission) decays and gets noisier when the MCL feedback path is turned on (see Attached StripTool screenshots).

I don't like AS110 or AS55. Neither of them are designed for DC and so the DC readout chain is hokey. How about use an actual transimpedance PD with a 100-1000 Ohm resistor and a 3 mm diode? This would eliminate the alignment sensitivity and the drifts due to electronics and room lights.

I had Rich show me his approach to a chassis for the Acromag modules. The document tree for his design can be found on the DCC. Note that he's using the high densitymodel ES series, which is available as a bare board variant with pluggable screw terminals:

He can fit up to 4 of these in a 2U chassis and has outsourced the wiring from front panel Dsubs to the board connectors to an external company. At the 40m (and in West Bridge) we currently only have the rail mounted XT series

At first glance the specs are very similar. Both A/D and D/A flavors have 16-bit precision in both cases. The high density ES series with Rich's layout can achieve 128 A/D per 2U, 64 D/A per 2U, or 384 DIO per 2U. Into a 4U chassis of the type we have currently we can fit ~32 XT modules (assuming two rows), which results in very similar numbers, except for the DAC, of which we could fit more.

XT1221-000 (8 diff. channel 16-bit ADC)                          $495.00      $61.88/ch
XT1541-000 (8 channel 16-bit DAC and 4 discrete I/O )    $525.00      $65.63/ch
XT1120-000 (16 channel DIO)                                         $320.00     $20.00/ch

ES2162-0010 (32 diff. channel 16-bit ADC)                     $2050.00    $64.06/ch
ES2172-0010 (16 channel 16-bit DAC)                           $1400.00    $87.50/ch
ES2113-0010 (96 channel DIO)                                      $1100.00    $11.46/ch

It's cheaper to stick with the current XT models, but they need the bulkier 4U chassis. The good news is that actually all these models have 16 bit precision, which wasn't clear to me before. Lydia and I will work out what connectors we want on the boxes, and how many modules/channels we need where. Rich also got me in touch with Keith Thorne, who handles the analog I/O Acromag at LLO, and I will ask him for advice. From his documents on the DCC it seems that he is using yet another series: EN. The 968EN-4008 for example is a rail-mounted ADC with pluggable connections, but looses quite clearly in price per channel.

For a generic multipurpose DAQ interface box the ES series is the best approach in my opinion, because it offers a more compact design. We could for example fit 1 ADC, 2 DAC, 1 DIO in a 2U chassis for 32/32/96 channels. The combined price tag for this scenario would be ~$6k.

Vivitek D952HD sn2160130 was send out for warranty repair. It's hard to believe that it has a 5 year warranty...... RMA - WR16004483.....expected to be back by Friday, Dec 2

[yinzi, craig, gautam]

Yinzi had translated the Perl PID script used to implement the discrete-time PID control, and had implemented it with Andrew at the PSL lab. Today afternoon we made some minor edits to make this suitable for the FSS Slow loop (essentially just putting the right channel names into her Python script). I then made an init file to run this script on megatron, and it looks to be working fine over the last half hour of observation or so. I am going to leave things in this state over the weekend to see how it performs.

We have been running with just the MC2 Transmission QPD for angular control of the IMC for a couple of months now because the WFS loops seemed to drag the alignment away from the optimum. We did the following to try and re-engage the WFS feedback:

• Close the PSL shutter, turned off all the lights in the lab and ran the WFS DC offsets script : /opt/rtcds/caltech/c1/scripts/MC/WFS/WFS_DC_offsets
• Locked the IMC, optimized alignment by hand (WFS feedback turned off) /opt/rtcds/caltech/c1/scripts/MC/WFS/WFS_DC_offsets
• Unlocked the IMC, went to the AS table and centered the spots on the WFS
• Ran WFS RF offsets script - this should be done with the IMC unlocked (after good alignment has been established) /opt/rtcds/caltech/c1/scripts/MC/WFS/WFS_DC_offsets
• Re-engaged WFS servo

GV addendum 23Nov2016: The WFS have been working well over the last few days - I've had to periodically (~ once in a day) run the WFS reflief script to keep the outputs to the suspension PIT and YAW DOFs below 50cts, but the WFS aren't dragging the alignment away as we had noticed before. The only thing I did differently is to follow Rana's suggestion and set the RF offsets with the MC unlocked as opposed to locked. I've added a line to the script to remind the user to do so... Also, note that EricQ has recently cleaned up the scripts directory to remove the numerous obsolete scripts in there...

I made a very slighly updated version of Yinzi's script that pulls the channel names and actuator hardstop limits from an external .ini config file. The idea was to avoid having as many versions of the script as there are implimentations of it. Seems like slighly better practice, but maybe I'm wrong. The config files are also easier to read. Its posted on the elog (PSL:1758) with lastest on the 40mSVN .../trunk/CTNLab/current/computing/scripts . 

If you're working off her first implimentation 'RCAV_thermalPID.py' then there is an indent issue after the if statement on line 88: only line 89 should be indended. If you deactivate the debug flag then the script fails to read in PID factors and dies.

Last jump at rack Y2.

Over the weekend, I was successful in locking the DRMI with the arms held on ALS. The locks were fairly robust, lasting order of minutes, and was able to reacquire by itself when it lost the lock in <1min. I had to tweak the demod phases and loop gains further compared to the 1f lock with no arms, but eventually I was able to run a sensing matrix measurement as well. A summary of the steps I had to follow:

• Lock on 1f signals, no arms, and run sensing lines, adjust REFL33 and REFL 165 demod phases to align PRCL, MICH and SRCL as best as possible to REFL33I, REFL165Q and REFL165I respectively
• I also set the offsets to the 'B' inputs at this stage
• Lock arms on ALS, engage DRMI locking on 3f signals (the restore script resets some values like the 'B' channel offsets, so I modified the restore script to set the offsets I most recently measured)
• I was able to achieve short locks on the settings from the locking with no arms - I set the loop gains using the UGF servos and ran some sensing lines to get an idea of what the final demod phases should be
• Adjusted the demod phases, locked the DRMI again (with CARM offset = -4.0), and took another sensing matrix measurement (~2mins). The data was analyzed using the set of scripts EricQ has made for this purpose, here is the result from a lock yesterday evening (the radial axis is meant to be demod board output volts per meter but the calibration I used may be wrong)

I've updated the appropriate fields in the restore script. Now that the DRMI locking is somewhat stable again, I think the next step towards the full lock would be to zero the CARM offset and turning on the AO path.

On the downside, I noticed yesterday that ITMY UL shadow sensor readback was glitching again - for the locking yesterday, I simply held the output of that channel to the input matrix, which worked fine. I had already done some debugging on the Sat. Box with the help of the tester box, but unlike the PRM sat. box, I did not find anything obviously wrong with the ITMY one... I also ran into a CDS issue when I tried to run the script that sets the phase tracker UGF - the script reported that the channels it was supposed to read (the I and Q outputs of the ALS signal, e.g. C1:ALS-BEATX_FINE_I_OUT) did not exist. The same channels worked on dataviever though, so I am not sure what the problem was. Some time later, the script worked fine too. Something to look out for in the future I guess..

Following up on the discussion from last week's Wednesday meeting, two points were raised:

1. How do we decide what number we want for the coating on the AR side for 532nm?
2. Do we want to adjust T@1064nm on the HR side to extract a stronger POP beam?

With regards to the coating on the AR side, I've put in R<300ppm@1064nm and R<1000ppm@532nm on the AR side. On the HR side, we have T>97% @ 532nm (copied from the current PR3/SR3 spec), and T<50ppm @1064nm. What are the ghost beams we need to be worried about? 

• Scattered light the AR side interfering with the main transmitted green beam possibly making our beat measurement noisier
  • With the above numbers, accounting for the fact that we ask for a 2 degree wedge on PR3, the first ghost beam from reflection on the AR side will have an angular separation from the main beam of ~7.6 degrees. So over the ~4m the green beam travels before reaching the PSL table, I think there is sufficient angular separation for us to catch this ghost and dump it. 
  • Moreover, the power in this first ghost beam will be ~30ppm relative to the main green beam. If we can get R<100ppm @532nm on the AR side, the number becomes 3ppm
• Prompt reflection from the HR surface of PR3 scattering green light back into the arm cavity mode 
  • The current spec has T>97% @532nm. So 3% is promptly reflected at the HR side of PR3
  • I'm not sure how much of a problem this really will be - I couldn't find the reflectivities of PR2 and PRM @532nm (were these ever measured?)
  • In any case, if we can have T<50ppm @1064nm and R>99.9% @532nm, that would be better

So in conclusion, with the specs as they are now, I don't think the ALS noise performance is adversely affected. I have updated the spec to have the following numbers now.

HR side: T < 50ppm @1064nm, T>99.9% @532nm

AR side: R < 100ppm @1064nm and @532nm

As for the POP question, if we want to extract a stronger POP beam, we will have to relax the requirement on the transmission @1064nm on the HR side. But recall that the approach we are now considering is to replace only PR3, and flip PR2 back the right way around. Currently, POP is extracted at PR2, so if we want to stick with the idea of getting a new PR3 and extracting a stronger POP beam, there needs to be a major optical layout reshuffle in the BS/PRM chamber. Koji suggested that in the interest of keeping things moving along, we don't worry about POP for the time being...

Alternatively, if it turns out that the vendor can meet the specs for our second requirement (which requires 1.5% of lambda @632nm measurement precision to meet the 10+/-5km RoC tolerance on PR3), then we can ast for T<1000ppm @1064nm for the HR coating on PR2, and keep the coating specs on PR3 as above. 

Attached is a pdf with the specs updated to reflect all the above considerations...

[Lydia, Johannes]

We connected and powered up the Acromag chassis today. It lives in 1X4 and is powered by the Sorensen +20V power supply in 1X5 via the fuse rail on the side of 1X4. For this we had to branch off the 20V path to the dewhitening and anti-image filter crate of the c1:susaux driven SOS optics. After confirming that none of the daughter modules in the crate draw from the 20V line, we added a wire leading to a new fuse we added for this unit and ran a power cable from there.

The diagnostic connector of the PSL laser is now connected to the unit and a tmux session was created on megatron that interfaces with the chassis and broadcasts the EPICS channels. We need to watch out in the coming days for epics freezes/outages, as in the past these seemed to occur around the same times we were toying with the Acromags.

The Rga was turned on yesterday.

Current Acromag chassis status:

I found out that Acromag offers DIN rail mounting kits for the open boards, so we can actually fit both XT series and ES/EN series in the same boxes, depending on the signal needs. The primary design driver will be the ES footprint, but if we find we don't need that many channels in some of the units, it's interchangable. For the wiring to the front panel - for which we will have a standard front panel express design, but may order modified ones for the custom needs of the 40m, I will contract the same company that Rich used for the wiring in his DIO box (Panel mount connectors terminating in loose wires/pre-routed plugs for Acromag units). We will either run a single DIN rail along the length of the chassis, or have two in parallel across.

Lydia and I took close looks at the breakout arrangements on the rack sides, and determined that because of the many cross-connects between non-DAQ ports it is not possible to redo and debug this in a reasonable amount of time without essentially shutting down the interferometer. So instead, we will connect the chassis directly to the slots that were previously leading to the slow machines. They come in two different flavors: The ADC modules have 64 pins, while the DIO and DAC ones have 50. There are a couple things we can do:

• For ADC: Put favorite 64+ pin connector on front panel. I would advocate for the 68 pin VHDIC (SCSI-5). This standard ist widely used, has a sturdy connector, and usually off-the-shelf cables have twisted pair leads.
• For DAC+DIO: Either use favorite 50 pin connector (there are 50-pin DSUB connectors, and also 50-pin IDC connectors with backshell), or also send the signals through VHDIC connectors, tolerating a few unused lanes. I would prefer the second option, after all it all goes to some 64 pin VME-crate backplane connector in the end, so if we ever get rid of the rack-side breakouts the wiring will much more uniform.
• For good measure, we will add a few (16 maybe) BNC connectors to the front panel.
• A standardized front panel could have a variety of different connectors by default: DSUBs, BNCs, etc., to be used when needed with some initial default wiring.
• Note that THEORETICALLY we could even connect all backplane EUROCARD ports to the Acromag chassis and do the cross-connect wiring entirely inside, although that would make the inside extremely messy.

Based on Rich's design I will get started on a parts list and wiring diagrams to send out to the cable company.

I wanted to get a clearer idea of the FSS servo and the various boxes in the signal chain and so Lydia and I poked around the IOO rack and the PSL table - I will post a diagram here tomorrow.

We then wanted to characterize the existing loop. It occurred to me later in the evening to measure the plant itself to verify the model shape used to construct the invP filter in the feedback path. I made the measurement with a unity gain control path, and I think there may be an extra zero @10Hz in the model.

Earlier in the evening, we measured the OLG of the MCL loop using the usual IN1/IN2 prescription, in which above 10Hz, the measurement and FOTON disagree, which is not surprising given Attachment #1.

I didn't play around with the loop shape too much tonight, but we did perform some trials using the existing loop, taking into account some things I realized since my previous attempts. The summary of the performanceof the existing loop is:

• Below 1Hz, MCL loop injects noise to the arm control signal. I need to think about why this is, but perhaps it is IMC sensing noise?
• Between 1-4Hz, the MCL loop suppresses the arm control signal
• Between 4-10Hz (and also between 60-100Hz for the Xarm), the MCL loop injects noise. Earlier in the evening, we had noticed that there was a bump in the X arm control signal between 60-100Hz (which was absent in the Y arm control signal). Koji later helped me diagnose this as too low loop gain, this has since been rectified, but the HF noise of the X arm remains somewhat higher than the Y arm.

All of the above is summarized in the below plots - this behaviour is (not surprisingly) in line with what Den observed back when he put these in.

The eventual goal here is to figure out if we can get an adaptive feedback loop working in this path, which can take into account prevailing environmental conditions and optimally shape the servo to make the arms follow the laser frequency more closely at low frequencies (i.e. minimize the effect of the noise injected by IMC length fluctuations at low frequency). But first we need to make a robust 'static' feedback path that doesn't inject control noise at higher frequencies, I need to think a little more about this and work out the loop algebra to figure out how to best do this...

TP3 drypump replaced at 655 mTorr, no load, tp3 0.3A 

This seal lasted only for 33 days at  123,840 hrs

The replacement is performing well: TP3 foreline pressure is 55 mTorr, no load, tp3 0.15A at 15 min  [ 13.1 mTorr at d5 ]

Valve configuration: Vacuum Normal, ITcc 8.5E-6 Torr