Jeff Kissel, Stefan Ballmer

We started to modify to local svn copy of the suspension models to include a cleaned-up version of the ODC parts and an IPC sender to the ODC master

Modifications to ODC library:
 - The ODC library parts were cleaned up and repackaged. They are in cds/common/model/ODC.mdl and cds/common/model/ODC_CHAN.mdl. There will be some further changes inside ODC.mdl, but the inputs and outputs won't change any more.
 - We decided to give up full backwards compatibility of the parts to get them clean up. In particular:
   - In ODC.mdl, we changed the number of inputs of the CHANS (bit combiner) block from 31 to 30, because the 31st bit cannot be used anyway (it is reserved for the parity bit)
   - In ODC.mdl, for the COMBINE_ODC block, we added one input for the bitmask and one output for a latched version of the ODC channel for epics.
   - ODC_CHAN.mdl contains a new all-in-one ODC block for use in front-end models (called CHANNEL). (see plots) 
 - Files
       ODC_CHAN.mdl
       ODC.mdl
  
Modifications to SUS master models:
 - ODC_CHANNEL_OUT is now recorded as uint32 at full rate
 - The new, all in one ODC library block (called CHANNEL) was installed (see plots).
 - Its output is routed to the last output of the master model.
 - Updated models:
       OMCS_MASTER.mdl
       HSTS_MASTER.mdl
       TMTS_MASTER.mdl
       QUAD_MASTER.mdl
       HLTS_MASTER.mdl
       BSFM_MASTER.mdl
       MC_MASTER.mdl
 - we also started to update
       HAUX_MASTER.mdl
   which did not have and ODC output at all. Jeff will finish this one tomorrow (as well as a possible merger with HTTS_MASTER.mdl)

Modifications to suspension models
 - Added an ipc link (called H1:{OPTIC}_ODC_IPC) to the new ODC output from the master model
 - Updated models:
       h1sustmsx.mdl
       h1sustmsy.mdl
       h1susprm.mdl
       h1sussrm.mdl
       h1suspr2.mdl
       h1susbs.mdl
       h1suspr3.mdl
       h1sussr2.mdl
       h1sussr3.mdl
       h1susetmx.mdl
       h1susomc.mdl
       h1susetmy.mdl
       h1susmc1.mdl
       h1susmc2.mdl
       h1susitmx.mdl
       h1susitmy.mdl
       h1susmc3.mdl
 - All of them did recompile, but we did not install or run any of them yet.

None of the listed files are in svn yet - Jeff will implement additional changes tomorrow.

We developed and tested a prototype ODC Master model for use in the corner station on the Test Stand.  Screenshots of the model are attached.  

We tested that the 32 bit word calculations all work properly, including the new C library part used to take the BITWISE AND of multiple inputs.  

We also tested the updated C library part for setting a high-order odd parity bit in a 32 bit integer channel, and showed that it worked properly.  

We plan to migrate the new code for ODC Master to the OAF FE machine next.

Additionally, we carefully tested the new DAQ channel type setting "uint32".  We found that the data appears to be written to disk properly in binary, but FrDump and the python FrUtils both display the data as signed integers, so a value of uint32: 2147483654 becomes int32: -2147483642.  

- moved IMC_CUST_OVERVIEW.adl form ioo/h1/medm to ioo/common/medm
- replaced all site-specific occurrences (LHO,lho,H1,h1) with their macro counterparts for 
   ioo/common/medm/IMC_CUST_OVERVIEW.adl and
   ioo/common/medm/IMC_CUST_ODC.adl
- verified that the screens work at bot observatories.

- updated
   sus/common/medm/bsfm/SUS_CUST_BSFM_ODC.adl
   sus/common/medm/hxts/SUS_CUST_HLTS_ODC.adl
   sus/common/medm/hxts/SUS_CUST_HSTS_ODC.adl
   sus/common/medm/omcs/SUS_CUST_OMCS_ODC.adl
   sus/common/medm/quad/SUS_CUST_QUAD_ODC.adl
   sus/common/medm/tmts/SUS_CUST_TMTS_ODC.adl
to include the COMBINE_ODC_BITMASK and a link to the corresponding subsystem. Also updated
   sys/common/medm/SYS_CUST_ODC_CHAMBER.adl
   sys/common/medm/SYS_CUST_ODC_ENDCHAMBER.adl
   sys/common/medm/SYS_CUST_ODC_HAM1.adl
   sys/common/medm/SYS_CUST_ODC_HAM2.adl
   sys/common/medm/SYS_CUST_ODC_HAM3.adl
   sys/common/medm/SYS_CUST_ODC_HAM4.adl
   sys/common/medm/SYS_CUST_ODC_HAM5.adl
   sys/common/medm/SYS_CUST_ODC_HAM6.adl
   sys/common/medm/SYS_CUST_ODC_SITE_OVERVIEW.adl

All changes have been propagated to LLO.

Dave B., Jim B., Patrick T., Richard M., Robert S.

This was an experiment to test a fix to the 'noise' documented in alog 8076.

The problem may be that the sampling method used in the previous code (device support met_one_227b-1.0.1) may have initiated a short stage to purge the air path through the dust monitor before the start of each sample, and this was not happening with the new code (device support met_one_227b_comp_ctrl-1.0.0).

The previous code used this command:
"d" Start Counting (counter controlled): The counter will begin counting and control the count cycle based on the front-panel setting for period (sample time).

From Robert's comments in his original iLIGO code (http://blue.ligo-wa.caltech.edu/apps/epics/3.12.2.pem.llo/release/r3.12.2.PEM.LLO/base/src/drv/ansi/drvDustMetOne227.c):
The sample time is set in the dust monitor. This routine uses the "d" command (dust monitor control) rather than the "c" command (computer control). The advantage is that you can start several monitors almost simultaneously rather than sequentially. The monitor purges for two seconds before starting the sample - important for time stamping the sample. 

The new code uses the following command:
"c" Start Counting (computer controlled): The counter will begin counting without waiting for an even second boundary (quick start). Counting will continue until stopped by the computer. The count cycle should be controlled by the computer.
It may be that the "c" command does not run a purge prior to counting.

I added the following separate command with a short sleep after it before the "c" command and started testing it with the dust monitors in the optics labs:
"g" Active Mode: The counter will enter a mode that prepares it for counting. For example, the air pump may turn on to purge the air path.
It is unclear if this runs for a set period or until it receives the next command.

I added a one second sleep after each iteration through the dust monitors to address the code's high CPU use as well. I also added one second sleeps after caught exceptions in drv227b.cpp to provide some time to read the printed error message.

13:30: I stopped the dust monitor code for the optics labs and started the modified code. The sleep after the "g" command was set to 2 seconds.
14:00: I stopped the modified code, changed the sleep to 5 seconds, and restarted it.
14:14: I stopped the modified code again, changed the sleep back to 2 seconds, and restarted it.
15:00: I stopped the modified code.
- I checked the modified code into subversion. -
18:03: I restarted the modified code in 'screen' on h0epics from /ligo/apps/linux-x86_64/epics-3.14.12.2_long_sc/iocs/dust/dust_met_one_227b_comp_ctrl-1.1.0.

I have attached a plot of the dust counts in the Vacuum Prep Lab over this period. The change appears to have helped.

I will let this run over the weekend.

[Stefan and Kiwamu] (Translated by J. Kissel)

Approximately 2 months ago (see alog 7490), Stefan and Kiwamu modified the output matrix of the IMC's ASC control system, such that the caviy-axis-basis ("DOF" basis) alignment is decoupled with the input pointing alignment.

Jeff K. asked me to write some more details for the procedure of how we did it

Here it is, translated by Jeff.

The procedure of the adjustment :

• Lock IMC and engage all the IMC ASC loops.
• Make sure the output matrix is as follows
  • DOF1 feeds to just MC2 (1.0)
  • DOF2 feeds to MC1 and MC3 (+,+0.5 for Pitch and +,-0.5 for Yaw)
  • DOF4 has all zeroed elements
  • There is only one non-zero element in the DOF3 drive matrix namely H1:IMC-OUTMATRIX_P(Y)_4_3. (Computing the rest of the elements is our goal.)
• Make sure the input matrix is as follows
  • DOFs 1 and 2 are a derived by a linear combination of  WFSA and WFSB
  • DOF 3 is only measured by MC2 TRANS (1.0).
  • DOF4 has all zeroed elements

We're going to use the closed-loop, control output signals for these cross-coupled, cavity-axis-basis, loops as our metric of where the cross-coupled sensors think the cavity axis lies. We will then offset the error point of DOF3 (which controls the input pointing, with the PZT) at DC, to see how this affects all our controlled degrees of freedom (again at DC). Before we get started, there is some residual DC value of this control signal because the static alignment had not yet been tuned, and the integrators in the loops found the best spot for us. To make things easier to calculate later, we can offload this DC alignment control signal to the mirror/actuator basis (after the output matrix that we're trying to make better), such that the cavity-axis-basis control output signal is zero at DC. We have a script that already does this:

• /opt/rtcds/userapps/release/ioo/h1/scripts/imc/mcwfsrelieve
  • This moves the DC output control signal from  H1:IMC-DOF_1(2, 3)_P(Y)_OUTPUT to H1:IMC-MC1(2,3)_PIT(YAW)_OFFSET and H1:IMC-PZT_PIT(YAW)_OFFSET.
  • Now the control output for all three DOFs is zero (or close to zero with some fluctuation since the servos keep running).
• Introduce a small offset at the DOF3 error point in either YAW of PIT (Let's say we do this in PIT)
  • We introduced an offset of 0.2 in H1:IMC-DOF_3_P_OFFSET, which moves the PZT, changing the input pointing
  • This shifts the error point of the DOF3 loop such that it is parked at -0.2.
  • But ALSO, it changes the alignment of the input pointing and therefore of the MC REFL beam on WFSA and WFSB
  • WFSA and WFSB interpret this as the cavity axis changing, which inform DOF1 and DOF2
  • So DOF1 and DOF2 loops move all the mirrors to move in order to realign the cavity eigen axis to where they think it should be.
• Record the newly accumulated, static offset in all three control signals after all the servos settle down
  • This is H1:IMC-DOF_1/2/3_P_OUTPUT
• We have only introduced an offset in DOF3, and we *want* it to only control DOF3. So, we take the control output of DOF1 and DOF2, and compute the [DOF3 > MC1], [DOF3 > MC2], and [DOF3 > MC3] matrix elements of the output matrix to make this true, which "diagonalizes" the cavity-axis-basis. We do so with some maths:
  • DOF3 > MC1 =     (DOF2 OUTPUT) / [DOF2 > MC1]    *   [DOF3 > PZT] / (DOF3 OUTPUT )
  • DOF3 > MC2 =     (DOF1 OUTPUT) / [DOF1 > MC2]    *   [DOF3 > PZT] / (DOF3 OUTPUT )
  • DOF3 > MC3 =     (DOF2 OUTPUT) / [DOF2 > MC3]    *   [DOF3 > PZT] / (DOF3 OUTPUT )
• (or in channel form)
  • H1:IMC-OUTMATRIX_P_1_3 = (H1:IMC-DOF_2_P_OUTPUT)  /  ( H1:IMC-OUTMATRIX_P_1_2) * (H1:IMC-OUTMATRIX_P_4_3 ) / (H1:IMC-DOF_3_P_OUTPUT)
  • H1:IMC-OUTMATRIX_P_3_3 = (H1:IMC-DOF_2_P_OUTPUT)  /  ( H1:IMC-OUTMATRIX_P_1_2) * (H1:IMC-OUTMATRIX_P_4_3 ) / (H1:IMC-DOF_3_P_OUTPUT)
  • H1:IMC-OUTMATRIX_P_2_3 = (H1:IMC-DOF_1_P_OUTPUT)  /  ( H1:IMC-OUTMATRIX_P_2_1) * (H1:IMC-OUTMATRIX_P_4_3 ) / (H1:IMC-DOF_3_P_OUTPUT)

• Turn off the DOF3 pitch offset, enter in these values, and DOF3_PIT is done. (We keep the mirror/actuator basis offsets in place, because this is what the original loops said was the best alignment and we want to consistently return there regardless of whether the cavity-axis-basis loops are on or off.)
• Move on to YAW and do the same process in YAW.

I have written a python script to monitor the eight beam tube vacuum gauges and send emails to cell phones if a vacuum excursion is detected. It runs on sysadmin0 machine in the control room. It samples every 5 minutes, sending emails to John, Kyle, Richard, Ski and myself (both ligo-wa email and cell phone text messages).

code is in /opt/rtcds/userapps/release/cds/h1/scripts/ve_btvac_alarms_emailer.py

btvac = Beam Tube Vacuum.

This script performs the out-of-hours supplemental monitoring linked to the control room Alarm Handler and Wall Display striptools.

After some discussion with the SEI group, I made the following changes to the macro file naming at LHO to make both observatories identical and following the same standard (defined in T1300881)

Removed
opt/rtcds/userapps/release/isi/common/medm/*macro.txt
from the svn - these were originally put there in error by Joe Betzwieser. MEDM screens were not affected by that on either site.

Moved
hpi/h1/medm/hepi/hepi$(chamber)_overview_macro.txt
to 
hpi/h1/medm/hepitemplate/H1_hpi$(chamber)_overview_macro.txt

This makes the hpi macro file names and locations consistent between the two sites.
I updated the H1 site map, which was the only medm screens affected.

Moved
isi/h1/medm/{bsc/ham}isi/isi$(chamber)_overview_macro.txt
to
isi/h1/medm/{bsc/ham}isi/H1_isi$(chamber)_overview_macro.txt

This affected the following medm screens:
cds/h1/medm/SITEMAP.adl  (the main sitemap)  
isc/h1/medm/H1IFO_ALIGN.adl
ioo/h1/medm/H1IOO-MC.adl
ioo/h1/medm/IMC_CUST_OVERVIEW.adl
all of them were updated and checked in.



ODC screen work

updated 
hpi/common/medm/hepitemplate/HPI_CUST_CHAMBER_ODC.adl
isi/common/medm/bscisi/ISI_CUST_CHAMBER_ODC.adl
isi/common/medm/hamisi/ISI_CUST_CHAMBER_ODC.adl
as well as the sys/common/medm/SYS_CUST_ODC*.adl files
to include the ODC mask and a link to the corresponding subsystem medm screen.
those files were also propagated to LLO, and the links now seem to work for both sites.

Next: same thing for SUS

JimW GregG

HEPI actuator attachment started today at BSC9. Jim and I were able to get all of the vertical actuators attached with very little movement on the system so far. However, the vertical actuators are still locked and any moves to HEPI should be avoided until the install is finished.

•	Moving ISCT EX table from LVEA to air lock at end x -> Apollo
•	The dust monitor (unit#3) that was in the "beer garden" has been moved to the East door side of the BSC3 cleanroom -> Richard 
•	Site fumigation -> Sprague
•	Installing cables into HAM 5 and 6 IO chassis -> Aaron
•	The front end computer was powered down to allow EE work on the IO Chassis of h1sush56 -> David 

Alog from yesterday. Did in-rack cabling for the STS-2 Seismometer Distribution Chassis for BSC1, BSC2 and BSC3. Each distribution chassis fans out to the HAM ISI chambers (HAM2-HAM6). The following schematics were used for reference: D0901301, D1000298, D1101576, and D1101584.

See attached for the physical location of each seismometer.

Filiberto Clara 

to permit EE work on the IO Chassis of h1sush56 the front end computer was powered down at 13:47 and back up at 14:26 (PDT) when the work was completed.

Clipping problems encountered in the PRC at LLO have led to the discovery that the PRC scraper baffle "Z Plates" were built with an aperture approximately 0.75" too small. The PR2 scraper baffle assembly installed in H1 will require removal and re-work. For more detail please see the following link to the LLO log.
https://alog.ligo-la.caltech.edu/aLOG/index.php?callRep=9200

For the record, Richard is moving the dust monitor that was in the "beer garden" to the East door side of the BSC3 cleanroom.  Data drops and high alarm levels will be seen this morning during the move.

Mitchell, Gerardo, Apollo (Scott, Randy, Tyler), Thomas

We got the suspension and the box into chamber and mounted onto stage-0.  The next steps are:

- Photodiode cabling
- Fine-tune alignment
- Check balance
- Set the eddy current dampening gap
- Wipe down
- Swing Back for first contact removal

Over the last few days I've been taking some passive noise spectra of the ISI-CPS on HAM6 to try and determine what the best grounding strategy is (Jeff made a nice cartoon of the options in T1300871).

Along the way I found out a few things

  1) The HAM 6 CPSs had two problems that canceled each other out

                 There as a master in each crate  (the box near the chamber that hold the CPS controllers)

                 and the cable which brings the sync signal from the power board was either bad or plugged in backwards (it isn't keyed)

                I tunred one of the masters into a slave and replaced the cable

  2)  All of the racks that I have tested have there chasis grounded, so placing them on a cable tray creates a ground and possible ground loop.

  3) Currently it is pretty likely that ALL of the CPS that are installed have grounds in the electroncs room rack AND at the chamber (Ground Loop Gaore)

  4) The resistance to ground from the power cable ground to the chamber is 0.6ohms

  5) the sync cable carries a ground from one rack to the next

 After fixing the problems I took spectra under three conditions

   Multiple grounds (electroncs room and chamber) these were bad

  Single ground in the electronics room, red and green lines in the attached png (clean room on/off are the differences)

  Single ground at the chamber, blue lines in the attached png

   Given that set of data i think that we should plan on converting to a single ground at the chamber.

   The attached pdf shows one spectra converted to m/rthz and the expected noise (we have measured something close to the expected noise on the bench for every unit in use)

  I don't really know what to say about this except yuck

[Betsy Travis Arnaud]

Today, since BSC2 chamber was accessible, we checked the polarity of the four magnets located on the second mass (M2) of the beamsplitter with the electronic polarity tester.
As we were expecting (see 7693), three of them were flipped (UL UR LR), so Betsy reversed them as in E1100108.

Also, we took the serial numbers of the 10 bosems. They are referenced in the updated chart in E1200343

Attached are some nice pictures of ITMX ETMX BeamSplitter and the baffles

Both HAM2 and HAM3 chambers are now fully controlled (HEPI+ISI). 

Those chambers feature:
- Standardized ISI damping loops (defined in: SeiSVN/seismic/HAM-ISI/H1//Scripts/Control_Scripts/Version_2/Step_4_Damping_Filters_H1_ISI_.m)
- Standardized ISI blend filters (SeiSVN/seismic/HAM-ISI/Complementary_Filters_HAM_ISI/HAM-ISI_Generic_Comp_Filters.mat)
- Hybrid aLigo/eLigo blend filters are available (SeiSVN/seismic/HAM-ISI/Common/Blend_filters/Filters_HAM_ISI_Blend_Hybrid_24_Jul_2013.mat)
- All 3 levels of ISI isolation loops (UGF=10Hz, 25Hz, 35Hz - LHO aLog # 5355). Performance is mostly limited by the blend filters. Level 2 controllers give similar amount of isolation as Level 3. Level 2 controllers are prefferred for now - ISI Perfomance can be found in LHO aLog #7414

- HEPI position blend filters (IPS signals are multiplied by 1, L4C signals are multiplied by 0)
- HEPI position loops. UUG = 2 Hz. Should be set to 5Hz soon. HEPI position loops performance plots, and controllers, can be found in LHO aLog #8037.

Trun on Process:

Prepare:

1. Use the command button to shut down any kind of control on both HEPI and ISI.
2. Make sure the border of the ISI screen is green. If not, check the Watchdogs (model, Dackill, IOP dackill, SUS) and the master switch.

Damp ISI:

1. Use the command button to turn on the damping loops on the HAM-ISI.

Turn on HEPI:

1. Make sure that the following are in the correct state for HEPI: Watchdogs (model, Dackill, IOP dackill, SUS, ISI) and the master switch.
2. Open the IPS screen. Be careful there are 2 of them. One for the vertical sensors, and one for the horizontal sensors. 
3. In this screen, make sure that the alignment offsets are loaded and engaged. We want them to be set at the opposite of the locked hepi IPS redouts referrenced in LHO aLog# 7180
4. Open the DC bias screen and make sure that everything is disengaged - it should be.
5. Repeat (4) for the Sensor Correction screns, the ISC mon, the feedforward (FF) and the Twist.
6. Make sure that offsets are disengaged ont he OUTF screens. Be careful there are 2 of them as well. 
7. Push both gain sliders to 1.
8. Engage the Blend filters named "Pos"
9. Make sure signals come out of those filters. If not, refer to LHO aLog #7299, and follow Sebastian's instructions
10. Check the IPS 2CART matrices. For now, they should look like that. If not, use SeiSVN/seismic/HEPI/Common/MEDM_Functions_HEPI/Populate_Matrices_HEPI.m to fill them out. Do not use Populate_MEDM_Screen_HEPI.m instead, as it will reset a lot of the things you just worked on.
11. Use the "Isolate <Chamber> Lvl1" button of the HEPI command screen, to engage the position loops

Engage the isolation loops, on the ISI

1. Open the Sensor correction block and make sure that everything is disengaged.
2. Open the blend filters block. Make sure that filters named "01_28" are selected. These are the generic blend filters mentioned above.
3. Open the command screen of the ISI ou are working on
4. Wait for the readouts to stabilize
5. Use the "Reset CPS offset button" to reset the CPS biases. 
6. Press the "Store Target offsets" button. This will store the current position so the ISI can be steered back there once the isolation loops are engaged
7. Use "Turn on Lv. 2 isolation loops for ...." button, and wait for the isolation loops to get engaged.

Arnaud P. Betsy W. 

Did B&K hammer measurements on Friday feb 8th on the BeamSplitter that will be installed in BSC2. 
Config: BSC-ISI Unlocked, QUAD Unlocked, Baffles ON, Vibration Absorbers ON, Damping ON.

Attached, pictures describing the position of hammer impacts, and accelerometer + corresponding pulse files.