Applied Clean all (don't save configuration) and rebuild all to all PLC project.
Rescaned all PLC projects.
Activated new system configuration.
Created new boot projects.
Reactivated system configuration.
Open OPC configurator and saved/activated it.
Restarted iocShells.
All OPC channels now connect.
0510 hrs. local -> Arrived at site an found LN2 dripping on ground at the broken weld joint on the ambient air vaporizer -> piping from CP5 to ambient air vaporizer frosted -> piping from dewar to ambient air vaporizer was warm (not frosted) -> GN2 exhausting out exhaust piping and exhaust check-valve bypass valve closed -> closing then opening instrument air to LLCV confirmed that LLCV was, in fact, stroked fully open as indicated by CDS (100% open) -> magnehelic diff pres level gauge at 36" = 87% which agreed with CDS indication of 86% full -> discrepancy between observed pump level (LN2 present in regen line and pressure difference between vapor volume and liquid volume in pump(????)) Conclusion: Actual pump level is "over full" -> PID response was correct for the incorrect level indication -> I don't have an explanation (yet-still waiting for the coffee to kick in) as to why the magnehelic and CDS source pressure transducer indicated levels don't indicate an over full condition ???? -> For now, I'll ignore the indicated pump level and set the LLCV at 75% open (manual) and let the actual pump level fall off -> indicated pump level may correct when over full condition is absent????
(Late Post) The new edge switches at EX and EY were connected to the core today. The wireless AP at EY has been moved to the VEA and attached to the new switch - also verified that the signal is still strong in the entry area, so most (all) of the building that needs wireless coverage should have it. The wireless AP for EX will be installed tomorrow. Once we have a couple days running experience with the new switches, we will start moving the vacuum and FMCS systems to the new switches in advance of decommissioning the old switches. (Jumbo frames enabled, IOS 15.0.2SE2 at install.)
I modified the BSC-ISI master model to introduce the sensor correction path from the ground instrument (STS-2) to the stage 1 of the ISI. I added the 9 STS-2 inputs (3 signals per STS-2 and 3 STS-2s at the corner station). Now, 50 sensors signals feed to the BSC-ISI models.
The input filters of the STS are named: IFO:ISI-CHAMBER_ST1_GNDSTSINF_Group_DOF. In the existing stage 1 sensor correction block (the block was created to perform the sensor correction using the L4Cs of the HEPI), I added to the filters names dealing with the HEPI L4C the prefix HPI_L4C.
In this block, I also added filters (FIR and IIR) for the sensor correction path using the STS-2s. The prefix used for these filters is GND_STS. I integrated in the Master overview screen the STS and modified the stage 1 sensor correction path. A dozen of the screens were also created. While I was modifying the model, I also increased the sample rate of the excitation channels to 4K. I only installed the new model on ISI-BSC6.
I started to work on the Version 3 of the commissioning scripts to consider these changes in the models. For the moment, I have upgraded the “Routine 0” of the scripts to introduce the calibration filters of the STS-2. The STS-2 signals are added to the CPS after passing through a series of filters. I created a routine to load the IIR and FIR filters used to create the high pass filter. The companion filters ISI_4k_FIR_companion_filters_GND_STS_20130509.mat in userapps/release/isi/common/filterfiles. These filters are only usable for the ground path sensor correction (Different names for HEPI L4C)
I also created a routine to compute and load the “Match” filters used to correct the calibration errors between the STS-s2s, the T240s and the CPSs. The CPSs are used as reference. The reference is the simulated Cartesian transfer functions with the ISI damped.
In attachment, you can find the new overview screen (BSC-ISI_Overview.png). The modifications are visible are in the red box.
After measuring transfer functions from the STS-2 to the T240s (ISI damped) and using the damped transfer functions, I computed the match filters for the ground path sensor correction. The match filters are just gains close to 1 (+/- 3% correction).
I measured some spectra (calibrated in nm/s for the T240 and nm/s for the GS13 above 1Hz) with and without the sensor correction (STS-2). In the figure H1_ISI_ETMY_Sensor_Correction_20130509_155200.png, spectra are measured in the following configuration:
- High gain on L4Cs and GS13s
- On stage 1, blend at 250mHz with the T240s in the super sensor in X, Y and 750mhz without the T240s in Z, Rx, Ry, Rz.
- On stage 2, blend at 250mHz in X, Y and 750mHz on Z, Rx, Ry, Rz.
- UGF: 15Hz for all DOFs
- The plots with the "ref" label indicates that the sensor correction was engaged.
Results are pretty good especially in the Z direction (the blend frequency is 750mHz whereas it is 250mHz in the X and Y directions). Tomorrow, I will increase the UGFs of the isolation filters on the 2 stages.
We leveled the optical path from the laser to the first MCL PZT mirror, but things didn't improve much.
The first Faraday transmission is fine (first picture, EOM and the second Faraday removed from the path, picture taken in front of the second lens).
EOM transmission is ugly when the beam passes through the middle of the input and output aperture (second picture).
After fiddling around with the EOM alignment using 4 DOF stage, we always ended up with the beam very high on the input side. (no picture).
After the EOM, when the second Faraday was inserted, it makes many ghost beams /halo in PIT (third picture). This vertical pattern doesn't change much with alignment.
Using an iris, we can get rid of most of the ghost beams (last picture). The central part of the beam is not totally round, this comes from the EOM. This looks somewhat better than OAT days (sadly).
Anyway, since it is convenient to open up the iris to see the retro reflected beam from ETMY, we left all irises wide open.
The Faraday makes many ghost beams for the refl beam too, but all of these seem to fall on the PD.
We measured the power level right at the laser, downstream of the first Faraday, downstream of the EOM and downstream of the second Faraday. Jax will post the numbers.
We have removed some unused optics from ALS table.
One 50/50 green splitter (labeled as such on the mount) and one green high reflector (ditto) were put in the ISC cabinet in the squeezer bay together with things labeled "ALS WFS stuff".
One green high reflector didn't have any label, but it was easy to see that it was a high reflector. To avoid future confusion, I removed the optic and put it back in the original mirror box from Precision Photonics.
It seems like there are other unlabeled optics in "ALS WFS stuff" bunch, and maybe on our tables, but we should do better. PP optics are not labeled anywhere, the only visual clue for the reflectivity is the color of the coating, and some people might have a difficulty identifying which is which.
Make labels and put them on mirror holders or posts.
Power measurements as promised:
From laser: 42.5 mW
After first Faraday isolator: 35.5 mW
After EOM: 31 mW
After 2nd Faraday isolator: 28 mW
After 2nd PZT mirror: 26.3 mW - after inserting another iris this measured 27 mW, likely within the bounds of measurement error.
Note that there is a 10% pick-off for Hartman sensor path between the 2nd PZT mirror and the periscope.
All in all, the efficiency from the laser to the chamber is something like 0.9*27mW/42.5mW = 57%, which is consistent with June 2012 alog (61%) as well as July 2012 alog (0.9*20mW/30mW=60%, note that the measurement was done upstream of 10% pick off in July 2012).
With approval from Keita to proceed, Kiwamu got the Mode Cleaner locked and tuned up a bit. We then moved the ISI-in-a-can to as close as possible to WHAM4. It now sits north of HAM4, I say the center of the load is about 5m away from the center of the chamber. We monitored the lock state of the IMC and it remained locked through the duration. The load was lifted and started moving toward WHAM4 at 1456PDT(2156utc) and was set down at the above location at 2201utc.
Damping loops were turned on for the ISIs of the IMC to support locking effort.
The payload screen of the HAM-ISIs now feature links to the MEDM screens of the related suspensions. For example, one can now click on the MC1 button of the HAM2-ISI payload screen, and get re-directed to MC1's MEDM screen. This is especially useful if the suspension(s) tripped the ISI. One can then open the payload sceen, find out which suspension tripped the ISI, and un-trip it in a couple clicks.
Kyle has been pumping the Vertex volume for about 9 days now. The trend is attached. Note that about 1/2 of the annulus systems are under vacuum. The plot shows the affect of pumping GV2 annulus. This indicates a leak in this annulus system - likely not a serious problem. As pump carts free up they will move to other annulus systems.
Lightpipes (i.e. "light hoses") were installed on ISCT1's (1)REFL/POP & (2)ALS (green) paths. This completes all (3) light pipes for ISCT1.
The backup file server cdsfs1 required a reboot, it had become unresponsive. It was found that the tape backup machine had also lost connection with its SCSI library, so it was also rebooted which restored the connection.
Since we had not performed a disk-to-disk backup of cdsfs0 nor h1boot since yesterday evening, I did a manual rsync of these at noon. I then checked to see if the rsync of h1boot still causes front end epics slow down problems, and found that this problem is now not seen. So I have increased the frequency of h1boot's backup from daily at 5am to every four hours starting at 1am. We'll see if the epics slowdown problem reappears.
* Because the TMTS is significantly different from the OMCS, I started a new Mathematica model for it, based on the OMCS but with two (not four) blades at the top mass, blade lateral compliance at the top mass blades (just in case, because it's been important for quad blades in the past), and toe-out of the bottom wires in both the front-back and side-side directions (the front-back wire separation sl parameter has been forked to upper and lower values slu and sll). This can be found in the SUS SVN at ^/trunk/Common/MathematicaModels/DualLite2DBLateral . * As is now my standard practice, I used the Mathematica version to export state space matrix elements in Matlab which I incorporated into the existing Matlab double model - if the flag pend.db is defined and non-zero, the ssmake2MBf.m file uses matrix elements for TMTS, otherwise it defaults to OMCS. The double model lives in the SVN at ^/trunk/Common/MatlabTools/DoubleModel_Production . I suspect there might be a tiny error in the way I allow for the wire stiffness in the Matlab arising from the tricky double toe-out of the bottom wires - the relative error between Matlab and Mathematica mode frequencies is better than 0.00007, but with other models I've routinely gotten 1E-12 or better. The exported matrix elements are correct because when used with no wire flexure correction (pend.stage2=0) they agree at the 3E-14 level with the Mathematica "Stage0A" results (also without wire stiffness). Any problem must be in the ssmake2MBf.m file. I'll revisit this as I have time and inspiration. * I got Szymon to adapt the OMCS plotting scripts for TMTS ( ^/trunk/TMTS/Common/MatlabTools ) and take some test data (^/trunk/TMTS/H1/TMSY/SAGM1/Data/). Jeff K also helped. An issue we grappled with is that the Mathematica coordinate system is traditionally with +x normal to the plane of the top wires, whereas the natural coordinate system for use in MEDM screens has +L (longitudinal) in the cavity direction. So we put in a 90° rotation at a strategic point in the code. * I started separate parameter sets for the first article and production builds: ^/trunk/Common/MatlabTools/DoubleModel_Productiontmtsopt_firstarticle.m etc). * I quizzed Ken Mailand for mass and MOI data for both first article and production builds. For ease of reference, I rearranged what he sent us and posted it at https://awiki.ligo-wa.caltech.edu/aLIGO/Suspensions/OpsManual/TMTS/Models . * I got most of the rest of the parameters using the measure tool on the eDrawing at D0900419 for the first article. (The production version may well be different, but we haven't given it any attention yet.) To get the top mass d's (d0 and d1) I assumed the COM was 1.401" above the bottom surface of the bottom plate. To get the telescope d (d2) I assumed the coordinate origin was level with the round bars (where the marker seemed to be in the screenshot). This could very well be a few mm off. * I put in blade lateral compliance the same as for the UIM blades in the quad model. * With all of the above, the fit to TF data was pretty good except for the "R" (Mathematica "pitch") DOF. See the attachment at https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=6066 , especially p4. The fundamental R mode is predicted a bit too low, and the highest R mode is predicted a bit too high. * To fix the fundamental R mode, I added a tweak of +4.3 mm to d1 (the blade d). This is plausible because that's what you would get if the blades were sitting a bit lower than expected, but the mismatch could also be due to stiff cabling. See https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=6239 . This tweak doesn't make any other modes significantly different, either better or worse. So the top R mode is still about 10% off, and I haven't been able to account for it in terms of plausible perturbations to parameters. * A summary of the current best parameter set is on the aLIGO wiki at https://awiki.ligo-wa.caltech.edu/aLIGO/Suspensions/OpsManual/TMTS/Models/20130501TMTS_FirstArticle . Parameter values with comments giving sources are listed at the bottom. In particular, the flexure corrections are flex1 -> 0.00450482 and flex2 -> 0.00870794. Updated versions will be linked to from https://awiki.ligo-wa.caltech.edu/aLIGO/Suspensions/OpsManual/TMTS/Models .
This entry is just for a record.
I took a look at the ALS fiber path on the PSL table since I happened to be there for some other works today. The drawing below shows a power budget of the current fiber setup.
Started reworking on EY ALS table.
Green beam was injected into the chamber and PZT servo was enabled to center the beam on QPDs. Whitening gain of QPDs were very high again so we had to decrease the servo gain by a factor of 100. Whitening gain MEDM screen was all white so we left it as is.
After confirming that the beam was centered on all four irises, we started removing the Faraday from Drever lab and realigning.
Some observations:
We'll level the beam path and start over tomorrow though we don't know if doing so will improve the transmission quality.
I was hoping this (and power cycling the end link chassis) would get rid of errors on the bridge terminal, but there are still errors on the bridge terminal even in the init state.
GregG & Hugh We put the eight HEPI actuators of HAM2 in bleed mode and opened the 4-way valves and then ramped the fluid pressure back up to max. We did this 4 times to be cautious about fluid leaks--we did this at each pier and carefully checked for leaks before ramping the pressure back down and opening the valves at the next pier. The introduction of additional resistance has us at 28psi now with the motors running about 92% and the servo on. We will continue these steps at WHAM3 and possibly HAM1 and/or BSC1 over the next few days. Additional chambers will allow us to get higher pressure and so more nominal performance. While the HEPI is locked, we did see some readout change on the IPS. The signs are all the same and the largest is about 300 counts or ~0.0005". I suspect this is mechanical distortion of the Actuator itself with the pressure introduction. Since the IPS is all attached to the actuator both the Flag (unsuspended) and Sensors (suspended), I don't think this can be checked independently. I would venture that the Platform has not actually moved. After a couple days of bleeding, we'll switch the Actuators to run mode. This will likely change the pressure/IPS state again so hold on! We'll likely want to have enough Actuators on line so that we can servo the pressure at nominal level(80psi, 70diff) so the introduction of additional actuators will not change the actuator distortion/IPS 'zero'.
These are the TF/Spectras Test Results for HSTS I1-PR2
These measurements presented above are in excellent agreement with the model and consistent with other HSTS's tested, therefore PR2 is approved for LIGO India. Great Work!
[Dick, Kiwamu]
We took a look at the IMC demod phase shifter [1] which had been found to be out of Beckoff's control the other day last week [2]. The conclusion is that it magically recovered by itself and became under Beckoff's control again. Anyway we set the delay by the hardware switches and disabled the digital switching as this is relatively safer than setting it digitally. There should be no impact on the IMC locking.
The optimum delay setting
The delay is set to be 14.4375 nsec (= 1/16 + 1/8 + 1/4 + 2 + 4 + 8 nsec)
where the I signal is maximized. This corresponds to a step setting of 231=(1+2+4+32+64+128) in the digital system. Note that the IMC modulation frequency is at 24078360 Hz.
The Symptom
The symptom we saw last week was that the delay time didn't change when we changed the digital values in the corner Beckoff. I don't know what had happened to the electronics and digital system. Somehow we were able to change the delay from the Beckoff today.
The right half of the device is not functioning
This is not important. A sad thing we found is that the right half of the delay line device doesn't seem working for some reason. We are not going to fix this because we haven't been using it and also we are not going to use it. Dick found that the loss of the circuit heavily depends on the amount of the delay and sometime it can reduce the ~15 dBm input signal down to -20 dBm or even less at the output port. We are guessing that there is some kind of loose contact in the circuit.
[1] DCC S1103425
[2] LHO alog 6265 "IMC locked, relocking well, visibility is good, but not TRANS beam, as it is clipped by it's bellows."
This was reported to Bugzilla as bug99.
https://services.ligo-wa.caltech.edu/integrationissues/show_bug.cgi?id=99
