Laser Interferometer Gravitational-Wave Observatory

LIGO Caltech 40 meter prototype Upgrade

WEEKLY REPORTS ARCHIVE

May 27, 2004

• Osamu, Bryan, Seiji and Alan have been working to confirm and check the observation that the "sidebands on sidebands" applied by two EOMs in series cause a large contamination of the arm signals (CARM and DARM) onto the short dof double-demod signals (PRC, SEC, MICH), severely complicating lock acquisition and control of the Advanced LIGO optical configuration. This is documented in a set of ppt slides from Osamu, his SPIE paper, and a detailed analysis by Bryan Barr.
• Bryan was able to get reasonably good separation between the arm and short degrees of freedom by careful choice of demod phases, at the expense of large offsets and exceedingly tight demod phase tolerances. Not a practical solution.
• Fortunately, a practical solution is easy to implement: a Mach-Zehnder interferometer on the PSL table, with one EOM in one arm and another in the other. This requires a control system to lock the M-Z via internal modulation (two beamsplitters, an RFPD and demod board, servo filter, pzt mirror actuator), all of which are being acquired now. A very preliminary draft design by Osamu, Bryan, Seiji and Steve is here. It includes an analysis of the frequency and intensity noise introduced by the M-Z; we think it will be negligible. We hope to assemble the system in the next week or 2.
• Osamu will re-measure the PSL frequency noise and intensity noise, before and after implementing the M-Z.
• The PSL, mode cleaner, and FPMI interferometer all continue to lock well and stably. All oplevs and QPDs are centered and the spots come back to where they were before, between locks (we don't leave the FPMI locked routinely).
• Jay reports that the LSC rack cross-connect wiring changes for the double-demod boards are done, all the modules are in hand, and code for the new lsc auxiliary crate 2 will go in next week.
• Rolf is ready to put the new front end LSC software in next week. Rob is working on a first draft of the InputMatrix40m code.
• Rob is also thinking about machine-learning methods for finding the ideal demod phases for double-demod length sensing.
• Ben has been working on the LSC/ASC RF distribution drawings. These drawings are intertwined with Jay's LSC drawings, and usher in a new wiring for the 1x2 rack. This rewiring will be accomplished by Tuesday, along with the cabling of the cross-connects to the modules. This will make the hardware be in place in time for the software's arrival.
• Osamu found a problem with an LSC PD Interface board. Ben took it back to the shop for troubleshooting. He fixed it and it now works fine. There was a bad logic chip that disabled the 1st two PDs.
• Bob has installed heliac cables for all long RF runs.
• Osamu continues to work on the frequency stability servo, implementing some changes that he has planned for a long time.
• Flavio and Peter King have been working on the ISS for the 40m PSL, implementing changes to the servo filter and the current shunt actuator that were developed on the LASTI PSL; the block diagram is here. Preliminary results look very nice. The changes to the current shunt caused the power of the PSL to drop from 10.8 W to 10.4 W.
• The readout board for the TRY QPD is not working; Ben will look at it.
• Ben Olsen and Steve have been diagnosing problems with the STACIS seismic isolators. There is one bad controller (of 4), and at least 3, maybe 4, bad isolators (out of 12). At least one shows evidence that the PZT stacks are burnt out. They will continue to diagnose the problems, before we start sending pieces of the system back to TMC for repair.
• Steve had two Tropel optical spectrum analyzers converted from green light to 1064; they are now back and ready for use. A third Tropel is in continual use monitoring the light transmitted through the mode cleaner.
• Fumiko has implemented a simple servo filter for the oplev P/Y servos, and has taken transfer functions at different gain settings. She finds that the pitch pendulum Q can be reduced from > 10 to < 0.9 before oscillation sets in. The continues to optimize and document the system.
• Shihori continues to work on developing a procedure for diagonalizing the suspensions' output matrix.
• Monica Varvella of Orsay has spent the last few weeks here at Caltech learning how to run e2e, build models, understand the Han2K and SimLIGO models, and thinking about the modifications required for AdvLIGO and the 40m. She will also work on modeling the Mach-Zehnder, as a simple and useful exercise. She leaves tomorrow, but will return for the month of July.
• Bob is busy with many bake jobs for Janeen and the triple pendulum MC crew, and for Larry Jones.
• Bob is baking out and scanning the new squeaky-clean bake oven, in the hopes of achieving the AdvLIGO cleanliness specs.

May 20, 2004

• Osamu and Bryan have uncovered what we believe to be a fundamental design problem in the Advanced LIGO length sensing scheme which the 40meter has been working towards testing. The design calls for two pairs of RF sidebands, f1 and f2 (for AdvLIGO this might be 9 & 180 MHz; for the 40m, it's 33 & 166).
  The CARM and DARM degrees of freedom are sensed at f1 and f2, respectively, while the short dofs (PRC, MICH, SEC) are sensed at f2-f1, f2+f1, or both ("double demod"). The idea is that the short dof sensing is derived only from the sidebands and is undisturbed by the arm-amplified carrier phase shifts.
  We have modeled this scheme using Twiddle and Finesse, and it agrees with the matrix presented by Peter Fritchel in an LSC meeting back in 2000 or so.
  We have implemented these two pairs of sidebands at the 40m using two EOMs in series. This produces "sidebands on sidebands", ie, local oscillators at f2-f1 and f2+f1. These were apparently neglected in all the modeling. They exist, and they can beat with the carrier at f2-f1 and f2+f1. This is not a negligible effect; to the contrary, it completely swamps the signals for the short dofs. Osamu has confirmed this using a modified FINESSE model. Osamu's results are here.
• It turns out that Bryan was unable to lock the Glasgow 10m RSE interferometer for exactly this reason. Bryan's summary of the Glasgow work is here.
• We are considering several possible solutions to this problem:
  • eliminate the sidebands on sidebands by placing the two EOMs in parallel, using a Mach-Zender
  • eliminate the sidebands on sidebands by employing 2 more EOMs to kill f2-f1 and f2+f1
  • Generate f1 and f2 but not f2-f1 or f2+f1 by sending an appropriately tailored RF signal into one broad-band EOM
  • find a different combination of demod phases and other signal processing to allow us to extract clean signals for PRC, MICH, SEC with the existing EOM configuration (it is not clear that this will be successful, but we will try).
  • some more radical solution which might require changes to the interferometer or the length sensing optics and electronics.
• Dave Tanner & Dave Reitze have experience with complex modulation of broadband EOMs, and they say that we can indeed reduce the sidebands on sidebands by, maybe 10 dB. We need more like 20 dB suppression, as well as a system to monitor and actively kill the sidebands.
• Bryan and Osamu have started looking into the finesse model in a bit more depth to see if we can reduce the effect of the arm cavities by tweaking the demod phasing and other such tricks. It looks as if it might be possible if we're not too fussy about offsets (at least initially) but needs some more in depth examination. If it is possible then the diagonalisation of the central part will not be as good and the problem of lock acquisition will still be just as bad due to large signals in the central part signals around the arm cav carrier resonances - only cancels at the operating point as far as I can see. We need to look at this a bit more but it is unlikely to change the conclusion that we need to use a Mach Zender.
• The best course for now seems to be to use a Mach-Zender interferometer because it has no sidebands of sidebands from beginning. Frequency noise by a doppler shift will be considerable, but it should be of the same order as one produced by the other mirrors.
• To control the MZ, Osamu asked Jay to get a PZT driver like PMC which has a fast response up to ~100kHz. Internal modulation using 33MHz should work well to get signal to control the MZ. We will probably put a pick off BS for parallels EOMs just after PMC, so MZ won't change the path length.
• Osamu is writing a proceeding paper on the 40m program for SPIE. The effect of the sidebands of sidebands is already included. A draft can be found here.
• Rolf has prepared new LSC software for the 40m dual-recycling configuration. Some more work is required before it can be installed; Rolf will return to this after his 2 week sojourn to LLO. New MEDM screens and the EPICS sequencer are in progress.
• For lock acq, this code expects to include InputMatrix40m.c; the development of this code is the subject of much active thinking by Rob and everyone else in our group.
• Ben has delivered two 33MHz Wavefront Sensors to the lab, and is getting two boxes for them from the machine shop. These complete the PD requirements, and join the other recently delivered QPD, double demod PDs, and 66MHz LSC PDs at the lab.
• Don has been installing the cross connects for the rewiring of our LSC rack to accommodate the double-demod signals.
• Ben has traced and characterized the PSL fast channel DAQ readback paths. To the design, they are being treated properly by the interface card. The next step is to see if the gain that was dictated by the design is appropriate for the signal levels that we have here.

May 13, 2004

• We did not hold a 40m weekly meeting this week; AJW is out of town. This is an incomplete weekly report.
• Osamu and Bryan have discovered that the length sensing modeling done with both Finesse and Twiddle share a common bug: they don't simulate the sidebands placed on the sidebands by the two EOMs operating at 33 and 166 MHz. These sidebands will change the error signals dramatically. Under intense study.
• All core suspended optics are now well aligned, including the two recycling mirrors.
• Osamu and Steve have used a Tropel optical spectrum analyzer to look at the light tranmitted through the mode cleaner, and then out the asym port. The 33MHz and 166MHz sidebands can be seen clearly at both locations, with expected magnitudes.
• The double-demod (DD) signal at the sym port has about 10mV with maximum modulation index and 15% transmission of SP EOS. The double demod signals generated by 133, 199MHz demod boards are going to whitening filters and they can be seen by dataviewer.
• We expect to see an offset in the DD signal due to imperfect demod phases (from Seiji's and Osamu's calculations) but it is not seen at all. Osamu and Bryan tried to change the phase of 133MHz and 199MHz local oscillators (using different cable lengths; the digital controls are not yet available) to see offset change, but saw no offset. This may be due to the Finesse bug referred to above. Under study.
• Jay has completed drawing up the mods to the length sensing and control (LSC) wiring to accommodate the double-demod LSC photodiodes. He has compiled a list of wiring changes, and work on these is commencing.
• Rolf, Jay, Ben and Alan met to discuss the LSC software required for dual-recycling lock acquisition. New software is in progress and should be ready for initial testing early next week.
• From Rolf, on his new LSC code: Added another Linux kernel module to do linked list DMA on the new VME P4 processor boards. This allowed me to port the present LSC software over. The LSC code under Linux on the new CPU board runs in 28usec (Pentium 3 code at sites runs in 49usec). This LSC code is now being modified for the 40m lab (12 input photodiodes, 8 DOF, and dither locking).
• Alex got the new CPUs to boot Linux over the network (diskless node). Therefore, we no longer require direct connection of a disk drive.
• While Rolf's initial intention was to change over the 40m lab DAQ system to be consistent with the sites at the same time (makes LSC code easier), he is opting to defer that until later.
• Bob has been working on the RF cabling for the LSC photodiodes. All RF signals at 133, 166, and 199 MHz, and also long 33 MHz runs, will be carried by heliax cables. Bob has installed some of these and is ordering the heliax for the remainder; in the meantime, he has installed temporary coax cables.
• Ben is acquiring two WFS photodiodes for the 40m arms. He brought two boxes to the machine shop they should be back soon.
• Ben continues to chase down problems with the PSL fast channel DAQ readback. A connector on the back of the Generic DAQ Interface card was out of position, and has been fixed. This accounts for some signals not lining up with their software names. Now the next step is to do a survey of the incoming signals, to see if they are being treated properly by the interface card.
• Bob cleaned and baked a load of small parts for Janeen.
• Bob bonded and baked (first bake) 10 Hybrid OSEMs, and cleaned and started their hot bake.
• Bob started wrapping the new oven with heating tape and he will switch the SRS RGA from oven B to the new oven.

May 7, 2004

• Bryan Barr has arrived from Glasgow, and Monica Varvella from the Orsay VIRGO group. They will focus on dual recycling lock acquisition strategies at the 40m and, via e2e simulation, AdvLIGO. Please welcome them!
• Commissioning (Miyakawa, Vass, Ward, Kawamura):
  • Osamu locked the power recycling cavity (PRC) along with MICH, for the first time. He used SP 33I for PRC and AP 166Q for MICH. This is a good check that the length control path to the power recycling mirror is working correctly.
  • Osamu is working on establishing signals with the double-demod (DD) photodiode, at 133 and 199 MHz. The signals are very small. He believes that the DD PD amp should be modified by notching out the 33 MHz, which otherwise swamps the DD signals. Jay will look into this.
  • Osamu is measuring the 166 MHz at the AP using an optical spectrum analyzer.
  • Seiji, in Japan, is developing methods to set the double-demod phases correctly for the PRC, MICH, and SRC signals.
  • Rob continues to study and understand the lsc front end code, and is talking with Rolf about how he can help develop it.
• PSL (Miyakawa, Vass, Ward):
  • Osamu found a 10 kHz oscillation on the frequency stability servo (FSS) error signal, and traced it to a back-reflection of 30-40% of the light from the AOM. He adjusted the 1/4 wave plate near the AOM, and eliminated the back-reflection and the 10kHz oscillation.
  • Osamu then found a 100 kHz oscillation on the FSS error signal, and traced it to a bad op-amp which was put on the FSS card to cancel the PMC pole (after we moved the FSS pickoff to be after the PMC). He and Ben replaced the opamp, and the 100 kHz is gone.
  • Osamu then found a 58 kHz oscillation on the FSS error signal, traced to a resonance of the PZT on the new NPRO. The FSS card has a notch at 71 kHz; with the new NPRO, this needs to be moved by changing a capacitor. Osamu and Ben will do that soon.
  • There remains a 500 kHz oscillation on the FSS error signal, whose origin is unknown; it may come from the RFPD amplifier. But, it's small and has a small effect on the performance of the FSS.
  • Osamu thinks that the FSS PZT loop filter has one more pole at 10kHz than it should. He will confirm with Rich Abbott, then remove the extra pole.
  • After all this FSS work, the PSL is very stable and performing as well as ever. The mode cleaner is locking robustly, and staying in lock for many hours at a time.
  • Osamu will remeasure the FSS OLTF and use the MC to re-measure the PSL noise, and compare with what we had 6 months ago.
  • We intend to adjust the PSL slow loop temperature change in the EPICS state code.
• Optical sensing (Kawazoe, Sakata, Ward):
  • Fumiko has installed digital filters on all the oplev servos, and is taking noise spectra with the servos off and on. She sees some significant in-loop noise reduction in the 0.5-1.0 Hz range, but sees little or no change in the OSEM P/Y noise spectra. She's also taking swept-sine transfer functions with one or 2 optics. She's finding several problems, and is not sure whether the problems are with FOTON or something else. Under investigation.
  • Steve is ordering replacement mirrors for two tropel optical spectrum analyzers that were left over from the old green-light 40m days.
  • Steve & Bob set up a video camera to view the area around the MC2 chamber.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • The group is reviewing Jay's drawings for mods to the LSC rack to accomodate the new double-demod signals.
  • Steve has ordered RF splitters for distributing the RF local oscillators to our long list of LSC demod boards.
  • Steve, Bob, and Rob set up remote controls for the picomotors that dial in beam blocks in front of the four test masses. All 4 beam blocks are visible on the video cameras viewing the test masses.
  • Ben, Jay and Rolf looked at our inability to read out the fast PSL channels correctly into DAQS. They traced it to an ICS110b ADC that was not in synch. They rebooted that crate and it appears to work ok now. Ben will go through the channels to see what gains they need on the DAQ interface board to keep them in range.
  • Rolf says that he will have time in the next couple of weeks to get our new LSC code working. He will run it in a new Pentium IV vme Cpu. Since he doesn't have a VxWorks BSP for this board, he will try to run it on realtime linux.
  • Rolf also plans to upgrade our DAQS system like he did at the sites, including a new frame builder and better handling of channels at different read rates.
  • Undergrad Ben Olson is taking transfer functions of all 12 STACIS systems and is finding that most of them are working well. He's trying to isolate and diagnose the problems.
  • EPICS code to shutter the PSL during vacuum transitions will be implemented soon.
  • Mohana and Shihori continue to work on setting alarm limits on all EPICS channels in the EPICS alarm handler.
• Facility (Vass) and South Annex Bake Ovens (Taylor):
  • Bob is vacuum-prepping OSEM clamps and LASTI MC parts for Janeen.
  • Bob is installing thermocouples on his new super-clean bake oven. He's working with Larry Jones on getting the noise down on RGA scans, etc.

April 30, 2004

• The 40m lab welcomes two visitors on Monday: Bryan Barr from Ken Strain's group in Glasgow, and Monica Varvella from the Orsay VIRGO group. They will focus on dual recycling lock acquisition strategies at the 40m and, via e2e simulation, AdvLIGO. Please welcome them!
• Commissioning (Miyakawa, Vass, Ward, Kawamura):
  • Work on the PSL, and on the LSC electronics, has trumped any experimental progress in dual-recycling lock acquisition this week.
  • Seiji has written a technical note documenting his proposed dual-recycling lock acquisition sequence, which features dither-locking of MICH as a first step. This document is now under review by the 40m group.
  • Rob has been studying the lsc front end code all week, and is starting to feel comfortable with modifying and maintaining it.
• PSL (Miyakawa, Vass, Ward):
  • Osamu has been struggling with the frequency stability servo (FSS) all week. Our new NPRO (SN 238) is much noisier than the old one (SN 332), forcing the FSS to work hard.
  • He re-aligned the entire FSS optical path, but still the FSS was not staying in lock robustly.
  • He found some design flaws in the FSS board, starting from rev D (our board says "rev C" on it, but it looks like it's rev E or F). He has fixed these on our board, and is working with Rich Abbott to generate a new rev (drawing to DCC) for this board.
  • He found saturation in the PC loop for the FSS, and changed some gains to fix it.
  • He sees ~ 10 kHz oscillation and saturation in the PZT loop for the FSS. He changed the dynamic range by a factor 3, which greatly reduced, but did not eliminate this problem.
  • He noted that the FSS EPICS gain sliders don't do what they're supposed to do. It seems to be an EPICS database problem; Ben and Osamu will diagnose and fix it.
  • After all these changes, the FSS is now performing much better, and the FRC and MC lock stably for hours. It's still not perfect: the 10 kHz oscillation still appears sometimes.
  • But just now (Thursday afternoon) Osamu found the source of the oscillation: the AOM QWP was out of adjustment, and much of the light that was supposed to go to the FRC was returning to the laser, introducing noise and optical feedback. He has fixed this, the 10 kHz oscillation is gone, and the FSS appears to be as stable as ever.
  • Osamu will remeasure the FSS OLTF and use the MC to re-measure the PSL noise, and compare with what we had 6 months ago.
  • We would like to get the old NPRO (SN 332) refurbished and made available as a hot spare in the 40m lab.
  • Jay reports that Flavio and Peter King will want to look for a 2.5 MHz peak in the ISS current shunt actuator response function, that has been seen in the LASTI PSL.
  • We intend to adjust the PSL slow loop temperature change in the EPICS state code.
• Optical sensing (Kawazoe, Sakata, Ward):
  • Fumiko continues to study and understand the oplev servo filters employed at the sites, and how to modify them for the 40m. She will talk to oplev servo experts.
  • Bob and Steve are installing the remaining ccd video cameras on beamlines, and cables to the video switch. They checked all cameras using the laser beam and/or flashlights. Mods to the video switch programming and the EPICS screens will be done by Lisa.
  • Steve will attempt to adjust the camera views so that the beam flags (in front of the 4 test masses) can be seen.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • Ben got the three RFPDs for double-demod from Todd; they are tested and ready for installation. The 66 MHz RFPD for SPOB will be ready by the end of the week.
  • The large fluctuations in the MOPA output power, reported by the EOT photodiode in the MOPA box, are now believed to be due to improper grounding of the photodiode signal to the ADC (VMIC 3123). Ben fixed this, and will do the same for the NPRO power monitor.
  • Ben is working on a switchbox to control all the picomotor beam flags from the control room.
  • Jay reports that drawings and database changes for signal recycling have been started and should be complete by Monday. These are mods required for our existing length sensing (LSC) rack, to accomodate more RFPD signals. It looks like it will be a substantial piece of work, and an additional aux crate is required. We will pursue this work with the highest priority, followed by a serious search for 60 Hz pickup.
  • Bob installed heliac cable for the LSC RFPDs in the POB and AP beamlines.
  • Ben has been in email contact with Vern regarding the RFPD redesign. Things seem to be coming along, and when Vern's design is reviewed, Ben can continue on the production.
  • Ben will check that perennial problems associated with rebooting (XYCOM boards not set right, EOS state not right, polarities for mechanical shutter toggles set right) are now being automatically done right.
  • Steve and Ben Olson are diagnosing the STACIS problems prior to a visit to the 40m lab by TMC reps on Monday.
  • EPICS code to shutter the PSL during vacuum transitions will be implemented soon.
  • Mohana and Shihori continue to work on setting alarm limits on all EPICS channels in the EPICS alarm handler.
• Facility (Vass) and South Annex Bake Ovens (Taylor):
  • Bob is in the middle of the first of a long list of bake jobs for Janeen, for the AdvLIGO MC triple pendulum suspension.
  • Bob is assembling and vacuum-prepping LIGO I osems for Janeen and for the 40m.

April 23, 2004

• Commissioning (Miyakawa, Vass, Ward, Kawamura):
  • Work towards implementing a full LSC system for locking the dual-recycling interferometer is progressing rapidly; see below.
  • The mode cleaner servo is performing less robustly. Osamu thinks that the demod phase has somehow changed (maybe when RF cabling was done in that rack a couple of weeks ago), and will re-tune the demod phase with different-length cables.
  • Rob continues to study the deep complexities of the front-end LSC code, with the aim of modifying it for dual recycling.
  • Seiji is writing a technical note summarizing his studies of dual-recycling lock acquisition, in which he concludes that dither-locking MICH can provide a robust path towards lock of the full IFO.
• PSL (Miyakawa, Vass, Ward):
  • Our MOPA continued to misbehave in new and strange ways. Ever since Steve & Osamu turned down the current on the NPRO (to reduce the overheating), the MOPA power monitor has been fluctuating *upward* from 11 watts up to 13 or 14 watts. When that happens, the mode cleaner falls out of lock.
  • Steve and Rob obtained an NPRO heat sink from the shop (ordered by Peter King weeks ago) and installed it on the NPRO in our MOPA laser. They also removed the cover from the NPRO. The TEC temperature monitor fell to 1 volt, the MOPA power stopped fluctuating, and the mode cleaner held lock for hours. Late last night, there was some more MOPA power fluctuation, but it hasn't happened since.
  • In the next few days, we will close the MOPA cover (it is now propped open a few inches) and declare victory on the heating problem.
  • However, there are still problems with the FSS. There appears to be a big oscillation (3MHz) in the FSS fast PZT loop which is limiting the gain and causing the mode cleaner to fall out of lock. Osamu burned the card out while testing it, by plugging it into a bad extened board. Ben fixed the FSS board this morning, brought over a new extender board, and checked that the FSS board works ok. But it still has the big oscillation. Osamu thinks it is an electronics problem that can be fixed. But he's also checking and tuning up the optical path to the reference cavity.
  • Osamu and Rob found that the EPICS sliders for the FSS common gain and fast gain do not work as they should. They will enlist Ben to help find and fix the problem.
  • We intend to adjust the PSL slow loop temperature change in the EPICS state code.
• Optical sensing (Kawazoe, Sakata, Ward):
  • Fumiko continues to work on implementing an oplev servo filter based on what she found at the sites (modified with FOTON to be more appropriate for our 5" optics).
  • Shihori is documenting her commissioning of the electro-optic shutters.
  • Steve will attempt to adjust the camera views so that the beam flags (in front of the 4 test masses) can be seen.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • Ben has been accumulating the remaining hardware needed for length sensing for dual recycling, including: two more double-demod RFPDs, another 33 MHz RFPD, a third LSCPD interface board, a fourth whitening board, and a fourth anti-aliasing board.
  • Jay is working on mods to the LSC rack drawings to get the wiring for these additional boards.
  • Ben continues to work on the redesign of the RFPD amplifier for fast RF and low power.
  • The RFM network went down on Wednesday at around 6am for unknown reasons. Rob spent much of the day trying to get it back up (by rebooting the vme cpus) with no success. Jay determined that the problem was that the c1epics linux cpu which connects EPICS with the front-end servos was not performing its task. He restarted the task and everything else could be brought up by the end of the day. Thanks, Jay; this is going into the procedures manual!
  • Ben will check that perennial problems associated with rebooting (XYCOM boards not set right, EOS state not right, polarities for mechanical shutter toggles set right) are now being automatically done right.
  • Steve has had no luck locating our STACIS servo extender board, and TMC wants $1K for a new one. We'll see if we can borrow one, and/or attempt to optimize the servo gains without it.
  • Steve located some old picomotor controls which work with our picomotor-driven in-vacuum beam flags, and he and Bob have run cables so that all beam flags can be controlled from the control room. Ben modified a switch so that it worked with this system.
  • A controller for a uniblitz shutter that was non-functional out of the box will be fixed by the manufacturer at no charge.
  • Bob ordered heliac cables and connectors for the long fast RF runs, and they are on their way.
  • EPICS code to shutter the PSL during vacuum transitions will be implemented soon.
  • We plan to do a careful search for 60 Hz pickup in our LSC crate, which is anomalously large.
  • Mohana and Shihori continue to work on setting alarm limits on all EPICS channels in the EPICS alarm handler.
• Facility (Vass) and South Annex Bake Ovens (Taylor):
  • Bob plans to deliver 8 LIGO-I OSEMs to Janeen (and spares for the 40m) by the middle of next week.
  • Bob is finished plumbing up new "clean" oven C, and is pumping on the RGA line.
  • Bob has decommission "dirty" oven D and surplused it. It is now removed from his lab.
  • Bob is exploring building a new very large bake oven for AdvLIGO.

April 16, 2004

• Commissioning (Miyakawa, Vass, Ward, Kawamura):
  • Osamu is preparing to lock the PRC with double-demodulation (DD) at the SP. A DD RFPD is in place, the DD will be done with a pair of IQ Demod boards, and the LO RF signals are in place (at 166+33 and 166-33 MHz). We need to do some wiring of the LSC rack to get the signals into the LSC system, and are now looking into the needed software changes, if any.
  • Jay, Osamu, and Rob are going over the requirements for LSC hardware and software for the full dual-recycling lock acquisition, and Rolf knows that we're thinking about it.
  • Rolf and Alex are thinking about implementing a general-purpose "math block" in the digital controls software, which we could use for dither-locking the PRC.
  • Locking of the Fabry-Perot-Michelson (FPMI) is currently hampered by a (new) problem with the PSL (see below). It can be done, but there is apparently a problem making the transition between locking of XARM/YARM and CARM/DARM that is not understood and is under investigation.
• PSL (Miyakawa, Vass, Ward):
  • The PSL is starting to misbehave again, in a new way. The frequency reference cavity (FRC) error signal is oscillating in a funny way, suggesting saturation. Osamu suspects the FSS fast (PZT) loop. Under investigation.
  • Steve & Osamu decreased the PSL NPRO #238 power output from 712mW to 504mW by adjusting the doide current potentiometer from 2A to 1.7A. DTEC fell from 1.7 v to 1.25 v (that was the point; the PSL trips off when DTEC exceeds 2 v), and the MOPA power fell from 12.1 W to 11.0 W. The PSL has been performing stably since then. We think we're out of the woods with regard to heating, but we still await an NPRO heat sink from the shop. That will hopefully arrive and be installed next week, at which time we will fully close the MOPA cover and hope that the heat (DTEC voltage) doesn't rise unacceptably.
  • We intend to adjust the PSL slow loop temperature change in the EPICS state code.
• Optical sensing (Kawazoe, Sakata, Ward):
  • Fumiko has tried to implement an oplev servo filter based on what she found at the sites (modified with FOTON to be more appropriate for our 5" optics). She was unable to see a reduction in P/Y noise (as measured by the OSEMs). She will pursue this work systematically.
  • Rob has finished laying out the POB beamline, and is hooking up the electro-optics: a ccd camera, an RFPD, and a QPD.
  • We now have 12 QPDs installed and being read out: IPPOS, IPANG, IOPOS, IOANG, MCR, MCT, TRX, TRY, POX, POY, AP, SP. We need one more, POB, and an additional readout channel. Ben has delivered the QPD itself; we have to either instrument a new readout channel or borrow POY's.
  • Shihori has aligned all the QPDs except for POB, and TRX and TRY. She will do those next.
  • We now have 3 electro-optic shutters (EOS): at MCR, SP, AP. The MCR one has been working for many months. Shihori and Osamu got the SP one working, and Shihori measured the transmittance vs HV; a nice cosine curve. She will document it. There are problems with the one at the AP, which may be due to bugs in the controller. Under investigation.
  • We now have 4 mechanical shutters under EPICS control (PSL, MCR, AP, SP) and 4 under manual control (MCT, POB, POX, POY). One shutter is not functioning and has been returned for repair/replacement.
  • EPICS code to shutter the PSL during vacuum transitions will be implemented soon.
  • Osamu, Steve, Rob and Shihori have cleaned up all the optical tables, removing all excess optical components for storage, and ensuring that all beamlines are fully instrumented and laid out as in the drawings.
  • The PSL table is now grounded to the vacuum enclosure, as are all other optical tables.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • We now have Marconi signal generators for all required frequencies (29.5 MHz for MCR, and 33, 166, 133, and 199 for the main interferometer, all phase-locked to multiples of 33 MHz).
  • Ben brought over LSC RF distribution boxes for the 133, and 199; not needed for 166, only one IQ demod board needs it. There are heliac cables for the 33 and 166 to the LSC rack, and Bob ran some temporary RG405 for the 133 and 199; more heliac cable is on order.
  • At Osamu's request, Ben removed (and jumpered across) the output attentuators at the end of the RF AM stabilizers that go to the EOMs. Osamu measured the maximum modulation depth for 166 MHz to be 0.3; more than enough, and easily turned down to 0.1 with the EPICS slider. The 33 MHz is probably much deeper, but will also be turned down to 0.1.
  • The 166 MHz and 133,199 MHz RFPDs show oscillation at 500 kHz. This is a known problem for RFPDs with low light level, with a known fix, and Ben will implement it.
  • Ben cured the long-standing problem of the varying monitoring readout of a couple of Thorlabs photodiodes on the PSL (PMC transmitted DCPD and FSS transmitted DCPD), by tying the "Guard" pin to the signal "Lo" pin (discovered after reading the 96 page manual for the 3123 ADC). He will now investigate the PSL DAQ channels, some of which have been railed forever.
  • Ben continues to work on the redesign of the RFPD board, with Vern Sandberg. They are working to incorporate a few new features that would be nice to have. Ben plan to give Dennis a schedule of development by the end of this week.
  • Ben is updating the system wiring schematics for the 40m Auxiliary rack, in order to reflect the as-built condition.
  • We plan to do a careful search for 60 Hz pickup in our LSC crate, which is anomalously large.
  • Mohana and Shihori continue to work on setting alarm limits on all EPICS channels in the EPICS alarm handler.
  • Ben Olson continues to diagnose problems with the STACIS seismic isolators. Two of the 3 isolators in the south end are working well, the NE isolator is behaving badly. Under investigation. Steve will locate the STACIS servo extender board or order a new one.
• Dual-recycling lock acquisition (Kawamura, Miyakawa): Seiji and Osamu have settled on two lock acquisition sequences (using double demod) that they think will work robustly. Seiji's involves dither-locking MICH then locking the PRC and then SRC with double demodulation at the SP and PO, respectively. Osamu's replaces the first step with an alternative scheme for MICH. We will work towards implementing these schemes. This is a perfect thing to test with e2e... new visitors are arriving in May who may want to work on this.
• Facility (Vass) and South Annex Bake Ovens (Taylor):
  • Steve is cleaning up the 40m lab, moving excess optics and equipment to cabinets in the east arm, and getting rid of surplus equipment such as old 40m racks full of obsolete electronics.
  • Bob tied down the monitors in the control room for earthquake safety.
  • Bob is plumbing up new "clean" oven C.
  • Bob will decommission "dirty" oven D, will surplus it, and will recommission old oven C as a new "dirty" oven D.
  • Bob is exploring building a new very large bake oven for AdvLIGO.
  • Bob is building OSEM diodes for Janeen (AdvLIGO MC triple) and for the 40m.

April 9, 2004

• Commissioning (Miyakawa, Vass, Ward, Kawamura):
  • Osamu and Rob locked the Michelson (l-) using the 166 MHz RFPD at the AP. They are now working on locking the PRC (l+) using double demodulation at the SP.
  • Rob and Osamu are studying Matt Evans' thesis and InputMatrix code, in order to generalize it for dual recycling.
  • Fumiko and Osamu have analyzed the MC transmitted beam using the Tropel optical spectrum analyzer. They measure a transmitted modulation depth of 0.13 for the 33 MHz sidebands and 0.06 for the 166 MHz sidebands, with all EPICS sliders set to maximum. they will work on increasing the modulation depth for the 166 MHz.
• PSL (Miyakawa, Vass, Ward):
  • Osamu re-aligned the beam from the (new) NPRO to the power amplifier. The power out of the MOPA is now up to 12.1 watts, and intensity fluctuations are now much reduced. He also re-aligned the beam into the frequency reference cavity and the PMC. The PSL is now performing very well!
  • We are still operating with our MOPA cover off to keep the NPRO from heating up.
  • Steve has ordered a heat sink on the new NPRO and will install it as soon as it arrives.
  • We intend to adjust the PSL slow loop temperature change in the EPICS state code.
• Optical sensing (Kawazoe, Sakata):
  • After tuning up the PSL, Osamu re-optimized the mode cleaner wavefront sensor beamline and digital alignment control system; it is now working well.
  • Shihori and Osamu have installed and aligned electro-optic shutters on the AP, SP, and MCR beamlines. The trigger photodiodes were wired from the auxilliary rack; this proved to be dangerous, and a couple of photodiodes have been fried. We are now running power to these trigger photodiodes using matched power supplies; SP and MCR beamlines have these, and a third will be installed soon on the AP beamline.
  • Shihori and Osamu have checked out 12 of 13 beamline QPDs, from photodetector through the entire electronics chain to EPICS displays. All are working well. One more QPD is required for the POB beamline. Ben is acquiring it (and a spare).
  • Rob is almost finished laying out and aligning the POB beamline. He awaits a QPD and a CCD camera.
  • Osamu and Rob have installed and tested mechanical shutters on the SP, AP, MCR, POX, POY, MCT, and PSL input beamlines. EPICS control of the SP, AP, MCR shutters have been tested; EPICS control of the PSL shutter will be tested soon. EPICS code to shutter the PSL during vacuum transitions will be implemented soon. Mechanical shutters for the other beamlines are controlled manually (as opposed to EPICS).
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • Jay has generated a block diagram and requirements for the LSC as it relates to signal recycling. There will be 11 LSC RFPDs feeding I&Q demod boards; a 22x8 input matrix; and an 8x8 output matrix (including a path from CM -> MC).
  • Ben continues to check out some varying monitoring readout of a couple (at least) of PSL channels (PMC transmitted DCPD and FSS transmitted DCPD) and now believes that the signals are being pulled down by a weak short to some other signal(s). He has a spare 3123 which he will test during the next PSL down time.
  • Ben helped Rob repair a particularly stubborn oscillation on one of the RFPDs. Adding the 1.5K resistor didn't do any good, so he tried the Inductor-Resistor fix, and it worked very well.
  • Ben received two signal generators (IFR 2024's) that he ordered for the LSC double demodulation, and has brought them to the 40m. Bob Taylor has made shelves to mount them between 1Y2-24, and 1Y2-28. This should make minimal impact on the existing rack drawings.
  • Todd has delivered the four I&Q Demod boards that are needed for double demodulation, and Ben brought those over today. We now have 13 I&Q Demod boards, and much work is required to wire them up for use. Ben will take the lead.
  • Bob has installed heliac cable for some long cable runs from RFPDs to I&Q demod boards, and will order and install heliac cable for all remaining long RF runs.
  • Ben is planning a redesign of the LSC RFPDs for higher frequency operation.
  • We plan to do a careful search for 60 Hz pickup in our LSC crate, which is anomalously large.
  • Mohana and Shihori continue to work on setting alarm limits on all EPICS channels in the EPICS alarm handler.
• Digital suspension controllers (Sakata, B. Abbott, Heefner):
  • The EPICS readback of the coils for ETMX were reading anomalously large values, although the optic was being damped just fine. Ben compared the database for ETMX with the one for ETMY, and found that the ETMX database was wrong (not known how or why) and that a recent reboot (not known when) caused the problem to surface and come to Osamu's attention. Ben fixed the database, rebooted, and all appears to be well.
• Safety (all): A 40m lab safety document T040092 has been completed and submitted to DCC.
• South Annex Bake Ovens (Taylor):
  • Bob continues work on new oven C.
  • Bob is building 40 or 50 OSEM diode boards for Janeen and for the 40m.

April 2, 2004

• Commissioning (Miyakawa, Vass, Ward, Kawamura): With the PSL problems under control (sort-of), Osamu and Rob re-established the Fabry-Perot Michelson interferometer (with misaligned PRM and SRM) and are re-measuring the noise spectrum. They are commissioning the new POB beamline. Rob is increasing the gain on the SP RFPD . Shihori is aligning and commissioning all 12 QPDs on the output beamlines. We are making plans to complete our LSC and Lock Acquisition system in preparation for locking the dual recycled IFO.
• PSL:
  • We are currently running with our replacement PSL NPRO (SN 238), installed 3/11/04, and our old MOPA power supply. The MOPA cover is propped open about an inch to allow it to cool. As far as we can tell, no cooling water is running to the NPRO through the needle valve; it appears to be stuck. In this configuration, the laser is running stably, and the NPRO DTEC voltage is ~ 1.6 volts. If the MOPA cover is not propped open, the DTEC voltage rises to 2 volts, and the NPRO trips off.
  • Steve looked at a history of the DTEC voltage for the last 800 days. About 150 days ago, it rose from 0.6 volts and hovered just below 2 volts while our old NPRO (SN 332) appeared to be dying. It crept closer to 2 v when we replaced it with NPRO SN 238, and the NPRO started tripping off. This indicates a long-standing cooling problem; we are guessing that we were flowing water cooling to the NPRO till the needle valve jammed 150 days ago, and have only now noticed it.
  • We don't have a heat sink on the new NPRO; we have ordered and will install one. Apparently, the sites all have heat sinks on their NPRO, and either have holes punched in their MOPA cover, or have the MOPA cover propped open. We have also ordered and will install a replacement water valve for the NPRO. We are in communication with Lightwave, but the rep there doesn't think that our problem is due to cooling...!
  • Intensity fluctuations out of the MOPA are 10 times larger with the new NPRO. We need to optimize the alignment from NPRO to power amplifier, to maximize power and minimize intensity fluctuations. And, we need to finish commissioning of the ISS!
  • The HV on the frequency reference cavity ion pump has had no failures ever since the 24 v power cable to the ion pump HV was replaced.
  • We intend to adjust the PSL slow loop temperature change in the EPICS state code.
• Optical sensing (Kawazoe, Sakata):
  • Shihori is working with Bob to wire up last EOS, in the AP1 beamline.
  • Rob noticed that the trigger PD for the EOS in the SP beamline was burned out; we have ordered a replacement.
  • Fumiko recentered the ITMY oplev. She is writing up her studies of oplev noise. She will move on to commission the oplev servos, next.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • Jay has begun generating a block diagram and requirements for the LSC as it relates to signal recycling. There will be 11 LSC RFPDs feeding I&Q demod boards; a 22x8 input matrix; and an 8x8 output matrix (including a path from CM -> MC).
  • Jay and Ben are ordering parts for more high frequency RFPDs and I&Q demod boards and more IFR RF generators.
  • Ben will start work on the design of high frequency (low power) RFPD boards, for double demodulation.
  • Bob is running heliac cables to all the high frequency RFPDs on the optical tables.
  • Ben is getting two more QPDs; one for installation on the new PO beamline, and a spare.
  • We plan to do a careful search for 60 Hz pickup in our LSC crate, which is anomalously large.
  • Mohana and Shihori continue to work on setting alarm limits on all EPICS channels in the EPICS alarm handler.
  • Jay and/or Mohana will implement EPICS code to automatically close the mechanical shutters in the event of a vacuum transition.
  • Ben continues to check out some varying monitoring readout of a couple (at least) of PSL channels (PMC transmitted DCPD and FSS transmitted DCPD) and believes it is due to a bad 3123 ADC.
  • Undergrad Ben Olson is measuring the transfer functions for the STACIS isolators. He has measured the three at ETMY; two work well, providing 22 dB of attenuation at 10 Hz; one is oscillating. He will learn how to adjust the gains on the loops. The STACIS controllers that are supposed to communicate with our EPICS system have many problems...
• Digital suspension controllers (Sakata, B. Abbott, Heefner):
  • Shihori has completed her writeup of the DSC input matrix diagonalization procedure and results and is submitting it to the DCC.
  • She is now preparing to diagonalize the input matrix of the last two suspensions (PRM and SRM) and is preparing the procedure for diagonalizing the output matrices.
• Dual-recycling lock acquisition (Kawamura, Miyakawa):
  • Seiji and Osamu continue to use Finesse to explore how the signals for the short DOFs (PRC, SRC, MICH) change during lock acquisition, in order to develop an efficient lock acquisition procedure. They are making good progress, and are optimistic. Seiji thinks that dither-locking MICH (l-) will work well, allowing us to aquire lock in the other two short degrees of freedom (PRC l+ and SRC ls) more robustly. Osamu is gearing up to test this.
• Safety (all): A follow-up safety audit was held last Monday March 29. All action items were addressed and closed. A 40m lab safety document is nearing completion. Many thanks to the safety audit team for their careful inspection and advice!
• South Annex Bake Ovens (Taylor):
  • Bob is wiring up the temperature controller for new oven C.
  • Bob did some bake jobs for Larry Jones' galling tests.
  • Bob is building 40 or 50 OSEM diode boards for Janeen and for the 40m.

Mar 26, 2004

• Commissioning (Miyakawa, Vass, Ward, Kawamura): There's been a break in commissioning work while repairing the PSL (see below). Now that the PSL is back up and all systems are working, we will re-establish the Fabry-Perot Michelson interferometer (with misaligned PRM and SRM) and remeasure the noise.
• PSL:
  • Our replacement PSL NPRO (SN 238), installed 3/11/04 to replace our old NPRO (SN 332) which was dying after many thousands of hours of service, started to misbehave 4 days after installation. The power drops to zero for around 10 minutes, then comes back up, and repeats every hour or 2.
  • The problem was initially associated with a 580 kHz oscillation seen on the FSS card. Rich Abbott said that this oscillation was also seen on an LLO laser. Jay and Osamu killed the oscillation with a capacitor on the FSS card, but the power problem persisted.
  • The problem did not appear when the stand-alone NPRO power supply was used, only when our MOPA power supply was used. Jay and Peter King swapped MOPA power supplies, but the power problem persisted.
  • We frantically searched for another spare NPRO; neither Peter nor LHO have any to spare.
  • The problem is also associated with raised temperature reported by the NPRO thermoelectric controller (TEC). Yesterday (Wednesday), Steve turned on the water cooling for the NPRO. As of today, the power problem has not reappeared, and the FSS, PMC, and MC remained locked throughout the night. We have switched back to the old MOPA power supply, are keeping the cooling on, and are declaring success.
  • The HV on the frequency reference cavity ion pump occasionally falls to 0. Steve and Bob replaced the 24 v power cable to the ion pump HV, and the problem has not reappeared.
  • These problems uncovered another problem: the PSL slow loop (which changes the temperature of the reference cavity when the FSS PZT signal drifts) changes the temperature by too much. Osamu will adjust this in the EPICS state code.
• Optical sensing (Kawazoe, Sakata):
  • Shihori and Fumiko are preparing to install the last EOS, in the AP1 beamline.
  • Ben is getting two more QPDs; one for installation on the new PO beamline, and a spare.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • Jay has delivered an RFPD tuned for duoble demodulation (peaks at 133 and 199 MHz, notch at 166 MHz).
  • Jay delivered two 2-omega IQdemod boards (for SPOB), and IQdemod boards for 133 and 199 MHz. Osamu will install and debug these in the LSC system.
  • Mohana and Shihori are establishing the alarm limits on all EPICS channels in the EPICS alarm handler.
  • Jay and/or Mohana will implement EPICS code to automatically close the mechanical shutters in the event of a vacuum transition.
  • Ben is checking out some varying monitoring readout of a couple (at least) of PSL channels (PMC transmitted DCPD and FSS transmitted DCPD) and believes it is due to a bad 3123 ADC. He will look for a replacement.
• Digital suspension controllers (Sakata, B. Abbott, Heefner):
  • Shihori has completed her writeup of the DSC input matrix diagonalization procedure and results; it will be released soon.
• Safety (all): Following through on the remaining action items from the safety inspection of 1/13/04:
  • All crane operators (Steve, Bob, Osamu, Rob) were trained and certified on Tuesday.
  • All 40 Meter personnel attended the lab safety briefing (led by Steve) on Thursday. AJW updated an evolving document summarizing all the material covered in the safety briefing.
  • A follow-up safety audit will be held next Monday March 29.
• South Annex Bake Ovens (Taylor):
  • Bob has recieved a new temperature controller, and 20" oven body, for the new oven C. He's working on wiring up the temperature controller.

Mar 19, 2004

• The 40m Technical Advisory Committee met on Thursday during the LSC meeting in LLO. Osamu presented the status of the project, as well as Seiji's and Osamu's results on strategies for locking the dual-recycled Michelson. Powerpoint slides are here (8 MB).
• Commissioning (Miyakawa, Vass, Ward, Kawamura): There's been a break in commissioning work while repairing the PSL (see below) and while Osamu is at LLO for the LSC meeting.
• PSL:
  • Late last week, Rob and Osamu installed a replacement NPRO (SN 238) in our MOPA laser; our NPRO was showing signs of dying. They aligned it, and tuned up the MOPA by maximizing the output power. They got up to 12.5 watts output.
  • Osamu was then able to lock the frequency stability servo, after readjusting the servo gains. He moved on to the mode cleaner, but ran out of time in optimizing the servo before he had to leave for the LSC meeting.
  • Four days after installing the new NPRO, it has been showing signs of misbehavior: the power drops to zero for around 10 minutes, then comes back up, and repeats every hour or 2. Osamu will diagnose the problem when he returns from the LSC.
  • The HV on the frequency reference cavity ion pump occasionally falls to 0. This is not being monitored by EPICS; Steve noticed it when looking at the controller. Fortunately, the FRC holds vacuum pretty well even when the pump isn't working. Steve and Rob think the problem is with the 24v cable to the HV supply, and Bob will make a replacement.
• Optical sensing (Kawazoe, Sakata):
  • Shihori and Fumiko are preparing to install the last EOS, in the SP beamline.
  • Shihori is preparing to install the remaining QPD, on the new PO beamline, and to cable up and test the MCT QPD.
  • Fumiko is writing up her oplev calibration and noise studies. She is starting to implement and test the oplev servos, and is studying the appropriate filtering.
  • Jay and/or Mohana will implement EPICS code to automatically close the mechanical shutters in the event of a vacuum transition.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran):
  • Mohana and Shihori are establishing the alarm limits on all EPICS channels in the EPICS alarm handler.
  • The monitoring readout of a couple (at least) of PSL channels (PMC transmitted DCPD and FSS transmitted DCPD) has been misbehaving badly for many months. This does not effect the servos. The cause is somewhere in the electronics rack, and Ben continues to investigate it.
• Digital suspension controllers (Sakata, B. Abbott, Heefner):
  • Shihori has measured the damping Q's for the side OSEMS on all of our suspended optics. The poor damping of ITMX side osem servo is clearly observed. Shee will now study the dependence of the Q on the gain, for all OSEM damping servos.
• Dual-recycling lock acquisition (Kawamura, Miyakawa):
  • Seiji and Osamu continue to use Finesse to explore how the signals for the short DOFs (PRC, SRC, MICH) change during lock acquisition, in order to develop an efficient lock acquisition procedure. They think they have a path for locking the PRC (l+) degree of freedom; the other two (SRC or ls, and MICH or l-) are proving to be more problematical. Osamu presented these results at the Advanced Optical Configurations group meeting at the LSC meeting in LLO.
• Safety (all): Following through on the remaining action items from the safety inspection of 1/13/04:
  • The custodian went through the basic LIGO laser training. The Safety Office has requested that they not switch custodians on us so often.
  • All 13 emergency lights in the lab have been tested. Two were not functioning right (the two that were supposedly fixed last month); they have now been fixed properly, and tested.
  • All the fire alarms (including the one newly installed in the South Annex clean room) and extinguishers will be tested on Friday.
  • All rack kill switches have been modified to provide a continuous ground path.
  • All optical tables are being grounded to the vacuum envelope. Not sure whether we should also ground the racks to the vacuum envelope.
  • Bob is setting up lockout / tagout systems for all the relevant vacuum equipment. The racks and power breakers are done.
  • All crane operators (Steve, Bob, Osamu, Rob) will be trained and certified next Tuesday. Signs listing all certified crane operators, with safety reminders, are now posted on all three crane controllers.
  • All 40 Meter personnel will undergo the lab safety briefing (led by Steve) next Thursday. AJW wrote up an evolving document summarizing all the material covered in the safety briefing.
  • A follow-up safety inspection has been scheduled for March 29.
• South Annex Bake Ovens (Taylor):
  • Bob continues to work on rebuilding oven E and building new oven C. Parts are coming in on schedule. He's now working on the temperature controllers and RGA filaments for the new ovens.

Mar 12, 2004

• The 40m Technical Advisory Committee will meet next Tuesday during the LSC meeting in LLO. Osamu will present the status of the project. We hope that there will be time in the AIC working group agenda to present Seiji's and Osamu's results on strategies for locking the dual-recycled Michelson.
• Commissioning (Miyakawa, Vass, Ward, Kawamura):
  • Late last week, Rob increased the gain on the POX RFPD to compensate for the relatively much larger 60 Hz noise that appeared when we installed and mis-aligned the power recycling mirror. The x-arm is locking robustly, with much-reduced relative 60 Hz noise.
  • Increasing the gain on the POY RFPD induced an oscillation; Ben has fixed this with the addition of a resistor.
  • Seiji, Osamu and Rob tried an experiment in which they modified the damping of the ITMX, ITMY, BS optics (comprising the Michelson) to give them lots of DC gain and a UGF at 4 Hz, in effect "tying" them to the optical tables on the seismic stacks. The hope was that, if there isn't much relative motion between the tables, such a scheme could make it easier to obtain lock in the short dual-recycled Michelson. They compared the Michelson error signal and transmitted light power before and after changing the suspended optic damping, and so no significant change. They conclude that there is too much relative motion between the tables to make this a useful approach.
  • On Saturday, the PSL frequency stabilization servo (FSS) lost lock, and we haven't been able to regain robust lock since. After carefully studying the data and performing a variety of checks of the MOPA using the PMC, Osamu, Steve and Rob found no evidence of problems with the FSS, PMC, reference cavity, slow or fast frequency actuation, chiller, cooling water, coolant piping, or the power amplifier. Reviewing the data, it appears that the inability to lock the FSS since last Saturday is due to the failure of the TEC or some other element of the NPRO. The TEC temperature is now back to normal, but the NPRO is not; it appears to have suffered permanent damage. It appears that the 40m NPRO is in its death throes, after 18,000 hours of faithful service. We have now obtained a spare 126 NPRO (SN 238) from Peter King's lab (many thanks, Peter!), and Osamu and Rob installed it into the Lightwave box. They're busy aligning and tuning the output beam. We will return our NPRO (SN 332) to Lightwave for repair.
  • As part of the diagnosis of the PSL problem, Rob, steve and Osamu shut down all electronics and computers in the lab and brought everything back up successfully. Rob is documenting the detailed procedure for doing this.
• Optical sensing (Kawazoe, Sakata):
  • Shihori and Fumiko have installed an electro-optic shutter (EOS) on the SP beamline, and checked out its operation under EPICS control. They had some problem with the trigger photodiode, and a spare one will be installed soon. They are preparing to install the last EOS, in the SP beamline.
  • Shihori has checked out the installed beamline QPDs, and will install the last two (of 12) soon.
  • Fumiko is taking noise spectra of the five oplevs that are currently in service (the ones for the SRM and PRM aren't working because the optics are misaligned). She will start implementing and testing the oplev servos.
  • Bob has cabled up all the beamline mechanical shutters One is installed on the PSL output beam; three others will soon be installed in the MCR, SP, and AP beamlines. Ben and Jay have made them work under EPICS control, although there are some questions about the wiring logic for fail-safe shuttering under loss of power or EPICS cpu reboots.
  • Jay and/or Mohana will implement EPICS code to automatically close the mechanical shutters in the event of a vacuum transition.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran)
  • Rob modified the POY RF photodiode to increase the gain, but it was oscillating. After Ben installed a shunt resistor to ground as advocated a while ago by Mike Zucker, the oscillations have stopped.
  • Ben fixed a flaky LEMO cable reading out the PRM side OSEM.
  • Bob installed some heliac cable for the 33 MHz RF signal from the PSL rack to the LSC rack. The 166 MHz signal is already carried by heliac cable.
  • The kill switches for all the electronics racks have all been modified so that the ground conductor remains connected when power is disconnected.
  • The monitoring readout of a couple (at least) of PSL channels (PMC transmitted DCPD and FSS transmitted DCPD) has been misbehaving badly for many months. This does not effect the servos. The cause is somewhere in the electronics rack, and Ben continues to investigate it.
  • Mohana and Shihori are reviewing the long list of EPICS channels in the EPICS alarm handler, to establish good alarm limits on all channels.
• Digital suspension controllers (Sakata, B. Abbott, Heefner)
  • The DSC system has functioned with no problems this week!
  • Shihori has diagonalized the input matrix for the ETMX suspension, and has re-measured all the cross-couplings for all the other suspensions now that the vent is over (except for the PRM and SRM, which are still misaligned). She has observed some significant changes in the cross-couplings in ITMX. We suspect that the side OSEM in ITMX is not working well. Under investigation.
• Safety (all): We are documenting a response to all of the action items from the safety inspection of 1/13/04, which have all been addressed.
• South Annex Bake Ovens (Taylor):
  • Bob is ordering temperature controllers and RGA filaments for the new oven he's building,

Mar 5, 2004

• Commissioning (Miyakawa, Vass, Ward):
  • We closed up after a successful 4 week vent to install the two suspended recycling mirrors, PRM and SRM. We pumped down last Thursday with no problems, and now are at 9e-6 torr with the maglev turbopump alone.
  • By Thursday evening, the mode cleaner was locked, and by Friday AM the FPMI was locked. Minimal re-alignment was necessary; the mirrors did not move much because of the pump down.
  • All error signals are now much weaker, because only 7% of the light gets through the mis-aligned PRM; and the asymmetric port has an 8% transmitting mis-aligned SRM. As a result, the 60 Hz noise is now relatively much larger, and is a real problem for locking.
  • Osamu hunted for the 60 Hz noise source. It's in the analog LSC crate, and it only goes away when all cables are unplugged and modules pulled out.
  • The band-aid solution is to increase the gains on the RFPDs sensing the FPMI degrees of freedom at the POX, POY, and SP ports.
  • Rob increased the gain on the POX RFPD, but couldn't do the POY; Ben is looking at it. Meanwhile, with this mod, the x-arm is locking robustly, with much-reduced relative 60 Hz noise.
  • Osamu is now modifying the whitening filter gains and re-tuning the LSC servo filters, to optimize the control with these lower light levels.
  • We will then want to recalibrate the FPMI and get a new calibrated noise curve.
  • This past Tuesday, the HV cable(s) to the PZT steering mirror(s) at the end of the PSL table were bumped, and the mirror(s) moved (not sure which one, or both). Before understanding what happened, Osamu re-aligned the MC to the input PSL beam; as a result, the beam to the main IFO was different. Fortunately, we have QPDs monitoring the beam position / angle at the PSL and at the MC -> IFO hand-off. Osamu has now used these to re-establish the correct beam pointing at the PSL, and has brought the MC and and main IFO alignment back to where it was.
• Optical sensing (Kawazoe, Sakata):
  • All 7 oplevs are working well after pumpdown.
  • Shihori is checking out the installed beamline QPDs, and will install the last few (of 12). Then she'll calibrate them all.
  • Fumiko did some more measurements to determine how well the oplevs sense the suspended optic pitch and yaw, above the noise. We remain unsure about how useful the oplev servos would be for us, whether we should bother to commission them, and over what frequency range they should act.
  • Because of the large amount of light reflected from the PRM to the SP when it is aligned, we need to finally install and commission the electro-optic shutter (EOS) at the SP and AP ports. Fumiko and Shihori are working on this, and Bob is helping with the wiring.
  • Bob is checking all the wiring for the remote control of the mechanical shutters for the PSL, SP, AP, MCR beamlines. We still need to complete the EPICS control software for these.
  • All camera views are established for all 7 core optics and all 8 output beams, and all signals are going into the video switch. We still need a few more camera power supplies to make use of all cameras. Steve has them on order.
  • All chamber illuminators are working and under EPICS control.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran)
  • The monitoring readout of a couple (at least) of PSL channels (PMC transmitted DCPD and FSS transmitted DCPD) has been misbehaving badly for many months. This does not effect the servos. The cause is somewhere in the electronics rack, and Rob and Ben are searching for the problem.
  • Jay reports that the PD needed for double demodulation has been tuned up and is ready for installation.The response has a notch at 166MHZ and peaks at 133 and 199MHz. The is ~30dB of suppression of the 166MHz. Better suppression will require a re-layout/redesign of the board.
  • Mohana and Shihori are reviewing the long list of EPICS channels in the EPICS alarm handler, to establish good alarm limits on all channels.
  • Steve and Bob spent some time cleaning up the PSL cable plant.
  • We continue to think about the necessary mods to the LSC software to implement lock acquisition code.
• Digital suspension controllers (Sakata, B. Abbott, Heefner)
  • On Wednesday, the DSC sensor ADCs (ICS 110b's) started to channel hop, and all the vertex suspension watchdogs tripped off. We rebooted the DSCs, and several suspensions came back, but not all. The DSC filtering computer c1suslinuxpc2, which services MC1, MC2, MC3 and SRM, wouldn't re-start; when manually restarted, it segfaulted (as if it was getting corrupted data from the sensors). Jay found a bad cable from the PRM side OSEM sensor; Ben fixed it; Alex restarted the DSC system. By the end of the day, all the DSC system is back up with all suspensions damping nicely.
  • Shihori is diagonalizing the input matrices for the ETMX, PRM and SRM suspensions.
• Safety (all): We still need to document a satisfactory response to all of the action items from the safety inspection of 1/13/04. All the action items have been addressed, we just need to assemble documentation. Steve gave a safety lecture to new undergrad student Ben Olson who is looking at the STACIS system.
• South Annex Bake Ovens (Taylor):
  • Bob continues to work with Larry and Dennis on rebuilding ovens C and F, aimed at achieving AdvLIGO cleanliness specs. He has placed orders for all parts for oven C.
  • Bob is building hybrid OSEMs or Janeen, as well as spare OSEMs for the 40m suspensions.

Feb 27, 2004

• Installation (Miyakawa, Vass, Ward):
  • We are closing up after a 4 week vent to install the two suspended recycling mirrors, PRM and SRM.
  • The installation went very smoothly and successfully. The installation crew (Steve, Osamu, Rob) installed both new suspended optics, got them damping well in the middle of the OSEM ranges, balanced the chamber optical table, established good alignment of all optics, centered the beams on the in-vac PZT steering mirrors with the PZTs in the middle of their mechanical and electrical range, established all output beams (AP, SP, PO, POX, POY, initial pointing), observed aligned spots and michelson fringes at the AP and SP beams, established all oplev beamlines, installed glass dishes to isolate the OSEM cables from the chamber walls and checked for any contact, established good camera views of all optics, checked everything, and took lots of pictures. A complete list of tasks is here. Terrific job, installation crew!
  • On Wednesday, all access connectors and doors were replaced, and pumping on the annuli commenced. On Thursday, pumpdown commenced and has been proceeding smoothly. As of Thursday 5pm, the maglev turbopump is pumping on the main volume at about 3e-5 torr. We should be ready to recommission the interferometer on Monday.
• Optical sensing (Kawazoe, Sakata):
  • All 7 oplevs are working well. The two new ones (for SRM and PRM) were set up by Rob. They require several in-vac mirrors each, and the resulting beams on the QPDs are dim, but adequate for sensing.
  • Fumiko has completed her calibration of the first 5 oplev qpds, and is measuring the pitch and yaw motion spectra in units of radians/rtHz, and rms motion in radians.
  • Fumiko is determining what changes are required to Mike ACad optical layout drawings in light of the recent installation work.
  • Shihori will be working with Fumiko to commission the electro-optic shutters for the AP and SP beamlines.
  • Shihori will commission and calibrate all 12 beamline QPD monitors.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran)
  • Jay reports that the 2 Omega I&Q demod board for SPOB is fabricated and in test.
  • Jay has a working "double resonant" photodetector circuit for testing, but it is clear that a new design is required to make it work well.
  • Mohana has completed implementing a long list of EPICS channels into the EPICS alarm handler. She and Shihori will establish good alarm limits on all channels.
  • We continue to think about the necessary mods to the LSC software to implement lock acquisition code.
• Digital suspension controllers (Sakata, B. Abbott, Heefner)
  • Several more problems with corrupted entries in the SUS AUX databases were found, and Jay fixed them. It's hard to check everything; problems will be fixed as they reveal themselves. We're not sure how or why these databases got corrupted, but care will be taken to ensure it doesn't happen again.
  • Ben fixed a flaky cable on carrying the BS LL OSEM SENSOR into the ISC patch panel.
  • Shihori is working on a final draft document describing her diagonalization of the suspension controller input (OSEM) matrix. She will diagonalize the input matrices for the ETMX, PRM and SRM suspensions next week. She is also developing her plan for diagonalization of the output (coil) matrix.
• Dual-recycling lock acquisition (Kawamura):
  • Seiji was in Aspen last week and gave a talk on dual recycling at the 40m. After highlighting the progress at the 40m, he described the challenge of finding a clean path to lock acquisition of the dual-recycled Michelson with the current readout scheme. He got lots of interesting suggestions, and quickly tried many of them with Finesse; none worked well. The path to lock acquisition will be a challenging one! We hope that expert members of the e2e team will contribute.
• Safety (all): We hope to complete a satisfactory response to all of the action items from the safety inspection of 1/13/04, next week. Meanwhile, all crane work at the 40m in the last two weeks (there was a lot of it) was performed by operators wearing steel-toed safety shoes.
• South Annex Bake Ovens (Taylor):
  • Bob continues to work on rebuilding ovens C and F, aimed at achieving AdvLIGO cleanliness specs.
  • Bob expects all ovens to be back up and running by the end of March.

Feb 20, 2004

• Installation (Miyakawa, Vass, Ward):
  • We are vented for the installation of the two suspended recycling mirrors, PRM and SRM.
  • Steve used the de-ionizing N2 gun to attempt to clean the dust off of the PRM, but it didn't do much. Then he tried the CO2 snow gun; it got some of the dust off, not all. Steve talked with Stan, Helena and Betsy; apparently, the sites have the same problem, and they live with it. So will we; we will NOT attempt to drag-wipe the optics to get the remaining dust off.
  • The installation crew installed the signal recycling mirror last week, and the power recycling mirror this week. They are both positioned within 1 mm (or so) of specifications.
  • The OSEM pigtail bracket had to be moved in order for the pigtails to reach it. Steve moved it without any problems.
  • In-vac mirrors for the oplevs have been placed on the BS optical table, and the table was leveled using counterweights.
  • Both newly installed suspended optics are now damping well, with all OSEMs nicely centered both electronically and mechanically.
  • The two in-vac PZT steering mirrors were adjusted so that they are in the center of their dynamic range, both mechanically and electronically. The settings are being auto-burted.
  • Alignment of all five vertex suspended optics is in progress, and the SP, AP, and PO beams are being established. Currently, the main beam is flashing through the mode cleaner, through the faraday isolator and MMT, through the PRM, beam-splitter, to ITMX, and back through to the asymmetric port. Work is in progress to re-establish the symmetric port beam.
  • We had been using the PRM oplev to monitor the BS table, and last week we observed a significant drift of this signal. Rob traced this to a bad oplev laser mount, and this has been fixed. The PRM and SRM oplev systems are now being aligned onto the newly-installed optics.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran)
  • Jay reports that the 2 Omega I&Q demod board for SPOB is fabricated and in test. It should be available by Monday.
  • Jay reports that work on the "double resonant" photodetector continues.
  • We continue to think about the necessary mods to the LSC software to implement lock acquisition code.
  • Jay suggests that we run the high frequency (33, 166, 199) MHz LSC RFPD signals to the I&Q demod boards using heliac cable. Bob will get the cable and connectors.
  • Alex Ivanov has developed an x-y display of digital signals using a modified dataviewer. Osamu and Seiji find the display of error-signal versus transmitted power to be a powerful diagnostic tool.
  • Mohana continues to work on the EPICS alarm handler to make it useful.
  • Caltech undergrad Ben Olsen is beginning to look at the STACIS system, to resolve problems with it.
• Digital suspension controllers (Sakata, B. Abbott, Heefner)
  • There was an earthquake last Saturday, 4.3 in central CA. All suspensions got shaken. Watchdogs tripped off for ETMX and ETMY, but for none of the others. This has been seen before. Jay looked at it, but couldn't figure out why the watchdogs for the vertex optics weren't working. Finally, he rebooted the AUX cpu that runs the vertex watchdogs, and then verified that they are now working correctly. This remains a mystery.
  • After rebooting the AUX cpu, some of the optics were not damping well, and some OSEMs were not being read out. Jay discovered and cleared up a few problems with database records caused by inadvertant replacement of files in the target area. We will be implementing procedures to minimize the risk of this happening in the future.
  • Shihori has completed a second draft document describing her diagonalization of the suspension controller input (OSEM) matrix. She is also developing her plan for diagonalization of the output (coil) matrix.
• Dual-recycling lock acquisition (Kawamura):
  • Seiji is in Aspen this week and will give a talk on dual recycling at the 40m.
• Safety (all): Almost all of the action items from the safety inspection of 1/13/04 have been followed up on, and a response report is being prepared. Now that the two suspended optics are no longer on the South Annex flow bench, work on installing a fire alarm in the South Annex will be scheduled.
• South Annex Bake Ovens (Taylor):
  • Bob is cleaning and baking a reworked SOS suspension for our BS, which has a notch cut out of it to allow the POB beam to exit without clipping.
  • Bob rebuilt oven F, and is now doing an RGA scan to assess its cleanliness.
  • Bob is doing bake jobs for Larry Jones in support of HEPI and AdvLIGO.
  • Bob continues to work with Larry Jones and Dennis Coyne on the design of new 12" bake ovens for the lab, aimed at achieving AdvLIGO cleanliness specs.

Feb 13, 2004

• Installation (Miyakawa, Vass, Ward):
  • We are vented for the installation of the two suspended recycling mirrors.
  • The installation crew replaced a Brewster prism mirror on the SP beamline with a 2" mirror, and re-established the SP beam through to the photodetector.
  • Steve set up a ccd camera on the SP for installation / monitoring.
  • The installation crew placed two in-vac mirrors to steer a pickoff beam (from BS to ITMY) onto the optical table near ITMY, and established this PO beam on that optical table.
  • The installation crew carefully measured the distances between the ITMs and the BS. Osamu and Rob carefully calculated the optical path length for the power and signal recycling cavities and the Schnupp asymmetry to an accuracy of 1 mm (Seiji established the tolerance on this to be within a few mm of the design values for the two recycling cavities, and several cm for the asymmetry).
  • Steve used the di-inonizing N2 gun on the SRM. Then, the installation crew moved the SRM suspension and installed it in the chamber, positioning it precisely within 1 mm.
  • They connected the cables, and it damped well, immediately. Well, the side damping was a bit noisy until the OSEM was adjusted and the table balanced; then all 5 OSEMs damped the optic nicely.
  • The table was balanced using counterweights, and the OSEMs were adjusted to read around 1 V each (the middle of their range). Visual inspection of the magnets in the OSEMs remains to be done.
  • Osamu and Rob noticed discrepancies between their two calculations, and when it was resolved, it was realized that the SRM was placed 2 mm off from design. The suspension was moved by that amount.
  • After a bit more table balancing, PSL flashes were seen in the initial point beams, in reflections from ITMX back to the symmetric and asymmetric ports, and at ETMX. After some touch-ups, the reflections from ITMX and the SRM were made to overlap with each other, and the incoming beam, at the symmetric port. Additional alignment of suspended optics and output beams are in progress.
  • Plan to install the suspended power recycling mirror on Friday.
• Electronics (B. Abbott, Heefner, Taylor, Mageswaran)
  • Jay reports that The 2 omega I&Q demod board needed for SPOB should be stuffed and tested by 2/13/04.
  • Jay reports that work on a "double resonant" photodetector is proceeding. Right now we can get a PD that has peaks at roughly 133 and 199MHz with a notch at 166MHz. A PD should be ready for installation and test by 2/20.
  • The signal from the PSL frequency reference cavity transmitted light photodiode is fluctuating wildly. The problem appears to be downstream of the photodiode, eg, in the cable to the 3113 ADC, the ADC, or beyond. Under investigation.
  • The PRM oplev signals are fluctuating; under investigation. We have requested a spare oplev QPD from the controls group, and we have spare oplev lasers. This oplev is currently monitoring the levelness of the BS chamber optical table, NOT the PRM, which has not yet been installed.
  • Ben notes that the Minco temperature controller on the PSL frequency reference cavity is oscillating. He and Bob will look at the filtering in that controller.
  • Mohana installed a first draft of the EPICS alarm handler. Much work is needed to make it useful.
  • Caltech undergrad Ben Olsen is beginning to look at the STACIS system, to resolve problems with it.
• Digital suspension controllers (Sakata, B. Abbott, Heefner)
  • The DSC watchdogs to not do their thing. When the rms motion exceeds the set limit, they should disable the coils; they do not. Under investigation.
  • Shihori has completed a first draft document describing her diagonalization of the suspension controller input (OSEM) matrix. She is also planning her diagonalization of the output (coil) matrix.
• Dual-recycling lock acquisition (Kawamura):
  • Seiji continues to use Finesse to explore how the signals for the short DOFs (PRC, SRC, MICH) change during lock acquisition, in order to develop an efficient lock acquisition procedure. He spoke at some length with Hartmut Grote of GEO about how they lock their dual-recycled IFO. He is trying some of the techniques they use, but it is not clear that they will be helpful to us.
  • Seiji will be in Aspen next week and will give a talk on dual recycling at the 40m.
• South Annex Bake Ovens (Taylor):
  • Bob is cleaning and baking a reworked SOS suspension for our BS, which has a notch cut out of it to allow the POB beam to exit without clipping.
  • Bob is doing bake jobs for Larry Jones in support of HEPI and AdvLIGO.
  • Bob continues to work with Larry Jones and Dennis Coyne on the design of new 12" bake ovens for the lab, aimed at achieving AdvLIGO cleanliness specs.

Feb 6, 2004

• Commissioning (Miyakawa, Kawamura, Sakata):
  • We are vented for the installation of the two suspended recycling mirrors.
  • We are carefully monitoring all oplev and OSEM signals as we work. Everything appears to be pretty stable.
  • The PSL beam (low power) is flashing through the mode cleaner, to the ITMs and back to the SP and AP ports. This beam is being used to align all optics.
  • Steve, Osamu and Robert replaced an in-vac prism mirror with a regular 2" mirror, and will use the prism mirror to correct for the astigmatism in the SP beam in air, where it can be more easily aligned.
  • Then they installed two mirrors to bring the beam from BS -> ITMY out to a nearby optical table.
  • The distances between vertex suspended optics are being measured with a ruler. We aim to install the power- and signal-recycling mirror suspensions with a tolerance of a couple of mm with respect to the ITMs.
  • Next is to install the PRM and SRM suspended optics and oplev beamlines, and align everything.
  • We decided to not center the beam better on ITMX. It is a bout 1cm from the center of the optic. Since it's not really a problem, and would be a big chore to correct, we decided not to bother.
  • After we pump down, we will re-establish the Fabry-Perot Michelson (FPMI), with misaligned power- and signal-recycling mirrors. Our original plan was to first commission a LIGO-1-like power-recycled FPMI. But we decided that we would not learn anything valuable from that exercise, and that it would be more efficient to go straight to a dual-recycled Michelson (without Fabry-Perot arms, at first). Opinions on this are welcome.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • The 2 Omega I&Q demod boards needed for the SPOB signal should be ready within the week.
  • Mohana is working on the EPICS alarms and vacuum interlock.
  • AJW, Jay and Alex have been looking at the LSC front end code (Rolf) and lock acquisition code (Matt), and are discussing the mods needed for the dual-recycling experiment.
  • A considerable amount of additional electronics (LSC RFPD's, double-demod boards, and associated wiring) still need to be specified, designed, and built.
  • The STACIS system still is down, awaiting some attention.
• Digital suspension controllers (Sakata, Goggin, B. Abbott, Heefner)
  • Shihori is documenting her diagonalization of the suspension controller input (OSEM) matrix. She is also planning her diagonalization of the output (coil) matrix.
• Simulations:
  • Seiji has released a note on Signal Extraction Matrix of the 40m Detuned RSE Prototype, T040010-00-R.pdf. This note explores different variations on the 5x5 signal extraction matrix for Detuned RSE: comparison of differential demodulation and double demodulation, dependence of each DOF on demod phases, allocation of ports for optimal signals, comparison of different PO ports, and cavity length tolerances.
  • Seiji continues to use Finesse to explore how the signals for the short DOFs (PRC, SRC, MICH) change during lock acquisition, in order to develop an efficient lock acquisition procedure.
• Safety (all): Following up on the action items from the safety inspection of 1/13/04:
  • Bob ordered steel-toed safety boot covers for crane operation, but they're too small, so they will be returned for larger sizes.
  • Steve installed drip trays under the vacuum roughing pumps.
• South Annex Bake Ovens (Taylor):
  • Bob continues to work with Larry Jones and Dennis Coyne on the design of a new 12" bake oven for the lab, aimed at achieving AdvLIGO cleanliness specs.

Jan 30, 2004

• Commissioning (Miyakawa, Kawamura, Sakata):
  • We vented on Wednesday, for the installation of the two suspended recycling mirrors, with no problems or incidents.
  • As noted below, we had many problems with the suspension controllers on Monday. By Wed morning, things had been stable for more than 24 hours. All oplev, OSEM, PZT, etc signals and servo settings were recorded before the vent.
  • During the vent, the drifts in the oplev and osem readings were carefully monitored. There was considerable motion in the oplev signals, much less on the OSEMS: consistent with the motion being due to the table motion (in yaw, mainly) as the seismic stacks respond to the vent. There is clearly a daily drift in the oplev readings, correlated with temperature, but it isn't very big (~ 0.2 volts, while deviations of ~ 1 volt are tolerable for maintaining alignment). After the vent, the P/Y biases on the OSEMS were changed to recover the pre-vent oplev readings.
  • Osamu completed the mode-matching into the IFO last week, by maximizing the ratio of the x-arm transmitted power to input power. He needed to make only one turn of the MMT adjustment screw, which means the original MMT alignment and adjustment was done very well (thanks, Koji!). The MMT did not steer the beam very much; the spots did not move visibly on the PZT steering mirrors, and only moved by around 1 cm 40 meters away. The PZT steering mirrors easily corrected for this.
  • We are now at atmosphere, and the access connectors have been removed. Osamu re-established the PSL beam through the mode cleaner, and flashes were seen through the faraday isolator. Remarkably, no steering of the PSL launch was required!
  • Now, Rob is establishing a pair of optical levers to monitor the levelness of the BS optical table, where all the installation action will be. Next is to replace an in-vac prism mirror with a regular 2" mirror, and will use the prism mirror to correct for the astigmatism in the SP beam in air, where it can be more easily aligned. Then install the pickoff beamline, and then intall the PRM and SRM suspended optics and oplev beamlines, and align everything.
• Digital suspension controllers (Sakata, Goggin, B. Abbott, Heefner)
  • Last week, the digital suspension controllers for the PRM and SRM were fully tested by having them damp the BS. This week, Steve & Rob used them to damp the PRM and SRM, on the South Annex flow bench. Now, everything's been tested except for the in-vac cables, which were tested when they were installed, and one satellite amplifier, which is at Wilson House for full checkout.
  • On Monday night, the c1suslinuxpc1 computer which filters the suspension controller signals for the BS, ITMX, ITMY suspensions went crazy: the cpu maxed out, the three suspensions were shaking all night long, and the watchdogs (on a separate aux computer) failed to shut them down. Rebooting all the relevant computers in the morning recovered the system, but the cause of it remains a mystery.
  • The shaking might have messed up the suspension alignment; the OSEM readings were different after damping was re-established.
  • It appears that the oplev signals are a better (short-term) reference than the OSEMs. After the shaking, Osamu found that the Michelson error signal was better when the P/Y biases were adjusted to recover the old oplev readings, rather than the old OSEM readings.
  • Shihori has re-measured the pos/pit/yaw frequencies and couplings on all the optics whose input matrices she has diagonalized in the last few weeks. She finds that the frequencies are unchanged, but that the pos/pit/yaw coplings have degraded a bit. A document is in progress.
• Optical sensing:
  • The oplev QPD readout for the PRM and SRM have been tested and seem to work well.
  • Fumiko recentered all oplevs prior to venting. She continues to work on calibrating the oplev response.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • The computer that does the filtering for the mode cleaner WFS system, c1asclinux, loses time synch regularly (every day or so). Sometimes it recovers upon reboot, sometimes not. On Monday, when it would not recover, it was power-cycled, which interrupted the RFM network, bringing down the DSC (digital suspension controllers) system. To avoid this, Alex Ivanov gave this computer its own slot in the RFM bypass switch.
  • Jay and Alex fixed the code that was reading out the QPDX signals incorrectly. Now, QPDX appears to be working well.
  • The 2 Omega I&Q demod boards needed for the SPOB signal are being fabricated.
  • Mohana is working on the EPICS alarms and vacuum interlock.
  • In the process of mode matching (see above), the PZT steering mirrors were turned on and off many times, and they always come back very close to where they were.
  • The new EPICS controls of the PZT steering mirrors work well. Osamu will adjust the biases to allow the desiring setting to be mid-range on the sliders, while ensuring that the voltage applied to the PZTs are always within the 0-10 volt spec.
  • We are startingf to look at the lock acquisition code. We will need Rolf's help, but it is unlikely that we will get it soon, since he is saturated with HEPI.
  • The STACIS system still is down, awaiting some attention.
• Simulations:
  Seiji is using Finesse to address several questions regarding the length sensing matrix:
  • The baseline for the 40m is to use fouble-demodulation at f2xf1 (f1 = 33 MHz, f2 = 166 MHz) to sense MICH and f2-f1 to sense PRC and SRC. Seiji finds that double-demodulation makes noticable improvements for PRC and SRC as well, so we tentatively plan on using double demod for all three length sensings.
  • Seiji is exploring the demod phases required for double demodulation: one has to find a pair of demod phases that produces no DC offset to the error signal while maximizing the error dignal for the desired length degree of freedom. He can find workable phases (in principle).
  • Seiji explored the relative merits of different pickoff ports for length sensing, and finds them all to be roughly equivalent in terms of error signal strength. We will instrument the pickoff of the light from the BS to ITMY, for convenience.
  • Seiji explored the sensitivity of the length sensing matrix to deviations from the ideal macroscopic cavity lengths. He finds length tolerances of 6 cm for MICH, 3 mm for PRC, and 3 mm for SRC. It may be difficult to position the PRM and SRM sufficiently accurately in air, which may require us to iterate (vent, reposition, pump down).
• Safety (all):
  Following up on the action items from the safety inspection of 1/13/04:
  • Bob and Steve continue to set up the lockout-tagout system.
  • Steve is getting a quote on the installation of a fire alarm in the South Annex clean room.
  • Steve asked the electricians to identify a main breaker for the south annex which is outside the lab and which thus could be accessed in case of a fire in the lab.
  • Steve asked the electricians modify the rack power kill switches so that the ground is not broken (as they currently do).
  • Steve will ask the electricians to ground all the optical tables to the 40m vacuum envelope (basically, anything but the racks).
  • Steve arranged for routine fire alarm maintenance, which was done this week.
  • Steve and Bob are getting info on formal crane operation certification from the safety office.
  • Bob ordered steel-toed safety boot covers for crane operation.
  • Steve will install drip trays under the vacuum roughing pumps.
  • All 40m personnel will get an update on lab safety from Steve in the next couple of weeks.
• South Annex Bake Ovens (Taylor):
  • Bob is working with Larry Jones and Dennis Coyne on the design of a new 12" bake oven for the lab, aimed at achieving AdvLIGO cleanliness specs.
  • The SRS RGA has been repaired and is being shipped back to us.

Jan 23, 2004

• Rob Ward has joined the 40 meter team as a graduate research assistant. Welcome!
• Commissioning (Miyakawa, Kawamura, Sakata):
  • We plan to vent for the installation of the two suspended recycling mirrors, early next next week.
  • Osamu remeasured the Michelson asymmetry using the 33 MHz RF sidebants, and obtained 0.462 +- 0.028 m, or 0.012+-0.028 m larger than design. It is hard to reduce the error on this measurement as it is currently being done.
  • Seiji used Finesse to determine that the length sensing matrix is not too sensitive to the Michelson asymmetry, and can tolerate a few cm difference from design.
  • However, Seiji also found that the lengths of the power and signal recycling cavities must be within a couple of mm of design, or the length sensing matrix will be significantly degraded. We're not sure how we can place the optics to that level of accuracy in air! We will try our best, and iterate (requiring vents) if necessary.
  • Osamu looked at the optical axis of the 29 MHz and 33 MHz pockels cells, and found they differed by 18 degrees! But, then he found that a HPW placed before the 29 MHz pockels cell in the PSL IOO path during the re-layout was actually a quarter wave plate! He has replaced it, and remeasured the optical axes of the 2 pockels cells; they are within 3 degrees, quite tolerable. He has realigned the beam through all 3 pockels cells (including 166 MHz), minimized the AM for the 33 MHz, and found that the AM on the other two frequencies was small and tolerable.
  • He then re-established the mode cleaner lock with the correct polarization and modulation depth for all 3 RF frequencies.
  • Osamu is testing the use of the in-vac mode matching telescope and in-vac PZT steering mirrors, in preparation for mode matching into the main IFO after installation of the recycling mirrors. We are relieved to see that moving the mode matching telescope focus steers the beam only slightly, and the spot doesn't move off the small PZT steering mirror downstream. Current plan is to mode match by maximizing the light transmitted through the arm cavities; not very sensitive, but good enough for us.
  • Before vent, we will recenter all oplevs, and document all the readbacks and settings for all servos, OSEMS, oplevs, PZT steering mirrors, etc.
  • Rob Ward has built a small Michelson interferometer for use as a hands-on demo. It works great!
  • Seiji has been looking at uses for double demodulation to improve the diagonality of the length sensing matrix, and finds that it can be used to improve the matrix for the PRC, SRC, and MICH degrees of freedom (l+, ls, l-), *if* the demod phases can be chosen and optimized well. We are not sure how to do that!
  • Bob has returned the dycor RGA head to Steve, and Steve has reinstalled it on the 40m main volume and calibrated it. The electron multiplier on this head doesn't work, but it isn't a big problem right now.
  • Steve has all the 2" mirrors and mounts needed to install a BS pickoff beam. He has all the in-vac oplev mirrors. he has all the weights for balancing the BS table (cleaned and baked). He has all the clean air for the vent.
  • Bob is getting replacement bulbs for the chamber illuminators (plus spares).
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • Jay and Peter Fritschel discussed the prototyping of the sensing beamline centering/stabilization servo at the 40m. Peter will send us some steering mirrors. Rob Ward will take responsibility for commissioning and characterizing the servo.
  • Jay has 133 MHz and 199 MHz resonant circuits working for the LSC RFPDs.
  • Jay has 2omega I&Q demod boards for SPOB signal ordered and are due by 1/29. They will then be stuffed and tested.
  • Mohana is working on the EPICS controls of the PZT steering mirrors, the mechanical shutters, and the vacuum interlock (ensure that the PSL shutter is closed, and the PZT mirrors are not energized, when making transition from vent to high vacuum and vice versa).
  • Rob and Steve discovered a 2 hz comb on the ETMY UL OSEM noise spectrum. Jay determined that it was due to ICS110b channel hopping, and he cold-booted the crate (power down crate; count to 60; power back up; restart processor) and it went away. A day later, it was ETMX LL, again fixed by cold booting.
  • The c1asclinux processor (for MC WFS ASC control) has been falling out of time-synch; the servo works fine, but the DAQ data is probably screwed up. Rebooting fixes it. Sigh.
  • Lisa Bogue has instructed us on how to back up the data on our main CDS computer (op140m), and we will do it weekly.
  • Work on ASC readout (especially the QPDs) is in progress.
  • The STACIS system still is down, awaiting some attention. Rob Ward is starting to look at it.
• Digital suspension controllers (Sakata, Goggin, B. Abbott, Heefner)
  • We have checked out the PRM and SRM digital suspension controllers (DSC's) by using them to damp the BS. Next, we will damp the actual suspensions with them (they are in the South Annex flow bench). Thus, everything will be well tested prior to installation, except for the in-vac cables (which were tested when they were installed last year).
  • Shihori has completed the diagonalization of the input matrix for all suspended optics except for ETMX (which has a badly-positioned OSEM which will be fixed during the upcoming vent). She obtains satisfactory decoupling of POS, PIT, YAW; the worst cases in the small optics are ~ 8% cross-coupling; in the ITMs and ETMs, ~ 3%. She is documenting the procedures and results, and will move on to the output (coil) matrix.
• Optical sensing:
  • We should have 12 QPDs looking at the main beam (as well as 7 oplev QPDs). Most are in place, have light on them, are being read out, and being logged to frames (16 Hz). Steve and Jay are getting two very important ones (initial pointing position and angle) all checked out. A few more will be completed after the vent. (Done: IOPOS, IOANG, TRX, TRY, POX, POY. In progress: IPPOS, IPANG. Later: SP, AP, MCR, MCT).
• Safety (all):
  Following up on the action items from the safety inspection of 1/13/04:
  • Bob has installed a lockout-tagout kit, and he and Steve are setting up the system. We will all be trained on it in a week or 2.
  • Steve and Bob are getting info on formal crane operation certification from the safety office.
  • Steve & Bob are getting steel-toed safety boots for all crane operators (Steve, Bob, Osamu).
  • Steve will install drip trays under the vacuum roughing pumps.
  • Bob has modified and reposted the emergency procedure for the South Annex bake-out facility.
  • An audible fire alarm will be installed in the back room of the South Annex.
  • Steve has removed obstructions in the areas around the electrical panels and the emergency exits.
  • All 40m personnel will get an update on lab safety from Steve in the next couple of weeks.
• South Annex Bake Ovens (Taylor):
  • Bob is baking glass dishes to hold the in-vac cables, to keep the ground braid isolated from the chamber walls.
  • Bob continues to work with Larry Jones and Dennis Coyne on getting the bake ovens clean enough for AdvLIGO specs. He's building a new super-clean oven, and is replacing the temperature controller for another.

Jan 16, 2004

• Safety (all):
  • There was a safety inspection of the 40m lab on Tuesday, January 13. Several items that were called out last year had not been addressed; All of these issues are now being addressed, aggressively, belatedly.
  • Bob has ordered a lockout-tagout kit, and will set up the system. All personnel will be trained in its use.
  • Steve and Bob are investigating more formal crane operation certification.
  • Bob is ordering steel-toed safety boots for all crane operators (Steve, Bob, Osamu).
  • Steve is ordering, and will install, drip trays under vacuum roughing pumps.
  • Bob is modifying the emergency procedure for the South Annex bake-out facility in case of fire, equipment overheating/failure or loss of electrical power, based on comments from the safety inspectors.
  • Bob had the South Annex fume hoods calibrated, and they are within specs.
  • An audible fire alarm will be installed in the back room of the South Annex.
• Commissioning (Miyakawa, Kawamura, Sakata):
  • We are planning to vent for the installation of the two suspended recycling mirrors, next week. More detail below. The current work is all preparation for this installation.
  • Steve is completing the installation and commissioning of all the QPDs for beam monitoring, especially the ones that monitor the initial pointing of the beam just before it hits the PRM.
  • A mechanical shutter has, at long last, been installed on the PSL periscope. EPICS control of this shutter, including an interlock to the vacuum state, will be implemented by Jay soon. Steve and Osamu will install three more mechanical shutters in the API table, for the MCR, SP, and AP beamlines.
  • The 166 MHz pockels cell has been installed on the PSL table. All wiring is in place, the pockels cell has been energized; modulation depth checked with a scope; signals appear at the 166 MHz I&Q demod board (at the SP beamline); and into the digital LSC system. The signal generator has been phase-locked to be exactly 5x the 33 MHz signal.
  • Osamu used the 166 MHz sidebands to measure the Michelson Schnupp asymmetry. He put a HWP before the 166 MHz pockels cell to induce AM as well as PM onto the light, and measured the transmission of the AM at the dark port. From this he calculated the Schnupp asymmetry to be (0.39+-0.05) meters, or (6+-5) cm lower than design (the design will tolerate a difference of up to 10 cm). The large errors are due to difficulty in measuring the AM in the presence of beats at lower frequencies. Osamu is working on a different, more sensitive method.
  • However, the 166 MHz RF PD seems to have sustained some damage, and Osamu is working on getting a replacement PD.
  • Osamu has developed a procedure for aligning the PSL beam into the Mode Cleaner (in vacuum or in air) using the MC error signal as a guide. This should make it much easier to use the PSL beam, in air, to align the interferometer during installation.
  • Osamu checked that we are able to robustly lock the arms (using POX and POY) even at 8% of our usual input power, thereby simulating the situation when the PRM is installed (but misaligned so that the power recycling cavity is not in resonance). Indeed, we can; however, the DC output is very low, and is in the digitization noise. Jay will look into increasing the gain on the PD board.
  • We will check out the damping of the PRM and SRM on the bench in the South Annex, using the real satellite amplifiers and digital suspension controllers; only the in-situ cables will remain to be tested during installation.
  • Steve has assembled all of the vacuum mirrors needed for installation (Lee Cardenas will assemple a few more in the next few days), for the pickoff beam, the oplevs, and the video views. He has also ordered more laser diodes for the oplevs. He has all the clean air tanks for the vent.
  • Steve and Bob will prepare clean glass dishes to hold the in-vac cables, so that the ground braids aren't electrically shorted to the chamber.
  • Osamu has assembled a detailed installation procedure / checklist.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • Jay retuned an LSC PD and I&Q demod board for operation at 166MHz. The higher frequency required us to use a smaller active area PD. The PD and demod board are installed and operational in the SPD2 position. (But see above; it is already damaged).
  • Jay installed 2 more power supplies in parallel with the +/-15V supplies to raise the maximum current to 8 amps. Requested sorenson power supplies from LLO to be used to handle future power supply needs.
  • Jay is fabricating another 2 Omega I&Q demod board for SPOB.
  • Work on ASC readout (especially the QPDs) is in progress.
  • Mohana is implementing the EPICS control of the in-vac steering mirrors. These will help us to re-establish correct initial pointing, during and after the installation.
  • Mohana is also implementing the EPICS control of the mechanical shutters, and the interlock with the vacuum.
  • Most of the VME cpus went down over last weekend for some unknown reason. After a few failed attempts, Jay and Alex got them all back up.
  • DTT has not been working well. Alex and Daniel Sigg have found bugs in the code (related to timing); they fixed one and are working on another.
  • The STACIS system still is down, awaiting some attention. Rob Ward is starting to look at it.
• South Annex Bake Ovens (Taylor):
  • Bob obtained a new RGA for his bake ovens.
  • The SRS RGA was removed from the 40m and is now in Bob's lab.
  • Bob has reviewed the draft specs on vacuum cleanliness for Advanced LIGO, from Dennis Coyne.
  • Bob continues to work with Larry Jones on making the mods to his ovens to mean the Advanced LIGO cleanliness specs; looks do-able.
  • See safety issues, above.

Jan 8, 2004

• Commissioning (Miyakawa, Kawamura, Sakata):
  • We are now routinely locking our Fabry-Perot-Michelson (FPMI).
  • During the long break, Osamu worked on filters and whitening/dewhitening to reduce the noise in the FPMI; the resulting spectrum can be seen here.
  • Seiji and Osamu tried getting a Michelson control signal using the 2nd harmonic of the 33 MHz RF; unfortunately, they could not turn up the modulation depth sufficiently, and thus a signal at 99 MHz (3rd harmonic demodulation) could not be cleanly seen.
  • Osamu, Steve, and Bob are installing the 166 MHz Pockels cell and associated cabling, for HF RF sensing of the dual-recycling length (and angular) degrees of freedom. Jay is preparing the 166 MHz LSC RFPD and I&Q demod board.
  • Osamu and Seiji are making plans to measure the Michelson asymmetry. One approach is to amplitude-modulate the 166 MHz RF sidebands, and measure the reflection to the SP. Another approach is to lock the arms one at a time using the SP RFPD, and measure the difference in demod phase required to zero the signal at resonance (as Peter Fritchel explained to us).
  • We are making plans to install the power recycling mirror (PRM) and signal recycling mirror (SRM) in the next week or 2. In consultation with Mike Smith, we outlined a tentative alignment procedure for aligning the recycling mirrors, using an autocollimator beam brought in from outside (so as to not have to use the PSL beam through the mode cleaner, while in air; this requires too much re-steering of the PSL, which then has to be undone after we pump back down).
  • We plan to vent in the next week: install the two suspended recycling mirrors; install mirrors for oplevs for those suspended optics; install mirrors for camera views for those suspended optics; fix a badly-positioned UR OSEM on ETMX; adjust the in-vac PZT steering mirrors; commission the PRC PO port; and align the beam through the DR IFO. We hope that we can complete all these tasks in 2 weeks.
  • We also need to replace an in-vac Brewster prism mirror (used to correct the astigmatism placed on the SP beam) with a regular mirror; we'll use the Brewster prism mirror in air, where we can fine tune its position.
  • During this time, we will continuously monitor the positions of the suspended optics with our optical levers. Fumiko is making sure that all the oplev beams are centered on their QPDs, and reliably sensing the optic P&Y.
  • We will then re-establish the FPMI with mis-aligned PRM and SRM. Before the installation, we will verify that we can control the FPMI with the expected loss of incident power through a mis-aligned 7% transmissive PRM.
  • We will postpone the cleaning of the optics (we know their surfaces are dusty and reducing the visibility of the high-finesse arms) for a later vent.
  • We will postpone fine-tuning the mode matching into the IFO until after we install the recycling mirrors. However, we will go through the mode-matching procedure (changing the MMT focus, and compensating for beam steering with the the in-vac PZT steering mirrors) before venting, in order to determine whether / how much to move the PZT steering mirrors positions.
  • Steve is working on the commissioning of the remaining QPDs for the initial pointing position, and then the QPDs and cameras at the SP and AP.
  • We are working on the commissioning of the LA code for power recycling and then dual recycling. currently, we use a fixed sensing matrix for lock acquisition of the FPMI.
  • Jay is working on the readout of the ASC QPDS, which we need for linearization of the sensing signals for lock acquisition.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • Jay delivered to us an LSC RFPD tuned to 99 MHz and associated I&Q demod board for the 3rd harmonic demodulation experiment referred to above.
  • Jay is now in the process of converting the LSC RFPD and I&Q demod board for 166 MHz. The PD will likely require that we use the smaller active area PDs that we have, to control parasitic capacitances.
  • Several of the chamber illuminators are not working. Bob will check; they may just be blown bulbs!
  • Jay will set up EPICS controls for the mechanical shutters, including an interlock to shutter the PSL beam if the vacuum pressure goes up.
  • The STACIS system still is down, awaiting some attention.
• There will be a safety inspection of the 40m lab on Tuesday, January 13. An update was made to our 40m PSL SOP (Standard Operating Procedure) document, as recommended by the safety inspectors last year.

Dec 19, 2003

• Commissioning (Miyakawa, Kawamura, Sakata):
  • After some effort, Osamu and Seiji have managed to get the ETMX and ETMY coil dewhitening filters turned on after aquiring lock, without losing lock. This required several changes: (a) when turning on the hardware dewhitening, the software inverse dewhitening gets turned on at the same time. That software (digital) filter was being turned on with "Zero history" instead of "always on", so there was an initial impulse which would kick the mirror, causing lock loss. They changed it to "always on", fixing that problem. (b) The feedback signal has large peaks at 3-5 kHz and 300-500 Hz, origin unknown. (They will trace these upstream to see where they come from). The 3-5 kHz oscillation was saturating the coil driver DACs. To reduce the effect of these peaks, they adjusted the filters, reducing the UGF from 300 Hz to ~100 Hz, adding more phase margin at 300 Hz, and reducing the LPF corner frequency. (c) They put in notch filters to eliminate the feedback at the peaks at 300-500 Hz.
  • After all this, they were able to switch on the ETM coil dewhitening filters (and digital inverse filters) without losing lock. They saw 30 dB of noise reduction in AS_Q from 40-300 Hz.
  • Next big effort is to mode match the beam into the arms. Our in-vac mode matching telescope uses fixed mirrors with a stepping motor for focus. When we adjust the MMT, the beam is steered. So, we need to complete the setup of the QPDs that monitor the initial pointing position and angle of the beam (after the MMT but before the IFO) to correct for this using the in-vac PZT steering mirrors. And, we are setting up remote control of these PZT steering mirrors. Osamu and Seiji are working out methods for sensing mode mismatch using reflected light through an iris to sense the TEM00 mode, as the MMT is stepped through its range. Osamu's first attempt to make this work failed, presumably due to the ringing of the reflected light power due to doppler effects. Seiji is developing another scheme using the Michelson which should be immune to this effect.
  • Seiji and Shihori checked one aspect of the displacement calibration: is the BW reported by dtt right? It is.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • Jay has been debugging the electronics readout of the QPDs that were set up by Steve. He found a bad 3113 and has replaced it.
  • Jay began the tuning of an LSC PD for 99MHz operation. I t is proving to be difficult due to parasitics in the circuitry. An I&Q demod board is also being tuned for 99MHz operation.
  • Bob is running cables for the remote control of the mechanical shutters for the input beam from the PSL and the output beams for MCR, SP, and AP.
  • Jay will set up an EPICS channel that can be used to shut the mechanical shutter at the PSL periscope if the vacuum pressure rises above some level.
  • Bob is running cables for slow EPICS control of the PZT steering mirrors.
  • Steve set up a ccd camera to provide a wide view of the entire PSL table.
  • Bob has run the last of the cables for the ccd cameras that view all the optics and all the beams (and the PSL table).
  • We are following with interest the discussions in the CDS group about (a) placing multiple RF sidebands with one EOM. (b) Progress with the PSL ISS at LLO. (c) The ASI servo at LHO. (d) Servoing the beams onto the photodiodes on the ISC tables.
  • The STACIS system still is down, awaiting some attention.
• PSL (Miyakawa, Kawamura, Abbott, Nocera):
  • The power from the PSL was observed to drop by a factor of 2 on 12/4. After much investigation, Osamu discovered that it was mode-hopping. He adjusted the FSS slow loop (temperature servo) DC offset until the laser re-locked onto the correct mode, and the power came back up.
  • The PSL ISS system is disabled until Flavio investigates the dynamic range problem.
• Optical sensing (Smith, Kawazoe):
  • Steve completed the instumentation of the POX and POY beamlines as well as the TRX and TRY, and is now turning to the initial pointing position QPD, and then the QPDs and cameras at the SP and AP.
  • Fumiko has made great progress in establishing the sensitivity of the oplevs to the motion of the suspended optics, and also the laser pointing jitter noise. She will soon be able to turn to implementing the oplev damping servos.
• South Annex Bake Ovens (Taylor):
  • Bob continues to work on cleaning the bake ovens and attempting to each the cleanliness levels required for AdvLIGO.
  • Several of Bob's bake ovens have RGA heads that are 15-20 years old and are starting to fail. He will purchase a new one.
  • There was a fire alarm test at the lab, and the alarm could not be heard in the south cleanroom in the south annex. Steve will follow up.

Dec 12, 2003

• Commissioning (Miyakawa, Kawamura, Sakata):
  • Osamu has been out with the flu, so there hasn't been much commissioning work.
  • Seiji wrote up a document describing the calibration procedure of the Fabry-Perot Michelson, resulting in a noise spectrum that bottoms out at ~ 2e-16 m/rtHz above 2 kHz.
  • Osamu and Seiji found that the demod phase required to lock the Michelson if only one arm was locked was different, depending on which arm it was. Seiji determined that this is due to contamination of a large ASI signal into ASQ when the demod phase is not tuned precisely correctly, to better than 1degree. This isn't really a problem, because one normally doesn't bother to lock the Michelson with only one arm locked. I suppose we will eventually need an ASI_CORR servo.
  • Our original plan was to attempt to lock the michelson using beats between the 33 and 166 MHz sidebands (ie, not using the carrier, and thus insensitive to the arms). However, Seiji realized that this could not work, because both sidebands are phase modulations, and no Michelson signal can be obtained using phase modulation only. He suggests using the 2nd-order sideband at 66 MHz, which is amplitude modulation; beating that against the 166 MHz will yield a signal at 99 MHz, sensitive to the Michelson. Seiji verified this with Finesse. Jay will see about getting an LSC RFPD and demod board for 99 MHz.
  • Osamu is documenting all the 40m procedures and the performance of the various components (PSL, MC, IFO). It's already a very substantial document, but still in progress.
  • Steve is working on installing the last of the beamline cameras and QPDs.
  • Next big step is to implement 166 MHz RF sidebands and demod boards at 166, 99, 133, and 199 MHz.
  • And, tune mode-matching into the main IFO.
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • Medm screens for the beamline QPDs have been made, and the QPD channels have been added to DAQS.
  • From Jay: Researching various ways to produce the 133MHz and 199MHz that the 40M double demod system will require. Our fall back is to use phase locked IFRs. I am also looking into how to make an LSC PD with resonant peaks at 133 and 199MHz and a notch at 166MHz. Theoretically it is possible, but practically it may be difficult.
  • Bob is running RF heliac cables for these new RF signals.
  • There are plans brewing to add steering servos to the sensing beamlines, as a prototype for the sites. It would be relatively easy to add these to the existing controls.
  • Bob continues to clean up and label the cable plant throughout the lab.
  • Osamu is preparing to install the 166 MHz EOM on the PSL table.
  • The STACIS system still is down, awaiting some attention.
• PSL (Miyakawa, Kawamura, Abbott, Nocera):
  • The PSL ISS system is disabled until Flavio investigates the dynamic range problem.
• Digital suspension controllers (Sakata, Goggin, B. Abbott, Heefner)
  • Shihori now has a suspension diagonalization procedure that seems to work well, and has completed the input matrix for ITMY.
  • She will move on to the other 9 optics, then work on the output matrices, and then optimize the damping.
• Optical sensing (Smith, Kawazoe):
  • Fumiko continues to work on understanding the oplev signals, to establish the calibrated rms motion of the sensed optics and distinguish it from oplev sensing noise.
• South Annex Bake Ovens (Taylor):
  • Bob continues to work on cleaning the bake ovens; some are so dirty that they have to be rebuilt. He's making progress, and the oven cleanup is improving the RGA scans.
  • At the moment, the ovens in the south annex (and at the sites) are not clean enough to meet the AdvLIGO specs. Bob is working with Dennis Coyne and Larry Jones to see what can be done.

Dec 5, 2003

• Commissioning (Miyakawa, Kawamura, Sakata):
  • Osamu, Seiji, and Shihori have been working on tuning the filters for length control of the Fabry-Perot-Michelson to achieve low-noise operation.
  • They did a careful calibration (documentation in progress) and a displacement noise spectrum. To reduce the noise, they are working on implementing whitening and dewhitening filters in hardware, with appropriate software compensation.
  • They succeeded in reducing the noise above 2KHz (to around 1e-16 m/rtHz) using the hardware LSC whitening filter, after adjusting offsets and gains.
  • They had trouble employing the dewhitening to the coil drivers; there was a large 100 Hz oscillation. They traced it to the MC servo cross-over (between MCL and VCO). They reduced the MCL gain, but MC locking became less stable; they suspect VCO saturation.
  • But, with the modified MCL/VCO gains, the noise in the 30-300 Hz region is now 2x better than before.
  • Another problem with the hardware dewhiting filter is that the board is RevB2; should be B5. This has been fixed; now, the dewhitening filter transfer function is according to design.
  • There's also some 4kHz oscillation in the coil driver. All under investigation.
  • The symmetric port beam seems to be clipping at an in-vac Brewster prism mirror. This might improve once we properly mode-match into the IFO (not yet done).
  • IFO auto-locking scripts have been developed and tested.
  • QPDs and cameras have been installed and aligned at the ends (TRX and TRY). Still need to install the remaining optics in POX and POY, and complete the sym port (SP) beamline, including electro-optic shutter.
  • Next big step is to implement 166 MHz RF sidebands and demod boards at 166, 133, and 199 MHz.
  • And, tune mode-matching into the main IFO.
  • Osamu is documenting all the IFO locking, tuning, and calibration procedures that he and Seiji are developing.
• Computing:
  • Our main "online" computer, op140m, crashed last week, bringing all commissioning to a halt.
  • Lisa Bogue came to our rescue, and with Jay, they took the opportunity to finally bring up our Sun E450's and completely re-arrange the computers in our control room. Bob, Steve, and Osamu all helped rearrange and recable all the computers there.
  • One E450 is now the new op140m, and has all the controls software (targets, dataviewer, dtt, etc). One of the old Blade 100's now has 2 screens in "cinematic" mode.
  • Finally, dataviewer and dtt now seem to run reliably on the E450. These essential commissioning tools were all but useless when running on our Blade 100's; they now seem to work!
  • Lisa is rearranging all the data and accounts in our martian network, cleaning up a very sloppy and barely functional system.
  • Many thanks to Lisa and Jay for getting our control room back in shape!!
• Electronics (B. Abbott, Heefner, Taylor, Nocera, Mageswaran)
  • From Jay: ASC aux crate testing is ~90% complete.
  • Ben has installed balancing pots on the last three Universal Dewhitening boards needed for the IFO. These and others that he had done previously, are now installed on: ITMX, ITMY, BS, PRM, SRM, ETMX, ETMY, and MC2. These optics are now finished on this issue except the four newly installed ones (ITMX, ITMY, PRM and SRM) which need to have their digital gains balanced by Osamu.
  • Jay continues to develop plans to form double demod signals (at the sum and difference of 33 and 166 MHz) for LSC. He suggests first using the sums and differences, then move to genuine double demod in a second step.
  • So, we'll form 133 MHz and 199 MHz by phase-locking the signal generators, then bring those signals over to the LSC crate demod boards.
  • Bob is running RF heliac cables for these new RF signals.
  • Osamu is preparing to install the 166 MHz EOM on the PSL table.
  • The STACIS system still is down, awaiting some attention.
• PSL (Miyakawa, Kawamura, Abbott, Nocera):
  • The PSL ISS system is disabled until Flavio investigates the dynamic range problem.
  • Osamu is working on completing technical notes on the performance of the PSL post-re-layout.
• Digital suspension controllers (Sakata, Goggin, B. Abbott, Heefner)
  • Shihori continues to work on improving the input matrix for ITMY and the remaining suspended optics.
• Optical sensing (Smith, Kawazoe):
  • Fumik