LIGO Weekly Report

Monday, May 27, 2002 is a holiday. There will be no Executive Committee Meeting.

Special Announcements:

LSC Issues (Weiss)

LIGO I Construction/LIGO Laboratory Administration (Lindquist)

There was a site teleconference held on Thursday, May 23, 2002. The following items were discussed:

Action 109: estimates for furnishing the new buildings have been provided. This action is closed. Suggest change requests should be submitted.

Brunt Construction (Livingston): objecting vociferously about items being added to punch list. However, these are items that are not new, but could not be checked previously (e.g., roll up door not tied down prperly and vibrates when air conditioning turned on). Rest of list is being done except for a few contentious items that probably will not get done (e.g., crane stops).

Audio Visual Procurement: received four proposals. Lot of disparity in cost of labor proposed.

The list of current actions revised to reflect the status of open actions assigned through May 23, 2001 may be found at ACTION LIST.

Web pages for the DCC give simple how-to's for document numbering, easy access to the latest on-line documents, and search capabilities for the DCC database. Take a look. . .

Advanced LIGO (Frey)

August 1 we are scheduled to submit an annual work plan to the NSF for LIGO FY 2003 Operations. We are modifying the first draft budget for FY 2003 based on the current staffing plan.

Quality/Safety (Tyler)

LIGO Hanford Observatory (LHO) Operations (Raab)

LIGO Livingston Observatory (LLO) Operations (Coles)

GC: Setting up a testbed for working with VPN configurations. The more I read about VPNs on the Cisco PIX the more evident it becomes that we will have to implement NAT and static routing on the PIX. I am going to try and set up a VPN independent of the PIX using Solaris' implementation of IPSec and SunScreen Lite. I will be using Solaris for Intel, an Ultra 10, and a blade 100 for the test. I am investigating LDAP for a global address book for the observatory. Fixing a couple of problems with the LLO web page. Working with Bonnie to show her what she will have to learn and what is involved in maintaining some of the web pages here at LLO. (Shannon Roddy)

Seismic Amelioration: Work continues on gathering quotes from vendors on the heat treat, coating, and joining of the spring assemblies. Additionally, Efforts to Wire EDM the pre-isolator spring have been started with Reliable EDM of Houston. I am waiting to here back from them very soon.

Most of my week has been spent working on implementing the pre-isolator design into the BSC and HAM stacks then utilizing techniques to successfully produce a finite element model that reflects the physical model. This is still part of the effort refine the 6x6 stiffness matrix that I have been working on.

Efforts also include the ongoing search for a materials test lab that can test to failure the pre-isolator springs. LSU has two tensile testers: One will test to 80 kips while a large on will test to 500,000 lbs and has a six inch stroke on the load cell. The professor in charge is out of town but will return Tuesday. I am still looking for a commercial solution.(Marcel)

SCIENCE & ENGINEERING SUPPORT (Coyne)

Seismic Upgrade Project

Dennis Coyne reporting

External Pre-Isolation

Mohana is doing the rack-mounted receiver interface module pcb layout and is in the process of choosing the chassis. The boards for the L4C/IPS interface are here as well as the chassis to put them in. We will stuff a board and test it immediately.
Ken Mason revised the spring/actuator assembly to incorporate a support brace from wall-to-wall to increase the stiffness. FEA on the stiffness is pending. This brace also permits (with set screws) complete assembly on the bench prior to installation. The coarse actuation changes have delayed the major pieces in fabrication by ~ 2 weeks.
Double-start, counter-wound spring (Marcel Hammond):

1. Fabrication (traditional NC machining) has begun on a single prototype in maraging steel (300?) from bar stock. Machining expected to be completed by 5/23. In parallel investigating EDM as a potentially cheaper alternative. Also getting quotes (price & delivery) for maraging in pipe form to reduce machining costs. A heat treat process has been defined by the maraging steel provider. Looking for a heat treater for the prototype (2 requests for quote out – pending reply). Plan to chromium plate the prototype, as recommended by Dan DeBra. Cleaning/passivation process defined by the maraging steel provider. Plan to lead-silver solder join the end caps to the spring body. Plating will be done before soldering – the soldered areas will be masked off.

2. Stiffness and frequency analysis: updates still in progress. Need to get a better “read-out” of the rotational compliance. Need to perform an analysis of the integrated spring/structure assembly.

Rich Mittleman will post the measurement data for the LASTI BSC SEI driving point impedance measurements for the team to review and compare to analysis.
Connectorized L4C interface: design still pending.

Hydraulic External Pre-Isolation (HEPI)

Hydraulic Pump Station (Ken Mailand, Rich Abbott):

Jonathan Kern is working with Ken Mason and Myron MacInnis at MIT this week. Corwin has completed the changes to the bellows shield assembly we agreed upon, and handed over to us his complete hydraulic actuator CAD model. Ken has imported Corwin's changes into his spring/actuator assembly for the Hydraulic External Pre-Isolator (HEPI) and we're working to check the integration.
Jonathan, Ken Mason & Myron are also going over the installation procedures for LASTI; seeing how much of the PEPI fixturing and procedures will be adaptable to HEPI. Our discussions have made plain that we need to develop a test and measurement to carefully equalize the tension of the EPI springs to assure no yaw is introduced when we release the stops.
Jonathan completed re-templating Amit's nozzle drawing, and ordered a quantity of nozzles for LASTI. Amit is working on a procedure for adjustment and calibration of the servo valve. Parker informs us that an order for the assembly of the servo valves with our custom nozzles will take 3 to 4 weeks.
Passive RC manifold (“block”): performing analysis. Will send a copy of the prototype to Ken Mailand.
Distribution manifold design: pending.

Electro-Magnetic External Pre-Isolator (MEPI)

Received electro-magnetic actuator order.

Greg has taken measurements of the interaction between the EM actuator and the L4C geophones. Data reduction pending.

CDS Software

Dennis Coyne summarizing from elogs at LLO:

Rolf installed new LSC software. This is essentially the same code as LHO2k, except TM and BS output filters, which are hardcoded in LHO2k, are now user defineable. This code change required an MV162 to be installed as the EPICS interface. The l1iscsuper is still in the crate, but is not being used.
The front end CPU is still a 1GHz Pentium, but it tends to become fully loaded with only 4-5 test points. Changed out l1lscl0 for a new 1.2 GHz pentium. Only gained a test point or two before we saw ADC noise again.
Rana locked the LLO IFO using the new LSC code. No real problems except for the 'too many test points making noise' problem. Implemented some new filters giving x10 more low frequency gain and more phase margin for the DARM loop. We were able to use the new 'on the fly' feature to switch the old filters with the new ones with the IFO in lock.
Installed new variable timing delay modules (“dual” timing, i.e. allows the DAC clocking with a time shifted clock). Installed the LSC front end code which makes use of the new timing module. Mike has modified the two pentek modules in this crate to use the separate DAC clock inputs. However, this version of the software only makes use of the timing module polling (instead of polling the ADC) and does not yet make use of the separate DAC clock. I want to test in this mode first to see if the ADC noise is reduced.
A summary of the software changes and the state of the commissioning of the software is given in this elog entry (username: reader, password; readonly)

CDS Hardware

Jay Heefner

Differential Driver/Receivers: Two of each type of module have been tested. The output/input referred noise is ~100nV/rtHz which is sufficient for operation with both the pentek ADC and DAC. The bandwidth is ~10KHz. Common mode rejection for freq<10KHz is greater then 60dB and is typically greater then 80 dB. 30 of each type of board are bring stuffed in preparation for installation.
Optical Lever Whitening: The schematic for the new board is 75% complete and should be ready for final review by next week.

Rich Abbott, Falvio Nocera, Ben Abbott

Progress made on installation of new VME based Intensity Stabilization Servo (ISS) by Flavio, Peter King and Rich Abbott. First look at RIN before the MC is at 2e-7 as measured by independent PD. Initial TFs being taken and more to follow. All VME based controls are functional (Gain, test relays and read back channels). Still need to work on stabilization after the MC which will be attempted on Thursday. We worked through some problems with the DCPD that we are using, and now it seems to function well. The DCPD is a fully differential design (by Ben Abbott) and incorporates features which allows the servo card to send out an active AC coupling signal to the DCPD to implement auto-zero function.

Sander Liu

Micro-Seismic Signal Processor:

Remote interface Box - 12 units are being build. Chassis design is done, will review all dimensions called out before sending out for fabrication.
Post Processor - Testing of the first article is complete. Will begin to build 12 units as soon as man power is available.

PSL

PeterKing

The PSL Lab's BeamScan diagnostic has been returned for repair to Photon Inc. A broken cable is the prime suspect reason as to why the head is spinning faster than the nominal 10 Hz.

Optics Analysis

Erika D’Ambrosio

I spent some time in revisiting the work done on the parametric instability. I sent out the proof to Physics Letters A, which has accepted the article and is ready to publish it.

I am also trying to collect the large amount of informations on the sidebands imbalance in order to present those in a simplified and organic form, for Bill to check over when he is back. I decided not to include any estimation of what the impact of the sidebands imbalance would be on the strain sensitivity. This delicate point is something Bill and I have a lot to work on, because there is an overall issue related with the bandwidth of the sidebands, for the isolated recycling cavity and for the full interferometer locked case, that in the past has been considered the same. In the meantime I have been checking some of the analytical results on the sidebands imbalance by the FFT-code, so that all the statements we are claiming, based on our mathematical analysis are tested by a more sophisticated numerical tool.

Optical Contamination Cavities

Lee Cardenas, Liyuan Zhang

Cavity#1 Locked.

Ring Down ~ 24.85 x 10^-6 microseconds ± 2.1 x 10^-8 microseconds. Vacuum pump ~ 4 x 10^-8 torr (previously ring down ~ 23.32 microseconds with pump 4 x 10^-6 torr).
Beat Frequency ~ 86.0815 MHz.

Cavity#2 Locked and with mineral test sample in it. Exactly one month of pumping.

4/18/02 Ring down ~ 33.1217 microseconds, Vacuum pump ~ 7 x 10^-9 torr (without test sample) - Beat Frequency ~ 74.4098 MHz.
4/22/02 Mineral oil test sample installed into cavity chamber.
4/23/02 Ring down with mineral oil test sample ~ 30.8409 microseconds, after one day of pumping Vacuum pump ~ 4 x 10^-4 torr.
4/29/02 Ring down --------------------------------- ~ 31.6308 microseconds.
5/22/02 Ring Down --------------------------------- ~ 32.0185 microseconds, after one month of pumping Vacuum pump ~ 7.5 x 10^-9 torr.
Ring down getting normal with time and vacuum pumping.
Beat frequency ~ 74.3987 MHz.

RGA is finally working and the sum of AMU(41,43,53,55 & 57) ~ 5 x 10^-9 torr.

[D. Coyne Note: The results from the mineral oil sample are still being evaluated. Another week or two of data will be taken. Initial results at high pressure indicated a significant increase in absorption rate. At low pressure the rate is not yet discernable. Results yet to be properly scaled to LIGO conditions.]

Transmission Test Set-Up.

Breadboard optical table 2ft x 4ft x 2 inches thick acquired.
Design and fabrication of a steel base support table is on going. Material ordered and welding will proceed.
Most of the Optics (1/2 wave plate, Isolator, mirror, lens, photodetector and lock-in amplifier) are ready.
Attenuator and chopper needed.
NPRO Laser Model: 126-1064-700, S/N 415, obtained from Shanti. Beam waist measurement will proceed.

Misc.. Tasks.

Installed a metro rack next to the cavity #2 and all the equipment from the floor are up on the shelf. This created more room for the transmission test set-up.
Cleaned and fixed the lab.

DMT

John Zweizig

This week I have been writing an alarm handler which will be used to summarize abnormal conditions seen by the DMT monitors and to notify operators of these conditions. I have also written the client interface API that wil be used by the monitors to communicate with the Alarm Handler.

40 Meter Interferometer (Weinstein)

Suspension Controllers and Mode Cleaner length controls (B. Abbott, Heefner, Taylor, Ugolini)

Optics (Billingsley, Armandula): Helena has compiled all the documentation on the polished and coated optics that she has inspected. REO will start coating the core optics on July 8th. It will take approximately 3 weeks to coat all the parts. The ETMs will be coated first, followed by the ITMs, the Beam Splitters and then the SRM's and PRM's. They are working on the tooling and everything is going smoothly so far.

Computing: Lisa Bogue and Mike Pedraza has set up a secure wireless access point on our martian network (40MARS), and installed a wireless laptop for use as an EPICS monitor screen. Two more laptops will be installed soon.

Thermal Noise Interferometer (Libbrecht)

This week we put together a number of measurements, made over the last few weeks, into a breakdown of the different noise sources that contrubute to the TNI. We have a number of total noise curves, going back to December of last year, and we thought it would be interesting to try and identify the limiting noise sources in various frequency ranges. Attached are two plots, one for the South Arm Cavity (SAC) taken in December of last year and one for the North Arm Cavity (NAC) taken less than a month ago. Each plot includes both a total noise curve, measured with the arm cavity locked, and a series of noise curves from different sources.

The first, and perhaps most obvious, noise source is laser-frequency noise. This is shown in purple on the enclosed plots and was obtained by measuring the free-running laser frequency noise (see last week's report ) and then dividing by the appropriate loop gain for the mode cleaner. In the North Arm Cavity (NAC), laser-frequency noise is dominant from about 5kHz to 20kHz, and the less-than-optimal phase margin at the unity-gain frequency (just over 10kHz) accounts for the observed bump in the total noise. SAC's laser-frequency noise is lower than NAC's because the gain of the mode cleaner was higher when the SAC data was taken, and the better phase-margin prevented a bump at the mode cleaner's unity-gain frequency. SAC's total noise in December was dominated by laser-frequency noise over a fairly narrow band between 6kHz and 10kHz, as expected.

The second noise source we looked at was laser-frequency noise imposed on the laser by seismic noise in the mode-cleaner cavity. We were only able to measure this up to about 45Hz, and it shows up at the same level in both measurements. This noise source is shown in red on both plots.

The third, and most interesting to us, noise source is from the electronics in the servo filters, shown in green on both plots. SAC's noise curve was dominated by electronic noise only at comparitavely high frequencies (above 12kHz), and the use of a passive notch filter in December gave rise to the bump at 27.5kHz.

NAC, on the other hand, is dominated by electronic noise over many frequencies and at a comparitavely high level! This is largely due to the fact that NAC's optical gain is lower than SAC's, but particulars to the design of the NAC servo that make it easier to lock than SAC's December configuration contribute also.

The straight lines are theoretical thermal noise expectations for SAC, also plotted with NAC's data for comparison: bulk thermal noise in red, and the upper bound set by SAC in cyan (blue-green). Niether curve accounts for coating thermal noise. It is evident from these plots that electronic noise must be reduced substatially in NAC before any thermal-noise measurements can be made, and SAC's electronic noise could be reduced as well.

LASTI (Zucker)

External pre-isolators: Jonathan Kern visited us this week from LLO. He visited the LASTI lab, reviewed the drawing packages, and also discussed installation staging and fixturing, piping standards, service procedures and hydraulic fluid leak containment with us.

Ken Mason is at Southbridge Sheet Metal today inspecting the first article EPI weldments; expected to ship next week if all is well. BEI has shipped the remainder of the MEPI actuators and they arrived in good shape (no chipped magnets seen so far).

Prestabilized laser: Dave and Jamie achieved success with the improved frequency stabilization servo. Unity gain bandwidth of 400 kHz was achieved using the new circuit topology, with further increase likely if some trivial photodetector-related limits are also taken care of. Unfortunately, an electronics fault (possibly a blown PA-85) happened right after these new results, so there is a delay in fully characterizing the new topology.

Data Analysis and Computing (Lazzarini)

LIGO II/Advanced R&D (Sanders)

Working on stiffening up the MC suspension structure.
Submitted drawing formats, instructions for use and drawing practices documents.

For the last 2 weeks I have been working at the IGR at the University of Glasgow. In the course of my visit I attended the PPARC funding review for the IGR; visited the Rutherford Appleton Laboratory in Cambridge, England; spent the day with Dr. Caroline Cantley, a senior Engineer who is about to start work in the Institute; attended one of the days at the annual GEO 600 Review; presented the drawings and plans for the mode-cleaner control prototypes; discussed the plans and time scales for the electronics that GEO will supply for the controls prototypes for MIT and CIT; and gave a presentation on how to use the MATLAB pendulum design model.

I will submit separate reports regarding my visit to Rutherford and discussions arising from the drawing review and electronics update.

I also helped Russell Jones, Alastair Grant, Mike Plissi work on several experiments and prototypes, for advanced LIGO, in the laboratory. Over the course of the 2 weeks we: -
1) tested several ways of adjusting the height of the cantilever blades.
2) designed and tested an improved method for adjusting and clamping the position of the coil former in its clamp
3) set up a planned test of some eddy current dampers on a triple pendulum. (Firstly we had to set up and install a new XYZ drive system under an existing prototype triple pendulum before we could install the new eddy current dampers and adjustable arms.)
All of the parts are almost in place for testing this effect in both the longitudinal and vertical directions.

As an aside Russell Jones, the young engineer at Glasgow is now trained in SOLIDWORKS to level I, same level as Janeen and I. They also hope to have a second young engineer trained in the next couple of months.

Rich Mittleman has implemented the proposed upgrade to the quad at MIT in order to try and solve the problem of the undamped pitch mode. The improvements are noticeable and Norna, Rich and I have been looking at these over the last day or so.

Janeen and I are working on the proposed changes suggested at the drawing review at Caltech and Glasgow. In particular I have been working with Russell Jones in Glasgow on the new drawings and an assembly of the new coil clamping and adjusting mechanism.

Phone conference with Goodrich regarding further compensating polish: We have decided that it will probably be a longer run and, therefore, cost more to do the measurement as specified, i.e., subtract side one from the bulk measurement. I will change the specification to state that subtraction of Side 1 is not required, but may be done to more accurately represent the bulk compensation. Of course if they can get under 10 nm rms without subtracting side one, all the better, it's certainly an easier measurement analysis.

Goodrich will provide surface measurements per the specification and their prior bid. These measurements may be done after the next round of polishing.

Goodrich does have data from a visual inspection which will be sent. This inspection was beyond their typical inspection. A 200W light was used and linear sleeks were found front and back. Linear sleeks are not something one would expect from CCP.

We at Caltech have done a visual surface inspection on the 250 mm sapphire piece: Side one is in worse shape than side two, it shows a lot of circular sleeks of diameter ~12 mm as well as a faint spiral pattern. Side two showed a stronger spiral pattern, but only faint circular sleeks similar to those on side one.
There were bubbles breaking through on both sides (no surprise there!) We did not find scratches on either side.

This is a significantly different finding than Goodrich. Perhaps it would be a good idea if I made a trip out there when the optic goes so that we can compare techniques.

Goodrich will put together an estimate for the next run, the previous guess was that the run would be about 11K.

Goodrich will put together an estimate for the cost of compensating Sapphire of similar quality, but 314 x 130 mm in size. They are to assume they would compensate the central 300 mm diameter, but will note roughly how the cost would change if only the central 215 were compensated. They are to assume an initial surface of 100 nm rms figure and 4 angstrom roughness. They will also assume a requirement of < 10 nm rms transmitted (S2/Bulk) and 4 angstrom roughness (similar to the current spec goals).

MLD's coatings break down when annealed at temperatures higher than 500 degrees C. To determine the optimum temperature to anneal the "Q" substrates coated with 30 alternate layers of Ta2O5/SiO2, witness pieces coated on the same run have been annealed as follows:

The next run at MLD will be 30 alternate quarter wave layers of NbO2/SiO2. The substrates are at their facility waiting to be coated.

1) Ordered and received resistors and thermistors to set up the mock heat capacity puck which will be then used as the test bed for preparing for the measurement. Mapped the resistivity puck channels to be used for the mock heat capacity test. Ready to put things together next week.

2) With the help of Phil calculated the resonant frequency for the Metglass (used the young's modulus and other constants for the MoRuB) to be 676 Hz for a 25mm long, 3mm wide (these dimensions were from the rough setup Mike and Kelin put together) and 27 micron thin reed. This was done for building the actuation setup for Q-factor measurement. Recalculation will be done, with different reed dimensions to bring the resonant frequency down.

1) Brian from Quantum Design (QD) installed yet another cryopump which had been tested at QD. At Caltech we found that the flapper valve failed once and the pressure gauge was faulty this time. Brian is returning with a newly tested, this time more thoroughly, cryopump next week.

2) Borrowed a Pirani gauge and controller etc. from Steve Vaas to continue systematic debugging of the system. At Eric's advice we are running measurements to pinpoint which unit, out of cryostat, controller and the computer, is really faulty.

3) Mike was appointed the point man here at Caltech to keep track of QD and cryostat in a meeting with Eric and Ken. Mark Seeback (Customer service head) and Stefano Spagna (chief Engineer) were contacted at QD to help solve the urgent problem. QD has offered us some time on their company cryostat in San Diego for conducting our measurements if the system is not fixed next week.

1) Changed different parameters, on the machine and sample, for splatting MoRuB samples in order to get an optimum amorphous surface from the splat. Finally produced a splat with almost 80% amorphous surface which is waiting to be scanned once I get some free time on the X-ray diffraction machine.

2) Prepared samples of appropriate size for MoRuBSi and MoRuBSiAlP to be splatted and checked. Preparing more MoRuB alloy as the one big chunk from Carnegie Mel. does not seem to be uniform in composition.

4) Checked the (AuSn)Ti braze by etching the Molybdenum alloy with Murkami's reagent and the braze with HF. The material was brazed under Argon atmosphere. The eutectic of AuSn is at around approximately 231 degrees Celsius but braze did not wet the alloy even at 250. I tried ramping the temperature by 20 degrees to 400 but that did not help either. Jan has a similar experience with braze and suggested that I go to higher temperatures. Will test the braze next week with all four different etchants I prepared and also will continue ramping the temperature until the right one is found.

5) Started getting ready the sample of MoRuB for the conductivity measurement at QD.

1) Setup an actuator test for Q-factor measurement. Used capacitive force to oscillate a Metglass reed but the maximum voltage supply available (32V) did not produce a deflection. Phil is providing with a power supply of over 100V next week.

2) Looked for a commercial actuator but did not find an appropriate one. On Phil's advice will just use a simple Cu plate for capacitive actuation.

3) Working on getting things ready for the installation of Q-factor measurement in Phil's vacuum chamber.

Exec. Comm. Agenda	Highlights	LSC	Administration	Hanford Observatory
Livingston Observatory	MIT	Caltech	Detector	40 Meter
TNI	LASTI	Data Analysis	LIGO II/Adv. R&D	Past Weekly Reports

05/23/02	Packages	Faxes
In	22	33
Out	12	25

Weekly Report for Week Ending May 23, 2002

Special Announcements: