40m.wbs 40m upgrade task list v1.0 AJW, 7/21/99 Here, for your perusal, is a first stab at a 40m upgrade task list. Its purpose is for you to tell me about gross oversights, errors, etc. There are varying levels of detail on the various subsystems. I plan on fleshing this out using MS Project software. Budget, schedule, manpower - in progress. building modifications - do detailed lay out of laboratory real-estate, including electronics racks: rack locations, power feeds to racks, cable trays, operator consoles, work areas - remove dog-house from 40m building - re-roof 40m building - re-roof 40m north annex building - move north wall of North Annex building northward, - add double door entrance - Remove existing partition 10' from existing north end. - If sharing with metrology lab, add partition in appropriate place. - remove partition between 40m vacuum system and present control room. - upgrade ac power in present control room, to handle electronics racks - seal or remove north door - seal all conduits from machine shop - lay additional cable trays - procure LIGO type racks vacuum envelope - remove all equipment and stacks from vacuum envelope - store all detector equipment, electronics, tools - connect output chamber to vacuum envelope - regenerate 5 ion pumps - bake-out entire system with heater jackets, under clean-room conditions - devise a better way of removing doors from chambers - thorough baffling design for the beam-splitter area and for extracting signal beams seismic stacks - order metal springs and flourel seats - There are existing 3-leg seismic stacks in the 5 core optics chambers and in one one-leg stack in the smaller input optics chamber - disassemble existing seismic stacks in 5+1 chambers - build one-leg seismic stack for new output chamber - design and machine leg masses - optical table - support beams? bellows? - vacuum-clean and bake all springs, seats, stack elements, tables - re-install all 7 seismic stacks under clean-room conditions vacuum controls - procure and install automatic gate valves for 5 ion pumps - regenerate 5 ion pumps - retain all existing valves, gauges, pumps, manifolds, interlocks - replace hardware-> computer interface equipment - gauge controller analog outputs -> patch -> VME SADCs - valve & pump on/off/readback relays -> patch -> VME bio - VME crate and baja cpu running vxworks, epics - epics database - epics gui - epics state machine code to enforce correct transition steps - epics alarm handling - epics archiving - hardware and software interlocks PSL - King & Abbott build 6w 1064 nm PSL, including control system, new CDS rack. - write laser safety SOP before installation - remove existing 40m green PSL from table - re-use all useful components! - remove old PSL cooling system - install new PSL cooling system - Install laser, optics, etc on existing 40m PSL table - Install PSL control system in nearby rack core optics - procure glass blanks - Corning glass for 2 ETMs, SEM (?). - Heraeus SV glass for RM, BS, 2 ITMs. - determine radius of curvature - superpolishing at General Optics, or CSIRO in Australia - determine transmission - design optical coatings, AR - use realistic Fresnel numbers! - apply optical coatings -> REO (Research Electro-optics), Boulder - clean and bake for vacuum - install on mechanical suspensions Input Optics - 1m mode cleaner - IO mode-matching telescope - fixed mirrors - pockels cell following MC - cooling in vacuum - commercial RF signal generator - faraday isolator - matching telescopes core optics suspensions - mechanical design - Scaled LOS - mechanical construction - OSEMs - cabling suspension controllers - spec and design filters, whiteners, signal conditioning - 7 satellite suspension controllers - 7 suspension controllers - suspension IO vme crate, cpu, ADC, DAQ, BIO Question: 6 susp controllers can be read out with one VME crate. Can we squeeze a 7th? I think so! ISC optical tables - Procure and install three new optical tables, one each for dark port, bright port, pickoff port - procure and/or build, for each table: - one LSC RFPD - WFS diodes (5 total for the 3 tables) - WFS Guoy telescopes (one for each WFS) - one QMPD - one ccd camera - 4 motorized mirrors - various fixed mirrors and beam splitters - associated cabling to the LSC, ASC, and slow control systems. ISC readout electronics - procure/build and install new QMPD processing units (3 units), (amplifier, AA-LPF, 4 ADC channels in ASC vme crate). DAQS channels per QMPU: 4 @ ?. - procure/build and install new WFS processing units (5 units), (PD amplifier, demodulator, AA-LPF, 12 ADC channels in ASC vme crate). DAQS channels per WFPU: 8 @ 16KHz, 4 @ 10 Hz. - SDAC in ASC VME crate to control 4 motorized mirrors / ISC table. - procure/build and install new LSC RFPD processing units (3 units) (amplifier, AA-LPF, 3 ADC channels (I,Q,P) in LSC vme crate). DAQS channels per LSC: lots. - procure/build and install CCD cameras for 3 ISC tables, associated electronics (?). Optical levers - The following reflection oplevs (G=global, L=local) already exist, each with laser, QMPD, fixed mirrors, xy electronics, and asc electronics: EEG, EEL, EVG, SEG, SEL, SVG, SVL, BSG(ref), BSG(tran), RMG (there is no EVL, BSL, or RML). - Plan to decommission the local loops, new suspension local control takes over. - retain (re-install) the 7 global loops (EEG, EVG, SEG, SVG, BSG(ref), BSG(tran), RMG) each with table, laser, QMPD, fixed mirrors. - procure/build and install new QMPD processing units (7 units), (amplifier, AA-LPF, ADC channel in ASC vme crate). (LIGO-T970062-00.pdf, p.14). - one additional global loop system for SEM. Other optical readout - tables and PD's exist for: EAT, SAT, MCT light power. - procure/build and install new electronic (3 units) (amplifier, AA-LPF, ADC channel in ASC (?) vme crate). ASC - ASC system receives input from 5 WFPU's, 3 QMPD's, and, optionally, 7 oplev QMPD's. - need 5*12+3*4+7*4 / 32 = 4 FADC's. - Need SDACs and BIOs for control of motorized mirrors - VME crate(s?), cpus, ADCs, DACs, BIOs - ASC software LSC - ASC system receives input from 3 LSC RFPD processing units, each with 3 ADC channels (I,Q,P) = 1 FADC. - VME crate, CPUs, Reflective memory, GPS, trigger boards - FDAC for signals to 6 (or 7) SUS controllers, and PSL. - LSC software DAQS - Signals to analog DCUs: - PEM : accel, seism, mic, mag, temp = 20 signals - SUS-COIL_readback: 6 signals * 7 suspensions = 42 - SUS-COIL_sensors: 5 signals * 7 suspensions = 35 - LSC: analog signals = 8 - PSL: analog signals = 16 - IOO: analog signals = 8 TOTAL: 129 = 4 or 5 FADCs - Signals to digital EDCUs: - LSC: digital signals = 18 - ASC_OPLEVs: 3 signals * 8 levers = 24 - ASC_QMPDs: 6 signals * 3 ports = 18 - ASC_WFSs: 10 signals * 5 sensors = 50 - ASC_controls: 2 signals * 7 suspensions = 14 TOTAL: 124 = 1 or 2 CPUs - frame builder VME CPU - VME crates, CPUs, Reflective memory, GPS, trigger boards - data storage - raid disk farm, tapes - data distribution - GDS ______________________________________________________________________________ From raab_f@ligo-wa.caltech.edu Thu Jul 1 17:51:19 1999 To: Alan Weinstein From: Fred Raab Subject: Re: 40m upgrade task list Remember to: -devise a better way of removing doors from chambers (The craneage in 40-m does not allow for proper, straight-up lifting or doors. Here we remove doors with a forklift that has a boom that reaches inside cleanroom curtains. Since your doors are lighter, an oddly proportioned engine lift should do well.) -write your laser safety SOP before you start to lay out your laser table and enclosure, otherwise it will cause you heartache and money later -use realistic Fresnel numbers when specifying your reflective optical coatings -do a thorough baffling design for the beam-splitter area and for extracting signal beams (We did not do enough to suppress scattered light in previous incarnations of 40-m and it burned us when we executed recombination. You should strive to do better and the LIGO implementation has evolved a design strategy that should be applied back to the 40-m.)