Download Large Optics Suspension Balancing Specification and more Study notes Optics in PDF only on Docsity!
COMPONENT SPECIFICATION
DRWG NO. REV. GID
TITLE
APPROVALS: DATE REV DCN NO BY CHK DCC DATE DRAWN: CHECKED: APPROVED: DCC RELEASE:
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY SHEET 1 OF 21
1 INTRODUCTION
1.1. Objectives and Scope
The scope of this document is to specify how to clean and assemble a Large Optic Suspension. It also details how to prepare an optic for installation into the suspension structure and how to hang and balance that optic.
1.2. Version History
8/21/98: Rev A by J. Hazel Romie. 6/28/99: Rev B first draft by M. Barton - near-complete rewrite. 7/13/99: Rev B final draft. Extensive revision of procedure for installing the wire.More detailed description of autocollimator usage. 1/28/00: Rev C. Major update. 4/18/00: Rev D. Add use of deionizing gun for static charge, extra precautions to identify autocollimator HR surface reflection, non-reuse of clamps.
1.3. Applicable Documents
LIGO-D960132: Large Optic Suspension Assembly, LOS1a LIGO-D970560: Large Optic Suspension Assembly, LOS1b LIGO-D970564: Large Optic Suspension Assembly, LOS1c LIGO-D970572: Large Optic Suspension Assembly, LOS1d LIGO-D970577: Large Optic Suspension Assembly, LOS1e LIGO-D970561: Large Optic Suspension Assembly, MMT3, 4k LIGO-D970578: Large Optic Suspension Assembly, MMT3, 2k LIGO-D970505: Large Optic Suspension Assembly, LOS2a LIGO-D970539: Large Optic Suspension Assembly, LOS2b LIGO-D970507: Large Optic Suspension Assembly, LOS
LIGO-D960133: LOS Structure Assembly, LOS LIGO-D970551: Recycling Mirror Structure Assembly LIGO-D970506: LOS Structure Assembly, Beamsplitter LIGO-D970508: LOS Structure Assembly, Folding Mirror
M. Barton, J. Hazel Romie
LARGE OPTICS SUSPENSION BALANCING
B 9/22/
C
D
E
E
E
J.R.
J.R.
J.R.
D.C.
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 2 OF 21
LIGO-D960145: LOS Height Adapter Assembly LIGO-D970571: LOS1c Height Adapter, Recycling Mirror 4k LIGO-D970579: LOS1e Height Adapter, Recycling Mirror 2k LIGO-D970554: LOS2a Height Adapter, Beamsplitter 4k LIGO-D970555: LOS2b Height Adapter, Beamsplitter 2k LIGO-D970569: LOS3 Height Adapter, Folding Mirror
LIGO-E960022: LIGO Vacuum Compatibility, Cleaning Methods and Qualifications Procedures LIGO-M990034: LIGO Hanford Observatory Contamination Control Plan LIGO-E970153: Large Optics Suspension Quality Conformance Worksheet LIGO-E990196: Magnet/Standoff Assembly Preparation Specification LIGO-E990197: Magnet/Standoff Assembly Quality Control Worksheet LIGO-T960074: Suspension Preliminary Design LIGO-T950011: Suspension Design Requirements LIGO-T970158: Large Optics Suspension Final Design (Mechanical System) LIGO-L970196: Part Numbers and Serialization of Detector Hardware LIGO-E000029: LIGO Large Optic Process Traveler Form LIGO-E000028: Core Optics Tilt Angles LIGO-E000033: Large Optics Suspension Table LIGO-E990035: Large Optics and COC Cleaning Procedures LIGO-E000061: LOS Installation Procedures for HAM Chambers LIGO-E000062: LOS Installation Procedures for BSC Chambers
2 VACUUM COMPATIBILITY REQUIREMENTS
2.1. General Handling
All procedures listed under this specification for assembly and balancing must be per- formed in a clean room environment while suited up in clean room clothing including, but not limited to: frock, shoe covers, bouffant cap, LIGO-approved latex gloves, and facial mask. This applies to anyone handling or near clean pieces or pieces being cleaned.
For further detailed handling requirements see M990034, LIGO Hanford Observatory Contamination Control Plan. In the terminology of that document, the optic, suspension structure and associated parts are Class A hardware (i.e., destined to be installed in vacuum) and once cleaned and baked should not come into contact with anything but Class A and B hardware.
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 4 OF 21
3 QUALITY ASSURANCE/CONTROL
3.1. Documentation
For each optic, make a copy of the LIGO Large Optic Process Traveler form, E000029. Obtain an E-series doc- ument number for the copy, using as a title “Process Traveler for ”, e.g., “Process Traveler for ETM4K, LLO”.
3.2. Quality Assurance Provisions for Sensor/Actuator Heads
Collect five sensor/actuator head assemblies which pass the following tests and record serial numbers and test data in the process traveler for the optic.
- Measure the photodiode output voltage in each sensor/actuator head with the standard current applied to the LED. (This can be done using the suspension controller test stand.) Reject sensor/actuator heads for which the voltage is below 1.5 V.
- Test for continuity of the plating by measuring the resistance from the face to the side of each OSEM.
- Check for shorts between turns of the coil by measuring the resistance between the ends of the wires. Reject coils whose resistance is not approximately 13 ohms.
4 FIXTURES
4.1. Drawing Numbers
D000256 Large Optic Holder, Part A D000257 Large Optic Holder, Part B
D000035: Magnet and Dumbbell Sanding Fixture
D000036: Magnet-to-Dumbbell Assembly Fixture
D990158: Magnet/Standoff Assembly Fixture D990186: “Ears” for use with ETM, MMT3, FM, Dummy Mass D990187: “Ears” for use with ITM4k D990189: “Ears” for use with ITM2k D990188: “Ears” for use with RM
D990190: LOS2 Magnet/Standoff Assembly Fixture (for use with BS)
D960147: Base Plate of Guide Rod Fixture Assembly and blocks for use with ETM, MMT3, FM and Dummy Mass D970574: Guide rod blocks for use with ITM4k D970573: Guide rod blocks for use with ITM2k
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 5 OF 21
D970568: Guide rod blocks for use with RM D970550: LOS2 Guide Rod Fixture for use with BS-4k, BS-2k
D960763: Fixture, Test Mass (a.k.a. the Dummy Mass) D970553: LOS2 Test Mass Fixture (an Al blank the size of the BS)
Microscope, Edmund Scientific, p/n J39- D000013: Microscope-OSEM Bushing PZT Buzzer D970180: Winch Fixture D960753: Fixture, Wire and Optics (Cradle Fixture)
D960145: Height Adapter for ETM, ITM4k, ITM2k D970571: Height Adapter for MMT3-4k, RM-4k D970579: Height Adapter for MMT3-2k, RM-2k D970554: Height Adapter for BS4k D970555: Height Adapter for BS2k D970569: Height Adapter for FM
4.2. Descriptions
- Magnet-to-Dumbbell Standoff Fixture
Used to configure and bond the magnets to the dumbbell shaped aluminum standoffs.
- Magnet/Standoff Assembly Fixture
Used to position and epoxy the magnet/standoff assemblies to the face and sides of the optic.
- Guide Rod Fixture
Used to position and bond a guide rod and a wire standoff to the side of the optic.
- Fixture, Test Mass and LOS2 Test mass Fixture
Used for the prototype test. These aluminum dummy optics have the same size, wedge, chamfer and approxi- mate mass as the ETM and BS.
- Measuring Microscope
Used to align the sensor/actuator plates to the magnet/standoff assemblies glued on the optic or dummy mass.
- Microscope Bushing
Mounted on the measuring microscope and used to adapt the microscope to the bore of the holes for the sensor/ actuator head assemblies in the sensor/actuator plates.
- PZT Buzzer
Used for sliding the wire standoff along the side of the optic to change the pitch balance of the optic. It is a rod or tube to which a PZT is attached. The PZT is driven while the vibrating rod is placed against the end of the standoff to produce small displacements of the standoff.
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 7 OF 21
- Clean fixtures: Clean the base plate and the side blocks of the guide rod fixture thoroughly per LIGO Vac- uum Compatibility, Cleaning Methods and Qualifications Procedures document, LIGO-E960022. Clean and bake the guide rod and wire standoff per the same specification.
- Prepare baseplate: Put the baseplate on a clean work surface with the serial number the right way up as viewed from the position of the worker. (NOTE: subsequent instructions assume this orientation. Engage the two side blocks with the dovetail slides on the guide rod fixture, taking care to match left and right. It is important to do this step before placing the optic, because the blocks foul on the optic when it is in position. Slide the blocks to their lowest position to give plenty of space for bringing in the optic. Cut three small strips (.50” x 1.00”) of Mylar film and place on the etched circle in the base plate. This Mylar is used to pro- tect the bottom face (the high reflective surface) of the optic.
- Position optic on baseplate: Note that one of the four etched lines on the sides of the optic has an arrow- head. For all large optics, this arrow is at the thinnest point of the wedge and points towards the high-reflec- tance or beam-splitting surface. Note also that all large optics except for the recycling mirror are suspended with the thin part of the wedge at the bottom. The optic should be laid on the baseplate such that (i) the face of the optic is concentric with the etched circle, (ii) the HR or BS surface down is down and (iii) what will be the bottom when the optic is later hung is towards the worker. That is, in all cases, the arrow should point towards the floor. Also, with one exception (the recycling mirror), the arrow should be on the side nearest the worker. (For the recycling mirror it should be on the side opposite the worker.) Make fine adjust- ments to the position and angle of the optic until the four etched lines on the side of the optic line up with those on the base plate. Be careful to allow for parallax. Before sighting on the line on the baseplate, posi- tion your eye so that the line on the near side of the optic lines up through the body of the optic with the one on the far side.
- Clean the sides of the optic. Wipe over the sides of the optic with a lint-free wipe dampened with acetone. Repeat with isopropanol or methanol.
- Position left and right blocks: Snug them up against the side of the optic and tighten the screws that hold the top blocks to the base plate to finger tightness.
- Prepare adhesive applicator: Clean the copper wire with acetone and alcohol with lint-free wipe.
- Prepare epoxy: Mix the two epoxy components of a Vac Seal “bipax” together thoroughly, approximately 2 minutes. Dispense from the middle of the container into a boat made of clean UHV aluminum foil. Degas the epoxy for 2 minutes at 10 torr (or the vacuum from any typical backing pump).
- Position and glue the wire standoff and guide rod: Position the guide rod in the smaller vertical v-groove. If there is difficulty inserting the guide rod into the v-groove, move the top block down a bit, along the wedge, insert the guide rod, and then cinch the block back into position, holding the guide rod in the v-groove securely. Dip applicator wire in epoxy and withdraw it, leaving a tiny amount of epoxy on the wire. Apply epoxy on the wire to the vertical line of contact between the guide rod and the optic that is furthest away from the worker. Be sparing in epoxy at this point as more glue will be used later to secure this guide rod. Take care not to get epoxy on the fixture. Insert the wire standoff in the other vertical v-groove and apply epoxy in the same way. Be sure to apply epoxy to the vertical line of contact between the wire standoff and the optic that is furthest away from the magnet/standoff assembly.
- Check the adhesive joints: If preparing an optic rather than a metal dummy mass, look through the optic at the two glue joints and make sure that the glue has spread over the whole area of the joint.
- Cure epoxy: Leave the assembly for 24 hours or more. Vac Seal cures fully in 72 hours so if that time is available, it should be used to allow the assemblies to fully cure. Note that the side blocks which existed as
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 8 OF 21
of this version of the balancing procedure are not rated for an oven cure (as used for the magnet standoff assemblies in the following section).
- Remove the guide rod fixture: After curing, unscrew the fasteners that hold the top blocks to the base plate, one side at a time, and slowly, carefully, move the top blocks down the wedges.
5.2.2. Applying Magnets (With Magnet/Standoff Assembly Fixture)
5.2.2.1 Materials
1 set of magnet/dumbbell standoff assemblies for LOS Perkin Elmer Vac-Seal epoxy resin D990192: Large Optic Holder D990158: Gripper Assembly D990190: BS Gripper Assembly D990186: “Ears” for use with ETM, MMT3, FM D990187: “Ears” for use with ITM4k D990189: “Ears” for use with ITM2k D990188: “Ears” for use with RM D960763: Fixture, Test Mass (Dummy Mass) or Optic or appropriate
solvents: acetone, and isopropanol or methanol 6” length of clean copper wire (never having had insulation) small vacuum chamber with backing pump UHV aluminum foil lint-free wipes oven mitts Teflon or PFA440HP, approx 400 mm x 400 mm x 3 mm (note: only PFA440 is approved for use in the airbake oven) air bake oven lab jack Newport Corp. p/n 281 plate, Newport Corp, p/n 290-TP extensions to bring plate level with oven rack when mounted on the lab jack
5.2.2.2 Fixture Assembly
- Preheat bake oven: Turn on the air bake oven and set the thermostat for 100°C.
- Clean fixtures: Clean the Large Optic Holder and top ring of the magnet/standoff assembly fixture thor- oughly per LIGO Vacuum Compatibility, Cleaning Methods and Qualifications Procedures document, LIGO-E960022.
- Prepare holder: Put the Large Optic Holder on a clean work surface with the two handles at front and back as viewed from the position of the worker.
- Position optic on holder in roughly the same orientation as for gluing the wire standoff and guide rod (see 5.2.1.2). The precise orientation is not critical.
- Prepare the magnet/standoff fixture: Back off the three 1/4-20 set screws in the top ring until the PFA440HP tips are flush with the the inner surface of the ring.
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 10 OF 21
plunger so that it releases the magnet/standoff assembly and withdraw it. NOTE: The plungers must be removed before the optic is allowed to cool, or differential themal expansion will cause joint failures.
- Let the optic cool: Where possible, just slide the optic back into the oven, turn the oven off and allow the optic to cool in situ. If the oven is required again immediately, use extreme caution in removing the optic to avoid burns to workers and damage to the optic.
- Remove magnet/standoff fixture: An assistant to act as a spotter is recommended for this step. Loosen the three 1/4-20 set screws that hold the Magnet/Standoff Assembly Fixture to the optic. Slide the ring off straight upwards, using extreme care to ensure that the ears maintain safe clearance from the side magnets.
- Test the strength of the bonds: For each magnet in turn, allow the flat surface of a clean razorblade to stick to the flat end. Pull the razorblade straight off along the axis of the magnet. If the magnet/standoff assembly survives the bond is adequate. Note the results of the test in the Process Traveler (Appendix A of this docu- ment).
5.3. Suspension Assembly
5.3.1. Materials
D960144: suspension block D970540: LOS2 Suspension Block (for BS) #4 flat washer, stainless #4 lock washer, stainless #4-40x1/2, silver-plated stainless SHCS 1/4 flat washer, stainless 1/4 lock washer, stainless D960134: LOS Clamp, Suspension Block D970180: Winch fixture (2 of) 4 silver-plated 1/4-20 x 1.0” socket head screws D990690: LOS Safety Stop
Figure 1: Polarities of the magnet/standoff assemblies.
StandoffMagnet
S S
S N
N S
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 11 OF 21
D960499: LOS Chamfer Stop D000040: Teflon Screw Tip, Small D000041: Teflon Screw Tip, Large D970563: Beamsplitter Safety Stop D970562: Beamsplitter Chamfer Stop D990691: LOS3 Chamfer Stop deionizing gun
5.3.2. Procedure
- Clean and bake all components of the suspension assembly, except the suspension wire, per LIGO Vacuum Compatibility, Cleaning Methods and Qualification Procedures, LIGO-E960022.
- Prepare the suspension block and winch fixtures: Set the suspension block on top of the LOS structure, and insert the attachment screws with their washers. Before tightening the screws, use a clean set of vernier cal- ipers to ensure that the edge of the suspension block is parallel to the top front edge of the structure. Using oversize washers and 1/4-20x1.5” long screws, attach a winch to each of the threaded holes on the top of the suspension block, with the rocker of the winch closest to the front of the suspension block. Fit suspension block clamps (6 in all) to the suspension block and the winch fixtures, leaving all screws very loose.
- Screw in the 8 chamfer stop screws and the 8 safety stop screws into their respective brackets until they pro- trude past the inside of the bracket by about .25”.
- Replace the Fluorel tips of the lower safety stop screws with the Teflon caps. Wrap the Fluorel tips in clean UHV foil and save. (The tips will be changed back again at the end of installation in the chamber.)
- Install the socket head set screws, into the Sensor/Actuator Brackets in preparation for the installation of the Sensor/Actuator Assemblies. Note: Do not install the height adapter yet. Although it is part of the suspension assembly, if it is installed before the optic is suspended, access to the suspension block is awkward at best and impossible at worst.
5.4. Optic Installation
5.4.1. Suspension Structure Assembly and Optic Hanging
5.4.1.1 Materials
Optic or dummy mass with magnet/standoff assemblies, guide rod and wire standoff from above. D960132: Structure for ETM (LOS1a), ITM4k (LOS1b), ITM2k (LOS1d), MMT-4k, MMT3-2k D970551: Structure for RM4k (LOS1c), RM2k (LOS1e) D970506: Structure for BS4k (LOS2a), BS2k (LOS2b) D970508: Structure for FM (LOS3) 0.012” diameter steel music suspension wire (0.008” for BS) D960755: Large Wire Standoff or appropriate Bubble level Perkin Elmer Vac-Seal epoxy resins 6” length of clean copper wire (never having had insulation)
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 13 OF 21
brackets) and put the cradle on top. Adjust the height of the top of the baseplate by placing one or more lint- free wipes under it so that it is level with the bottom plate of the LOS structure and the cradle can slide smoothly from one to the other.
- Move the optic onto the cradle: This step assumes the optic starts lying HR (or BS) side down on the base- plate of the gluing fixture. Standing the optic up is possibly the most risky part of the procedure and should be carefully rehearsed. The side magnets are at the 3 o’clock and 9 o’clock positions around the edge and are extremely easy to knock off. The sequence of holds described below keeps hands away from the mag- nets and eliminates awkward wrist angles, thus minimising the risk of injury to personnel and damage to the optic. One person should pick up the optic with hands placed at 6 o’clock and 12 o’clock. The stronger hand (i.e., typically the right hand) should be at 6 o’clock and the fingers should be horizontal, i.e., tangen- tially around the edge of the optic. The first person should then rotate the optic so that it is standing up, with the weight on the hand at 6 o’clock. A second person should then take the optic, using hands at 4:30 and 7:30 and with fingers pointing parallel to the cylinder axis. The second person should have the face magnets towards them so they are easy to see and cannot be knocked off if their fingers curl behind the optic. The second person should place the optic on the cradle with the arrow exactly at the bottom (or exactly at the top in the case of the RM).
- Move optic into place: A person to act as a spotter is recommended for this step. Rotate the cradle on its baseplate until the face magnets on the optic point towards the back of the structure. Slide the baseplate up to the bottom plate of the structure. Carefully slide the cradle off the baseplate, onto the bottom plate and into the structure, checking continually that the optic does not bump into any of the brackets, until it is sym- metrically located both left-to-right and front-to-rear relative to the safety stop brackets. Screw in the lower safety stops until they touch the optic. Then screw each stop in one turn further so that they lift the optic about 1 mm above the cradle. Push the cradle out from underneath the optic towards the front of the struc- ture (not the back as in previous versions of this procedure) and remove it. Screw in the chamfer stops and the upper safety stops so that there is a 1-2mm gap between the optic and the end of the stop.
- Post a person to safeguard the side magnets: Have a second person watching throughout the next few steps to make sure the wire does not come too close to the side magnets. This requires a lot of care because the wire is magnetic. The assemblies should withstand the wire being pulled off straight in any direction against the force of the magnetic attraction. However the assemblies are no match for a taut wire trapped on one side and wanting to be on the other side. Be especially careful to avoid the wire getting hooked in the waist of the dumbbell standoff.
- Take up the slack in the wire: Move the bottom of the wire loop forward into the gap between the front and rear safety stop brackets. Pull the ends of the wire so that the loop tightens around the optic. On the side of the optic with the wire standoff, let the wire drop into the groove in the standoff. On the other side, let the wire rest on the guide rod, a few mm to one side of the magnet standoff assembly. Tighten one upper sus- pension block clamp, taking care to keep the wire centered between the screws. Apply moderate tension with fingertips to the end of the wire on the other side and tighten the other upper suspension block clamps, again taking care to center the wire between the screws. It should be possible to pull the wire a few millime- tres away from the optic using only tweezers.
- Insert the second wire standoff: Hold the wire away from the side of the optic (and the magnet!) with a pair of tweezers and insert the second wire standoff below the guide rod but snugged up against it, using a sec- ond pair of tweezers. Position the wire standoff so that its central groove is directly below the centre of the guide rod, and make sure the wire drops into the groove. If the standoff is difficult to insert or if it will not
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 14 OF 21
hold its position beneath the guide rod, adjust the tension in the wire slightly with the winch fixture and try again.
- Suspend the optic: Slowly, lower the safety stops that support the bottom of the optic and suspend the mass. Try backing off the chamfer stops and note which direction the optic tends to tip. Use the PZT buzzer to micro-position the wire standoff so that the optic/dummy mass will hang stably at very roughly the desired angle (as judged by eye). An alternative technique is to emulate a diamond cutter and to use the handle of a screwdriver (or similar) as a mallet and a long pointed (clean) tool to transmit the force to the optic. Use the de-ionizing gun on the optic to make sure stray static charges have not affected the balance.
- Check the line of the wire underneath the optic: Check that the part of the loop below the standoff lies in the same plane as the part above, as nearly as possible as judged by eye. If the wire bends at the standoff when viewed from the side, it will cause a substantial pitch of the optic which is likely to change over time as the wire relaxes. A bend of the wire sufficient to displace the bottom of the loop about 5 mm from the centreline is equivalent to about 3 mrad of pitch. If the wire is bent, raise the lower safety stops to take most of the ten- sion off the wire, and push it into line using the PZT buzzer. Even if the wire is not out of place, gently apply the PZT buzzer to the entire length of the wire against the glass so as release any tension.
- Set up the winch fixtures: The cradle leaves the optic slightly higher than its working position so it is conve- nient to set the initial position of the winches so there is plenty of scope for lowering. (This may not apply for the BS, which is inserted manually.) For each winch fixture in turn, hold the back of the lever up under the head of the main screw (it tends to fall under its own weight) and adjust the main screw until the the lever is slightly nose up. Still holding the lever in position, put tension on the end of the wire and tighten the wire clamp on the nose of the winch.
- Do a preliminary adjustment of the position of the optic in roll and vertical using the winches. Put sensor- actuator head assemblies (without PAM screws) in the brackets. Rock them back and forward on the rails in the brackets until they settle snugly. Check visually that there is a full line of contact between the rails and the body of the assembly. Do not bring them close to the magnets at this point - the risk of damage is too great. Leave several mm clearance - somewhat more than the chamfer stop to optic distance. Sight through the sensor-actuator heads, controlling for parallax by positioning your eye so that the aperture at the far end of the sensor-actuator head is centred in the aperture on the closer side. Adjust the lengths of the wires on each side until the magnets on the face of the optic line are centered in the square holes for the sensor/actu- ator head assemblies. At this stage, concentrate on roll, that is, make sure that the magnets on opposite sides are in the same relative vertical positions with respect to the corresponding sensor/actuator heads. The final absolute adjustment in vertical can only be done after the final pitch adjustment, because the lever arm of half the thickness of the optic times a typical pitch angle amounts to a non-negligible vertical displacement of the magnets.
5.4.3. Optic Balancing
5.4.3.1 Materials
Optic suspended in Large Optic Suspension Structure Perkin Elmer Vac-Seal epoxy resins 6” length of clean copper wire (never having had insulation) UHV aluminum foil
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 16 OF 21
on an internal surface and is ten or twenty turns from the proper focus). With the corner cube back in place, check that the crosshairs are aligned with the dot in the centre of the scale. (If not, get the autocollimator realigned/repaired.) Remove the corner cube.
- Find a reflection from the optic in the autocollimator. Let the optic hang freely. Rotate the goniometer until a scale due to reflection from the optic is visible in the eyepiece. (Because of the by-eye horizontal align- ment of the structure and autocollimator specified above, you should only have to search vertically.) If you need to rotate by a large angle, undo the goniometer lock screw and rotate the goniometer stage with your hands - don't use up the limited range of the fine adjustment screw. Refocus the autocollimator (it should only take a turn or two).
- Identify whether the reflection is from the HR or AR surface. Get a rough idea of whether the scale is from the front (HR) surface or the back (AR) surface by looking by eye from the side and noting which surface is more nearly orthogonal to the body of the autocollimator. Confirm the identification by moving the autocol- limator to bring the other scale into view. For the usual case of a thick-side-up wedge (all optics but RM), to go from the HR reflection to the AR one the nose of the autocollimator has to be lowered by an amount equal to the wedge angle times the refractive index (1.46). Identifying the HR reflection correctly is partic- ularly important for ITMx2K, which has a very small wedge angle. The two reflections are normally differ- ent colours, e.g., orange and pink. Note the colour of the HR refection for future reference.
- Align the autocollimator and optic horizontally: As well as the scale from the HR face, there will be one from the AR face. Use the adjustment on the holder to improve the alignment of the autocollimator with the optic horizontally (still ignore vertical).
- Set up the reference: Method 1 (for standard metal optical table) - Clamp the optical flat to one of the refer- ence surfaces of the right-angle block. Set up the beamsplitter holder on a post in front of the autocollima- tor. Set the right-angle block on the holder with the optical flat towards the autocollimator and the other reference surface uppermost. Level the reference surface in both directions with the bubble level (not criti- cal). Method 2 (for granite table) - make sure the granite TRIsquare is sitting snugly on the table. Clamp the optical flat to one of the reference surfaces.
- Align the reference horizontally: Rotate the reference in yaw until the optical flat is facing the autocollima- tor as accurately as can be judged by eye. Rotate the autocollimator in pitch until it is as accurately horizon- tal as can be judged by eye. Look in the autocollimator and continue adjustments in yaw of the reference and pitch of the autocollimator until the scale from reflection of the optical flat on the reference can be seen. (Because of the by-eye alignment in the previous two steps, this should be close in both directions.) Refo- cus the autocollimator. Continue fine rotation in yaw until the scale is centred horizontally (not critical).
- (Method 1 only) Level the upper surface of the right-angle block as accurately as possible in pitch with the bubble level. The zeroing of the bubble level should be checked independently beforehand by the usual method of measuring a flat surface and then repeating with the level reversed. Do not reverse the level while it is on the right-angle block - the resulting mechnical perturbation will undo any benefits of the check. Leave the bubble level in place for the next step.
- Decide the autocollimator angle and check the sense of goniometer adjustment screw: Check the sign of the desired angle of the HR surface in E000028. If the HR surface of the optic is to face upwards, the nose of the autocollimator must dip to point directly at it. Conversely, if the HR surface must point down, the auto- collimator must tilt up. Check the sense (clockwise or anticlockwise) of the rotation of the goniometer adjustment screw that is required to tilt the autocollimator by an easily visible amount (a degree or more) in the desired direction.
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 17 OF 21
NOTE: there are two versions of the next step depending on the magnitude of the desired tilt. The autocollima- tor scale is somewhat more accurate than the goniometer scale and should be used where possible, but the goni- ometer scale has a larger range and is needed in some cases.
- Set the autocollimator to the desired angle: CASE I - the magnitude of the desired tilt is within the range of the autocollimator scale (30 minutes). Ignore the scale on the goniometer. Using the goniometer adjustment screw, first centre the autocollimator scale vertically (not critical for Case I). Then, turn the goniometer adjustment screw in the direction established in step 7 until the crosshair indicates the correct magnitude as given in E000028 (critical). CASE II - the magnitude of the desired tilt is outside the range of the autocolli- mator scale. Set the goniometer stage to exactly zero as indicated on the goniometer scale. Using the verti- cal adjustment screw on the autocollimator holder, centre the autocollimator scale vertically (critical for Case II). Turn the goniometer adjustment screw in the direction established in Step 7 until the goniometer scale indicates the magnitude of the desired tilt (if the rotation takes the goniometer scale to the positive side of zero, or 360 degrees minus the magnitude of the desired tilt if it moves to the negative side of zero and so wraps around).
- Remove the reference block.
- Find the HR scale again: The initial position of the optic will typically put the scale outside the viewfinder. Have an assistant gently touch the optic on the chamfer at top and/or bottom with a gloved finger until the scale from the HR reflection enters the viewfinder. Refocus the autocollimator. Note the direction that the optic needed to be pushed.
- Adjust the position of the wire standoff to tip the optic in the same direction as above until the scale is visi- ble with the optic hanging freely.
- Double check the right reflection is being used. Check that the colour is right for the HR reflection. View the optic from the side and double check that the relative tilts of the two faces are correct as near as can be judged by eye (have the correct sign, etc). Have an assistant press with a gloved finger on the face of the optic at the top or bottom to ensure that the AR reflection is where it is supposed to be. For the usual case of a thick-side-up wedge (all optics except RM), the HR face of the optic has to be tilted down to bring the AR reflection into view.
- Continue fine adjustments until the scale is centred to 0.5 minutes or better (critical).
5.4.3.3 Cementing Wire Standoff
- Prepare adhesive applicator: Clean copper wire with acetone and isopropanol using a lint-free wipe.
- Prepare epoxy: Mix the two epoxy components of a Vac Seal “bipax” together thoroughly, approximately 2 minutes. Dispense from the middle of the container into a boat of UHV aluminum foil. Degas the epoxy for 2 minutes.
- Apply epoxy: Dip conductor wire in epoxy and withdraw it, leaving a tiny amount of epoxy on the wire. Apply epoxy on the wire to the top side and ends of the wire standoff. Apply epoxy to the unglued end of the opposite wire standoff to secure it better.
- Gently clamp optic: When the optic is balanced, gently move the chamfer stops near the face of the optic, just until contact is made. Make sure that the alignment doesn’t change. (If the optic is fully clamped, the alignment will change upon the adhesive curing.)
- Cure epoxy: Let the suspension sit for at least 24 hours (72 hours or more if possible).
- Remove optic: Raise the lower safety stops until they support the optic and take the tension off the wire. Cut
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 19 OF 21
- Assembly
- Prepare the sensor/actuator heads and cables: Clean the assemblies along with the cables per LIGO Vacuum Compatibility, Cleaning Methods and Qualification Procedures, LIGO-E960022.
- Check the position of the safety and chamfer stops. A gap of 1mm must be maintained to protect the magnet/standoff assemblies during this procedure.
- Mount the sensor/actuator heads: Check that #10-32 set screws in the threaded holes that will hold the sen- sor/actuator assemblies in place are flush with the inside diameter of the sensor/actuator assembly mounting hole in the sensor/actuator bracket. Make sure that the optic/dummy mass is fully clamped. Mount the sen- sor/actuator assemblies in the proper configuration. Slowly, slide the sensor/actuator assemblies into the holes in the bracket until about 2mm of sensor/actuator assembly protrudes beyond the back of the sensor/ actuator brackets. The pigtails should be at the top side of the upper OSEMS and the lower sides of the lower OSEMS.
- Connect the sensor/actuator assemblies to the test stand. The orientation and order for plugging the pigtail connectors into the DB25 connectors is given in Fig. 2.
- Position the sensor/actuator assemblies: Using an oscilloscope, position each sensor/actuator assembly with respect to the magnet/standoff assembly on the optic such that the magnet shadows the photodiode and pro-
LR
UL UR LL
S
Wire direction
J
J
Pin 1 side
Figure 2: Sensor/actuator connections for the test stand, looking into the DB25 connectors. The small connectors are oriented so that the pigtail wires leave from the bottom.
COMPONENT SPECIFICATION
DRWG NO. REV. GID
CONTINUATION SHEET TITLE
CALIFORNIA INSTITUTE OF TECHNOLOGY MASSACHUSETTS INSTITUTE OF TECHNOLOGY
LARGE OPTICS SUSPENSION BALANCING SPECIFICATION
SHEET 20 OF 21
duces 50% of the unblocked voltage (measured earlier). Make fine adjustments in roll of the assembly until the screws in the back of the assembly are level in height, as nearly as can be judged by eye. After each adjustment in either position or roll, rock the assembly back and forward on the rails in the brackets until it settles snugly. Check visually that there is a full line of contact between the rails and the body of the assem- bly. Use the set screws to clamp the sensor/actuator assemblies in their optimum positions.
- Install the PAM screws: Check that the sensor/actuator assemblies damp properly and that critical damping may be achieved. Screw the assembled PAM screws into the back of the sensor/actuator assemblies until the distance from the ceramic body of the assembly to the underside of the PAM screw head is 0.85”.
- Adjust the PAM screws to correct for any pitch error shown by the autocollimator and optical lever.
- Iterate: Correcting a pitch error with the PAM screws will move the optic magnets away from their optimal positions relative to the sensor/actuator assemblies. However moving the sensor/actuator assemblies to cor- rect for this will move the PAM magnets and perturb the position of the optic. Repeat steps 5 and 7 at least once or until convergence is adequate. Note the final photodiode voltages and pitch angle in the Process Traveler.
5.5.1. Measure resonant frequencies
5.5.1.1 Materials
spectrum analyzer, e.g., Stanford Research Systems, SRS
5.5.1.2 Measurement
- Connect the spectrum analyser to the satellite box test output for a face sensor/analyzer assembly.
- Turn off the damping.
- Do a power spectrum with a start frequency of approximately 0.4 Hz and a stop frequency of 1.0 Hz. There should be three peaks corresponding to the pitch, yaw and pendulum modes at roughly the frequencies given for the optic in the table in T970158. Record the frequencies in the Process Traveler. If there are any major discrepancies, especially in the pitch frequency, investigate the position of the wire standoff and guide rod.
- Do a power spectrum with a start frequency of around 10 Hz and a stop frequency of around 14 Hz. Look for a peak corresponding to the vertical mode at the frequency given in T970158. Record the vertical mode frequency in the Process Traveler.
- Do a power spectrum with a start frequency of around 15 Hz and a stop frequency of around 19 Hz. Look for a peak corresponding to the roll mode at a frequency approximately sqrt(2) times that of the vertical mode. Record the roll mode frequency in the Process Traveler.
- Connect the PZT buzzer to the source of the spectrum analyser.
- Do a swept sign measurement from around 300 to 350 Hz with the tip of the PZT buzzer pressed gently against one of the wire locating pins of the suspension block. Look for a peak (or a cluster of very closely spaced peaks) due to the violin mode of the wires. Ignore peaks at multiples of 60 Hz. Record the violin mode frequency in the Process Traveler.
