h. mainaud durand, be-abp/su, 23/04/2010 status of the clic pre-alignment studies

Alignment Work in TT2 and PS During the Shut-down 2006-2007

H. MAINAUD DURAND, BE-ABP/SU, 23/04/2010STATUS OF THE CLIC PRE-ALIGNMENT STUDIES

2Survey and alignment for CLICFor the CLIC project, Survey people will have to align (measure the position and adjust) all beam components or their associated supports:

In all the area of the tunnels (Main beam injectors, drive beam generation complex, main linac, Beam Delivery System, return loops,)

72 000 components or supports (data on Sept. 2007) 52 km (data on Sept. 2007)

On the ground, on the ceiling (transfer lines), in loops (return loops, damping rings)

Within various precision and accuracy ranging from 3 microns to 300 microns.

3Survey and alignment studiesSome priorities have been given concerning the survey and alignment studies:

Prove the feasibility of the pre-alignment of the components of the main linac within a precision and accuracy of 3 microns over a sliding window of 200m along the whole linac

Propose a solution for the pre-alignment of the CLIC final focus (QD0)

Propose a global solution of alignment

Integrate this solution and make it compatible with other services.

4SUMMARY

General strategy and required solutionsIntroduction of the challenge and hypotheses considered for the studiesGeneral strategy Required solutions

R&D program and first resultsSolutions for the determination of positionSolutions for re-adjustmentSpecial case of MDI area

Solution proposed for the CDR Studies proposed for the TDR5Introduction of the main challengeWithin +/- 0.1 mm (1s)Mechanical pre-alignmentActive pre-alignment Beam based alignment Beam based feedbacksWithin 3 mm (1s)PRE-ALIGNMENT (beam off)

Pre-alignment of 3 microns along a sliding window of 200m, all along each linac

After computation, for a sliding window of 200m, the standard deviations of the transverse position of each component w.r.t. the straight fitting line must be inferior to 3 microns6Introduction: why active pre-alignment?At a micron scale:Ground motionNoise of acceleratorTemperature dilatationsContinuous determination of the position of the componentsRe-adjustment when necessaryConsidering the number of components supports to be aligned Considering the resolution of displacement requiredActive pre-alignment=Determination of the position of the components in a general coordinate system thanks to alignment systemsRe-adjustment thanks to actuators+7Budget errors: RF structures:

Introduction: hypotheses for study3 m over 200m (1) is a target for study and development.trade off with beam dynamics for realistic and achievable values for CDR.From D. Schulte, CLIC meeting, 04/05/2007.8

Budget errors: MB quad:Budget errors: MB BPM:From D. Schulte, CLIC meeting, 04/05/2007.9

From D. Schulte, CLIC Module Review, 15/09/2009.Budget errors Main linac mover requirements Addressed by the Stabilization Working Group10 m10SUMMARY



Solution proposed for the CDR Studies proposed for the TDR11

General strategy : re-adjustmentSeveral components will be pre-aligned on supports: Along the MB: Along the DB: RF structures on girders PETS + DB quad on girders MB quad on interface plateDB and MB girders will be interlinked with their extremities, based on so-called cradle. This allows a movement in the transverse girder interlink plane within 3 degrees of freedom (articulation point between girders). (Longitudinal direction adjusted thanks to a mechanical guiding). MB quad is mounted on an interface plate, allowing an adjustment along 5 degrees of freedom (longitudinal position will be positioned manually).

12General strategy: determination of the position of the componentsInstallation and determination of the surface networkTransfer of reference into tunnelInstallation and determination of the tunnel networkAbsolute alignment of the elementsSmoothing of the elementsActive pre-alignmentControl and maintenance of the alignmentFor the main linac

13General strategy: determination of the position of the componentsGeodetic Reference Network (GRN)

Backbone for all the tunnels and areasWill allow the installation of all services and of the MRN

1314General strategy: determination of the position of the componentsGeodetic Reference Network (GRN)Metrologic Reference Network (MRN)

As it is not possible to implement a straight alignment reference over 20 km: use of overlapping referencesTwo references under study: a stretched wire a laser beam under vacuum14

15General strategy: determination of the position of the componentsGeodetic Reference Network (GRN)Metrologic Reference Network (MRN)Support Pre-alignment Network (SPN)

1516General strategy: determination of the position of the componentsGeodetic Reference Network (GRN)Metrologic Reference Network (MRN)Support Pre-alignment Network (SPN)Alignment and fiducialisation of each component on the supports (AFC)

16 Required solutions: feasibility of the concept17Active pre-alignmentDetermination of the metrological network w.r.t the straight alignment referenceDetermination of the position of each sensor w.r.t metrological networkFiducialisation: determination of the zero of each component w.r.t the sensor (external alignment reference)Re-adjustment: displacement of the component supporting structure according to the sensor readingsStable alignment reference, known at the micron levelSubmicrometric sensors providing absolute measurementsMeasure 2m long objects within a few micronsSubmicrometric displacements along 3/5 DOFSTEPSISSUESOther issues:Compatibility with the general strategy of installation and operationCompatibility with other accelerator equipment or services1718SUMMARY



Solution proposed for the CDR Studies proposed for the TDR

Development of sensors whose resolution is submicrometricDevelopment of alignment methods associated with these sensors (double stretched wire method)Development of an active pre-alignment systemTests of the sensors and validation of an active alignment in a real environment (CTF2)Feasibility of controlling submicron movement.Required solutions: lessons from the past (1990->2002)

In the CTF2 facility, the components (CAS, PETS) were maintained aligned in a closed loop w.r.t. a stretched wire within a window of 5 microns, thanks to sensors and micro movers, in a very radioactive environment.Sensors and actuators were used for monitoring w.r.t a reference position and repositioning, not for pre-alignment

The pre-alignment in that case was manual and iterative (no fiducialisation : the position of the WPS was not known w.r.t the reference axis of the accelerating cavities) Small scale solution to align the accelerating cavities on the girders Mechanical design to update (modification of the size of the components considerable increase of load some question marks concerning the clearances and kinematics) A solution of fiducialisation within a few microns must be found A new solution of pre-alignment for the MB quad must be found (CTF2 solution not compatible with stabilization requirements)BUTRequired solutions: lessons from CTF221Introduction: state of the artAccelerator / ProjectAlignment requirementsStatusSolution for measuring the positionPerformanceSolution for re-adjustmentPerformanceLCLS (SLAC)Stability over 1 week period.Relative monitoring over 140m < 2mWorkingWPS (RF inductive)HLS (capacitive, ultra-sound)Relative monitoring < 2m over 1 weekCam mover2 m ?SLS (PSI)Monitoring of the girders positionWorkingHLS, HPSNever usedCam moverNever usedXFEL (PSI) = 100 m (inter girder) in lateral and heightUnder developmentIn vacuum laser reference line (or stretched wire)To be developpedDriving wedge0.1 mm (manual adjustment)LHC = 100 m rms over 100m (radial and vertical)WorkingStretched wire + optical levellingBetter than 0.1 mmIndian jacks< 0.05 mmLHC low betaMonitoring of a triplet within a few micronsWorkingWPS (capacitive)HLS (capacitive)Relative monitoring < 2m over 1 month

Motorized indian jacks< 0.01 mm (smallest step)DESY XFELRequired accuracy: 0.5 mm w.r.t 1000 m (photon beam lines)Under developmentPoisson alignment system (stretched wire as alternative)Development under progress??Monitoring requirements (after Beam Based Alignment ) and not pre-alignment (before BBA)Precision and not accuracy Precision over less than 200 m, during short periods

No solution answering the requirements exists22In 2005, when the CLIC studies on pre-alignment re-started, the following directions of studies were proposed:

Concerning the measure of the position of the components:Improvement of the stretched wire solutionDevelopment of a laser based solution (as B-plan)Find methods of fiducialisation and pre-alignment of the components on the supports

Concerning the re-adjustment:Find a new solution for the re-adjustment of the MB quadRe-design/Re-sizing the cradle inter-linking two girders Required solutions23SUMMARY


R&D program and first resultsSolutions for the determination of positionCollaboration with NIKHEFLAMBDA projectStretched wire solutionSolutions for re-adjustmentSpecial case of MDI area

Solution proposed for the CDR Studies proposed for the TDR24Status of the different solutionsMetrologic Reference Network (MRN)Support Pre-alignment Network (SPN)Alignment and fiducialisation of each component on the supports (AFC)

Stretched wire, modelized in vertical with HLS systemLong range RASCLIC (NIKHEF)Multi-point alignment system based on laser beam.No solution answering the requirements exists yet!Stretched wire RASNIK, short range RASCLIC (NIKHEF) Multi-point alignment system based on laser beam.

Not ready for CDRNot ready for CDR2425SUMMARY



Solution proposed for the CDR Studies proposed for the TDR26Collaboration with NIKHEF3 main objectives in 2005 :

A paper study on a low-cost RASNIK system to be implemented all along the linac as SRN (including an outlook for the relevant developments to be expected within the 10 years).

The development of a sub-nanometer RASNIK

The development of a laser alignment system, as MRN:Capable of giving a straight reference on more than 100m, with an accuracy of a few micrometers.Allowing to perform an inter-comparison with stretched wire measurementsAnd the study of its implementation along 40 km.

Not addressedRASNIK27Collaboration with NIKHEF

Results: A resolution of position of 20 nm was reached in TT1The system appeared to be a very good low-frequency seismographOn short distances, vacuum is no more needed: a short range RASCLIC could replace the RASNIK system foreseen as SRN.

Remaining issues: Vacuum is needed on long distances, which implies a sophisticated mechanics between the component to be aligned and the RASLIC partConfiguration: must be implemented as leap frog. No flexibility when distances are not constant.

RASCLIC28Collaboration with NIKHEFProposal for a new collaboration between CERN and NIKHEF around 3 WP:

Adapt the long range RASCLIC for CLIC and validate itCalibrationValidation through an inter-comparison with stretched wires (in TT1, TZ32)

Adapt the short range RASCLIC for CLIC configuration and validate itCalibrationValidation through an inter-comparison on the two beam module prototype (in lab and in CLEX)

Propose a solution for the monitoring of the QD0 on both sides of detector.

29SUMMARY




30LAMBDA Project

Technical note concerning the concept proposal ready Preliminary study concerning the mechanical shutters will start soon Definition of the subject for a doctorate student in progress

LAMBDA for Laser Alignment Multipoint Based Design Approach

31SUMMARY



Solution proposed for the CDR Studies proposed for the TDRStretched wireMain issue: long term stability of a wire (effects of temperature, humidity, creeping effects, air currents)Modelization of the wire using Hydrostatic Levelling Systems (HLS)

but only in the vertical directionbut HLS system follows the geoid which needs then to be known studies undertaken concerning the determination of the geoid

Is a stretched wire really straight (radial direction)?First idea: comparison with a laser beam under vacuum (NIKHEFF)Inter-comparison of different types of wires and technologies on short distances (50 m) this summer at SLACSubject of a thesis: Determination of a precise gravity field for the CLIC feasibility studies Sbastien Guillaume.32Stretched wire and MRNMinimum configurationTT1 facilityObjectives:To determine the precision and accuracy of a MRN consisting of overlapping stretched wiresTo study the behavior of wires of different lengthsTo study the modelization of a stretched wireTo build and validate the laser solution through an inter-comparison

33

34

Stretched wire and MRNFirst results in TT1Subject of a thesis: Analysis and Modeling of the effect of tides on the Hydrostatic Levelling Systems at CERN Julien Boerez.

(rad)=0.4 m , (vert)=0.6 mfor T=0.2C and humidity=7.7% 35Stretched wire and MRNResults on 500 mKnow-how developed for the installation of long wires (500 m)

36Stretched wire and MRNLatest results in TT1Precision on a 140 m wire: better than 2 microns over 33 daysAccuracy: 11 microns in vertical, 17 microns in radial. Can be improved!

Vertical residuals of the 2 longest wires: (wire 1) = 1.6 m (wire 2) = 0.5 m

Accuracy of the TT1 network adjusted by the least squares method in vertical: = 11 m r.m.s (27 m max. value)Subject of a thesis: proposal of an alignment method for the CLIC linear accelerator: from the geodetic networks to the active pre-alignment Thomas Touz.Sub-micrometric sensorsIssue: WPS sensor fulfilling the requirements absolute measurements (known zero w.r.t mechanical interface) no drift sub micrometric measurements

Upgraded WPSOptical based WPS (oWPS)

Capacitive based WPS (cWPS)Resolution: 0.2 mRange: 10 x 10 mmRepeatibility: 1 mBandwidth: 10 Hz

3738Status of the different sensors technologiescWPSoWPSTechnologyCapacitiveOpticalAccuracy (m)710Repeatability (m)17Precision (m)17Acq. Frequency (Hz)1000.1Resistance to radiation200 kGy (sensor)500 Gy (remote electronics)?WireCarbone peekVectranSag (mm) for 200 m76.5 mm45.5 mmCost5000 CHF500 $

3839Inter-comparison between sensors:Web site: https://clic-pral.web.cern.ch/clic-pral/

Status of the inter-comparison: WPS [SLAC, CERN] : inter-comparison at SLAC. Facility ready in July 2010.HLS [Fermilab, SLAC, DESY, SSRF]: long term stability tests at Fermilab, other tests at CERN.40Status of the different solutionsStrategy towards the feasibilityOnly one solution ready for CDR : stretched wire + WPS cWPS: design of a new mechanical interface and associated calibration bench improve the accuracy (target : accuracy < 5 m) cost study in definition for 50 000 units.oWPS:definition of the next generation under progress

Stretched wire for MRN and SRN

4041Status of the different solutionsStrategy towards the feasibilitySeveral studies launched to address the drawbacks of the WPS and stretched wireDevelopment of a concept allowing to stretch a wire without access to the sensors and the wire protectionA method was found to stretch two wires in the same optical sensor (to be validated on long distance)Determination of the local deviation of vertical (so to have a perfect knowledge of the geod)

4142Objectives:Find the best strategy and configuration of alignment systems for the pre-alignmentModel the impact of the pre-alignment errors on the beam emittance growth

First results:The pre-alignment tolerance could be achieved with wires longer than 425m.Beam simulations, based on these data showed that 400m wires were able to limit the long distance emittances.

Next steps implementation of a new model, closer to the realityThe requirements were defined in collaboration with the Beam Dynamics WGThese steps are now mathematically definedThe algorithms are being implemented and tested on the TT1 configuration (validation of the model on 150m)Development of a software allowing the modeling on the whole CLIC.

Simulations close to the reality4243Algorithm of simulations

44Summary concerning the determination of the position of the components for the CDROn surfaceIn the tunnel

4445SUMMARY


R&D program and first resultsSolutions for the determination of positionSolutions for re-adjustmentStrategyStatusFirst resultsValidation on mock-upsSpecial case of MDI area


Strategy of re-adjustmentMB Quad // cam moversDB and MB girders // linear actuatorsDesign of a new cradle (MME design office)9 high resolution linear actuators ordered this week (ZTS)Design of the mock-up (MME design office)5 high resolution cam movers ordered (ZTS)Validation of a SLS type cam mover (1 DOF mock-up)Validation of on a 5 DOF mock-upValidation on the two beam module prototype DR approved by the finance commitee Market Survey to be prepared and sentValidation on a pre-alignment mock-upValidation on the two beam module prototypeRe-installation of CTF2 re-adjustment solution Cost for 5 cam movers (one support): 52.5 kCHF Cost for 3 linear actuators (one girder): 42 kCHF46

47

Foreseen Tests will give answers to the following questions:

Max achievable accuracy with SLS mockup in case of 1DOF Modal behavior as a function of load mass (50kg 150kg) Analysis of backlash and clearance in entire assembly Further improvements on mechanical design and components

Validation of the cam system on a 1 DOF test setup47

48

Next steps:

Installation of capacitive sensorsReplacement of the pendulum bearingDisplacement curve (understanding of error due to planar tolerances on hardened surface and guidings)Repeatability and reproductibilityModal behavior as function of load mass

Status of the 1 DOF test setupObjective gained experience will impact the design of the first 5 cam system Mechanical assembly ready Command software readySome first mechanical improvements:

Replacing and optimizing pendulum bearing ( random error due to off centered gravity vector in case of non horizontal operation)Spinea drive as replacement for planetary gear box (high reduction ratio, zero backlash, high kinematic accuracy, high moment capacity, low cost motion, high stiffness)Reduction of longitudinal dimensionsReduction of transverse and vertical dimensionsAxial bearing pressure improvement with threaded flange and defined torque.48Under progress

49

Design of the 5 DOF setupNext steps:

Integration of the WPS sensors with wire stretcher and associated supports Upgrade of the design of the cam, according to the solution chosen

Designed to provide stiffness Deformation on self weight < 1 m in the center

49

Validation on mock-upsStudy of the CTF2 solution (old girder + equipment above ~ 40 kg)

Installation of 2 girders + cradles, with the associated actuators and sensorsConfiguration:

Objectives:Impact of millimetric displacements on cradle M on the other cradlesImpact of such displacements on the extremities of girdersBetter knowledge of the actuators and feedback for the next technical specificationBetter understanding of the girder behavior for the upgraded design

50Validation on mock-upsStudy and validation of the upgraded CTF2 solution (new girder + equipment ~ 800 kg)

Installation with 2 girders and 3 cradles, with Ves and dummy cavities (same weight)

Objectives:Validation of repeatable micrometric displacementsValidation of the fiducialisation strategyValidation of the pre-alignment strategyBetter knowledge of the costsFeedback for the CLIC two beam module prototype

51Test program on CLIC two beam module prototypeValidation of the repositioning concept (possibility of sub-micrometric displacements)Before the installation of all other systems (waveguides, vacuum,)After installation of all other systems (waveguides, vacuum,)Measurement of the eigenfrequencies of the girdersValidation of the fiducialisation strategyValidation of the stability of the components on the girdersImpact of the transport on a micrometric pre-alignmentImpact of variation of temperature, thermal cyclesFeedback for the CLEX test module, and all associated technical specificationsFeedback for the general strategy of installationFeedback for the schedule

End 201052Proposed solution for two beam module prototype

Inter-comparison between Support Reference Networks (SRN)This inter-comparison on short range would allow:To deal with the integration issues (book the space needed for the alignment systems)To deal with the environment issues (noise, EMC from other equipment in lab, radiation/magnetic fields in CLEX)To have a better idea of costsTo validate the procedure of pre-alignment, the schedule foreseenTo test the SRN on a real sequence of modules.To validate the concept of short range measurements on a real size mock-up and to get a better idea of all technical problems needing to be solvedTo validate the use of such alignment systems with the repositioning solution, to validate the algorithm of repositioning.535354SUMMARY



Solution proposed for the CDR Studies proposed for the TDRCase of the final focusDetermination of the position of QD0 w.r.t other components of the BDS (500 last meters)

stretched wire + WPS. (stretched wire of 500m, demonstrated in TT83)

Remaining issues: 10 microns (rms) concerning the position of the zero of QD0Integration

Monitoring of the position of one QD0 w.r.t the other: Solution based on RASNIK system, through the detectors (using dead space between detector areas)

Remaining issues:Perform simulations to find the best configurationValidate the proposed solution

Re-adjustment solution: cam movers for 5 DOF

Remaining issue:Integration5556SUMMARY



Solution proposed for the CDR Studies proposed for the TDRSummary: proposed solution for CDRDetermination of the position of the components: stretched wire + WPS sensors for MRN and SPN

Re-adjustment: MB quad: cam mover Girders: high precision linear actuators57Summary: alternative studies for TDRDetermination of the position of the components: In collaboration with NIKHEF (definition of the WP under progress):Design of a short range / long range solution adapted for CLIC requirementsIntegration of the short range solution on the two beam module prototypeInter-comparison of the long range solution in TT1 / TZ32 tunnels Validation of the concept of the multi-point laser based solutionRe-adjustment: Validation of the concept of articulation point with cam movers

Other studies (in order to reduce the number of sensors): Study of a mono-girder (DB & MB components) Study of longer girders58

59Work of a team:CLIC team (100%):Sylvain GRIFFET [pre-alignment on girders, fiducialisation] (fellow)Sbastien GUILLAUME [determination of geoid] (doctorate student)Pierre HUGON [transfer of reference into tunnel] (student)Juha KEMPPINEN [cam movers, mechanical shutters, linear actuators,] (fellow)Thomas TOUZE [Metrologic Reference Network] (doctorate student)

Julien BOEREZ [dynamic geod] (50% on CLIC, doctorate student)Friedrich LACKNER [cam movers, LAMBDA project] (fellow, < 01/03/10)

MTI (Micro Technology and Instrumentation, unit in ABP/SU):Patrick BESTMANN [oWPS]Andreas HERTY [inter-comparison, cWPS]Mark JONES [geodesy]Antonio MARIN [electronics lab]Sylvain MICO [electronics] (FSU)Michel ROUSSEAU [mechanical workshop]

h. mainaud durand, be-abp/su, 23/04/2010 status of the clic pre-alignment studies

Documents