1. logo area Long magnet manufacturing plan and QC by US HL-LHC-AUP Dan Cheng For the MQXFA team MQXF International Review – CERN, June 7 th – 10 th, 2016

2. logo area Outline  LARP magnet structures development at LBNL  Requirements and Reviews  Assembly Tooling and infrastructure  Assembly Breakdown Structure and documentation  Manufacturing plan  Status  Summary D.W. Cheng – MQXF International Review, June 7th-10th 20162

3. logo area LARP History D.W. Cheng – MQXF International Review, June 7th-10th 20163 LQ Long Quadrupole LQS01a-b/02/03 Some accelerator quality features Length with cos2  Accommodating variability in coil dimension Length LR Long Racetrack LRS01-02 Length demonstration HQ High-field Quadrupole HQ01a-b-c-d-e HQ02a-b / HQ03a Accelerator quality features Mechanical alignment High stress regime Time 2004 2014 2005 2006 2007 2008 2009 2010 2011 2012 2013 0.3 m 1 m 3.6-3.7 m SQ Subscale quad Concept TQ Technology Quadrupole TQS01/02a-b-c/03a-b-c-d Concept on cos2  Technology selection Ultimate stress exploration Scale-up & design optimization 2015 1.5 m MQXFS MQXFS1/D1 MQXFA Length scale up 2016 4.5 m

4. logo area MQXFA Structure Assembly D.W. Cheng – MQXF International Review, June 7th-10th 20164 Shell-Yoke Structure –Half-length subassemblies –Joined shell-yoke subassembly, full-length Coilpack –Coil pack subassembly Load pad stacks –Collar pack subassembly Collar stacks Instrumented and dressed coils Master Key packages –Load keys, alignment keys, shims Axial load –Axial rods, [end plates, wire guides] Splice Connection box (Not shown) Magnet support ring (Not shown) –Instrumentation connector skirt

5. logo area MQXFA Requirements  Requirements already agreed upon:  Aperture (T) = 150 mm  Gradient (T) = 143 T/m  Magnetic Length (T) = 4.2 m  O.D. (T) = 614 mm  Fringe Field (O) < 50 mT at 0.5 m  Operating Temp. (T) = 1.9 K  Yoke Channel Diameter (T) = 77 mm (for HX tubes)  Cooling Provisions (T):  Polyimide-free area: 40% inner coil surface  Free passage through the coil pole and G-10 alignment key, equivalent of 8 mm diameter holes repeated every 50 mm  Free Helium paths interconnecting the yoke cooling channel holes  Free cross sectional area of at least 150 cm 2  Maximum ΔT at cool down: 100 K  Ramp rate (T): 14 A/s  Maximum operating voltage during quench: 520 V to ground  Instrumentation (T): Voltage taps D.W. Cheng – MQXF International Review, June 7th-10th 20165

6. logo area MQXFA Requirements  Requirements already agreed upon (Cont.):  Quench memory (O): reach operating current after thermal cycle with no more than 1 quench  Ramp Down (T): Not quench while ramping down at 300 A/s  Radiation dose (O) < 30 MGy  Powering cycles (O) < 3,000  Quenches (O) < 50  Interfaces (T):  Cold Mass, CERN cryo system, CERN power system, CERN quench protection system, CERN instrumentation system  Details defined in interface document (to be written)  Safety (T): complies with CERN’s “Launch Safety Agreement” (LSA) for IR Magnets (WP3) D.W. Cheng – MQXF International Review, June 7th-10th 20166

7. logo area Reviews and Action Items  Summary of Internal Review July 2015 and Workshop Feb 2016  Key action items  Tooling effort, and how this was addressed  Alignment strategy D.W. Cheng – MQXF International Review, June 7th-10th 20167

8. logo area MQXFA Internal Review, July 2015  Summary of Internal Review and action items status D.W. Cheng – MQXF International Review, June 7th-10th 20168

9. logo area MQXFA Action Items from Feb 2016 Workshop D.W. Cheng – MQXF International Review, June 7th-10th 20169 DoneNotes Verify axial preload conditions with SST rods  Fabricate Nitronic 50 endplates for CERN  Not shipped yet Check coil stress with SST LHe containment  Check SG versus model during bladder operations ☐ In Progress Perform tolerance analysis on dummy coil case (and full tolerance analysis) Will also investigate impact of azimuthal tolerances ☐ Eng. Note in Progress Update analysis with adjusted modulus  Evaluate possible use of extruded bladders  CERN Check rigidity of integration table 

10. logo area Assembly Tooling and Infrastructure  Assembly processes flow  Bare coil -> dressed coil  QA  Yoke-shell (sub) assembly  QA/alignment checks(?)  Coil-pad assembly  QA  Final assembly  Alignment  QA  Pizza box  Tooling described  Lessons learned  Developed from the Prototypes currently being built D.W. Cheng – MQXF International Review, June 7th-10th 201610

11. Fiducializing the Magnet Structure Feb 3 2016 MQXFA Magnet Fiducialization and Alignment11  Yoke/magnet fiducial features must be accessible after magnet has been assembled, as depicted here Coil targets defined to their mechanical center Yoke survey block will be defined to coil targets Yoke features remain accessible after axial endplates are installed 1-2-3 survey block

12. Warm Magnetic Measurements Feb 3 2016 MQXFA Magnet Fiducialization and Alignment12  May use the existing warm measurement probe with a different arrangement of two pickup coils  we can measure the field angle from the first coil relative to that from the second coil as we scan along the magnet (We will check the possible resolution of this method)  Tilt angle sensors  The typical accelerometer with 12-bitADC has a resolution of 1 mrad (0.057 degrees). With inclinometers the resolution can reach 0.1 mrad  Survey targets can be attached to the measurement probe to track position and location of probe during measurements  Stretched wire measurements may also be performed  This information will be combined with the mechanical fiducialization data

13. Proposed Fiducial Plan Feb 3 2016 MQXFA Magnet Fiducialization and Alignment13 Yoke Fiducials, 3x- 4x per axial location (bottom may not be readily accessible) Coil Fiducials, Impregnated (4x), Both ends Measured (mech/mag) Centers

14. Magnet Assembly Tooling Layout Feb 3 2016 MQXFA Magnet Fiducialization and Alignment14 Magnet integration table will be surveyed and fixed into place Network of semi-permenent survey monuments for aligning the tables and defining coordinate system when surveying magnet structure

15. logo area ABS and Documentation  Assembly Breakdown Structure document  “Independent” tasks/subassemblies having  Dedicated procedures  Dedicated QA elements  CAD drawings, BOM  Travelers / documentation => MTF  Engineering notes D.W. Cheng – MQXF International Review, June 7th-10th 201615

16. logo area Manufacturing Plan D.W. Cheng – MQXF International Review, June 7th-10th 201616

17. Smallest Tooling Footprint--not desired  Not recommending, just documenting  Have only single line—need to swap tooling between Yoke Assembly and Coil Pack  Entire operation serial, no parallelization  Minimum Tooling Cost Move toward Wall Layout for MQXF-L (early 2016)

18. Start to Finish no parallel activity  This is what ‘minimum tooling’ rate probably looks like…  Does not include inefficiency of tooling re-configuration  If Yoke Shell sub-assy can be staged, perhaps shrink duration by 12.5d i.e. below 50d target (risky)  Conceivable to meet production rate, particularly if some learning curve is applied to both this and coil delivery  Would prefer independent Coil Pack assembly line to save additional ~10d in parallel (next slide) 18

19. Start to Finish moderate parallelism  Yoke Assembly finishes with Float, done in parallel  40days start to finish with sufficient manpower to work on coils in parallel  ‘Final Shim Adjustment’ may not be required, but not enough statistics to say it can be removed presently  Should not use this ‘now’ to save tooling cost 19

20. logo area Assembly Tables and Lifting Fixtures D.W. Cheng – MQXF International Review, June 7th-10th 201620

21. logo area Status  Procurement of long tooling finished, parts coming in  Procurement of first long structure complete, parts arriving  Draft procedures being written  Based on previous experience and final MQXF design  Interfaces with MTF in progress  Alignment strategy drafted; will be tested on prototypes  Layout of B77a “now” and at “full production”  Second Structure precurements will start later this summer D.W. Cheng – MQXF International Review, June 7th-10th 201621

22. logo area Status (May)  We experienced significant delays on key structural parts  Initiated yokes order September 2015  Yoke vendor encountered issues with machining tight tolerances, CMM inspections, 1st article turnarounds, several iterations  Loosened a few tolerances (comparable with CERN’s fabrication drawings)  Initiated a backup order for yokes with a 2nd vendor February 2016  Last month the re-tooled efforts of 1st vendor still did not hit tolerances  First yokes and collars are expected to be approved for production soon  Latest set of collars are being inspected this week  F/A yoke is due from 2nd vendor later this week  They invested in in-house inspection capability, so turnarounds could also be expedited  First batch of load pads have been approved and are currently in production  Shells received  Currently being instrumented  Bushings (collars, load pads, yokes) received  Nitronic 50 endplates received  First dummy coils are being delivered this week D.W. Cheng – MQXF International Review, June 7th-10th 201622

23. logo area Tooling Status (May)  We also experienced some delays in tooling design  At the July 2015 review expressed concerns about the amount of work the tooling represented  Designer resources finally stabilized during the first part of 2016, but much work still remained  CERN-style tooling required adaptation for use at LBNL  Short tooling designs had to be scaled up for longer lengths  Also required analysis for seismic events  Tooling is starting to arrive  Coil lifting fixtures  Integration table frame  Coilpack assembly table + side table frames  Procurements have been initiated with almost all tooling D.W. Cheng – MQXF International Review, June 7th-10th 201623

24. logo area Summary  Structure has been well developed through many years of LARP model magnet development  Procedures are based on the experiences and lessons learned from both short and long models  Manufacturing plan has been developed; based on the prototypes currently being fabricated  First long magnet test expected in 2017 D.W. Cheng – MQXF International Review, June 7th-10th 201624

25. logo area Acknowledgements  LBNL  H. Felice, E.C. Anderssen, D.R. Dietderich, R. Hafalia, M. Marchevsky, S. Myers, H. Pan, G.L. Sabbi, X. Wang, J. Wirdzek  CERN  P. Bestmann, N. Bourcey, A. Carlon, H. Dupond, N. Eyraud, J. Ferradas, B. Favrat, S. Izquierdo Bermudez, L. Lambert, P. Ferracin, P. Grosclaude, M. Guinchard, M. Juchno, F. Lackner, N. Peray, H. Prin, E. Rochepault, T. Sahner, E. Todesco, G. Vallone, R. Van Weelderen  BNL  M. Anerella, A. Ghosh, J. Schmalzle, P. Wanderer  FNAL  G. Ambrosio, R. Bossert, G. Chlachidze, L. Cooley, E. Holik, S. Krave, F. Nobrega, M. Yu D.W. Cheng – MQXF International Review, June 7th-10th 201625

26. logo area Backup Slides D.W. Cheng – MQXF International Review, June 7th-10th 201626

27. Feb 3, 2016 MQXFA Assembly & Tooling27 Coil Pack Assembly Prep--Coil Preparations  Coil lifting tooling has been scaled up from the MQXFS coil lifting  Rollover tooling is an extension of the table being used for the MQXFS  Lifting tooling will be scaled up from MQXFS fixtures  Single coil, OD-pick (BNL design)  Coil pair, OD-pick  Coil pair, ID-pick

28.  Coils will be received, CMM’d, instrumented  CMM measurements will also capture fiducial information (see later talk on MQXFA Alignment proposal)  Rollover tooling is used  To apply the coil GPI layers  To pair coils  To orient coils for handling Feb 3, 2016 MQXFA Assembly & Tooling28 Coil Pack Assembly Prep--Coil Preparations

29.  Install coil assembly support spuds  Spud engages with the holes in the RE end shoe  Tooling also clears the Kapton GPI layers, instrumentation wires, and leads (LE only) Feb 3, 2016 MQXFA Assembly & Tooling29 Coil Pack Assembly Prep--Coil Preparations

30.  The assembled loadpad-collar subassemblies are moved to auxiliary tables for positioning and assembly of the coilpack  Only load pads will be bolted in the assembly  Assembled coil pack is ready for insertion Feb 3, 2016 MQXFA Assembly & Tooling30 Coil Pack Assembly Prep—Load Pads & Collars

31. MAGNET INTEGRATION Feb 3, 2016 MQXFA Assembly & Tooling31

32. Magnet Integration Feb 3, 2016 MQXFA Assembly & Tooling 32 Insertion rails are attached to the coil pack (exploded view shown; not actual process) Rails will be split and pinned in the middle so that they can be pulled out from each end easily.

33. Magnet Integration Feb 3, 2016 MQXFA Assembly & Tooling 33 Coil pack assembled on assembly master

34. Magnet Integration Feb 3, 2016 MQXFA Assembly & Tooling 34 Coil pack assembled on assembly master Activation of the cylinders

35. Magnet Integration Feb 3, 2016 MQXFA Assembly & Tooling 35 Coil pack assembled on assembly master Activation of the cylinders Installation of the wheel racks underneath the coil pack

36. Magnet Integration Feb 3, 2016 MQXFA Assembly & Tooling 36 Coil pack assembled on assembly master Activation of the cylinders Installation of the wheel racks underneath the coil pack Actuator pressure released ~1 mm gap between assembly master and coil pack

37. Magnet Integration Feb 3, 2016 MQXFA Assembly & Tooling 37 Coil pack assembled on assembly master Activation of the cylinders Installation of the wheel racks underneath the coil pack Actuator pressure released Coil pack ready to be inserted

38.  Assembled coil pack is ready for insertion  Coil pack assembly table is shifted over to magnet integration table  Tables are aligned into place  Coil pack is inserted into yoke-shell structure Feb 3, 2016 MQXFA Assembly & Tooling38 Magnet Integration

39.  Assembled coil pack is ready for insertion  Coil pack assembly table is shifted over to magnet integration table  Tables are aligned into place  Coil pack is inserted into yoke-shell structure Feb 3, 2016 MQXFA Assembly & Tooling39 Magnet Integration

40.  Assembled coil pack is ready for insertion  Coil pack assembly table is shifted over to magnet integration table  Tables are aligned into place  Coil pack is inserted into yoke-shell structure Feb 3, 2016 MQXFA Assembly & Tooling40 Magnet Integration

41.  Assembled coil pack is ready for insertion  Coil pack assembly table is shifted over to magnet integration table  Tables are aligned into place  Coil pack is inserted into yoke-shell structure Feb 3, 2016 MQXFA Assembly & Tooling41 Magnet Integration

42.  Assembled coil pack is ready for insertion  Coil pack assembly table is shifted over to magnet integration table  Tables are aligned into place  Coil pack is inserted into yoke-shell structure Feb 3, 2016 MQXFA Assembly & Tooling42 Magnet Integration

43.  Assembled coil pack is ready for insertion  Coil pack assembly table is shifted over to magnet integration table  Tables are aligned into place  Coil pack is inserted into yoke-shell structure Feb 3, 2016 MQXFA Assembly & Tooling43 Magnet Integration Half-length (~2.3 m long) master keys and bladders are inserted.

44. Proposed Coil Ends Feb 3 2016 MQXFA Magnet Fiducialization and Alignment44 Either a coil fiducial plate, impregnated (Both ends), Or, 2-3 pin holes to be drilled into the end shoes and reamed after reaction This method will be a direct measurement to the coils, removing the uncertainty of measuring the collar ends

45. Fiducializing the Coil Mechanical Center Feb 3 2016 MQXFA Magnet Fiducialization and Alignment45 When coils are measured, CMM data will also measure the fiducials on both ends to locate them with respect to the theoretical center Perhaps pin holes in poles can also be used in the ID if necessary; requiring spot faces

46. Yoke Survey Tooling Block Feb 3 2016 MQXFA Magnet Fiducialization and Alignment46  “1-2-3” Style block  Magnetic hold down

47. Fixed, and aligned Magnet Integration Table Feb 3 2016 MQXFA Magnet Fiducialization and Alignment47 Magnet shell/yoke assemblies will be located onto pillars that have been surveyed in

48. What is Achievable? Feb 3 2016 MQXFA Magnet Fiducialization and Alignment48  This is more of a fiducialization process  Magnet alignment: No real “knobs to turn”  Assembly tooling will be aligned, though  Will likely use a laser tracker (FARO Vantage system)  Our survey crew will need to perform some tests http://www.faro.com/products/metrology/far o-laser-tracker/overview

49. Near-term B77a Feb 3 2016 MQXFA Magnet Fiducialization and Alignment49

50. Transition to Production E Anderssen

51. Overview  77A Layout Options  MQXF-L rates various space options…  Schedule Estimates

52. Transition to Long Magnets  Current location of table line—in line with traction system  Position OK for MQXF-S, but need to change for MQXF-L  Need to move traction system, and have one both ends  Space to wall overly generous… Current Shelf Loc. Loading Staging Traction winch Stage in Trays Current Table Line Approximate Layout Presently

53. Comments for Discussion  Modular tables are ‘generous’ (over wide/long)  3’ X 18’ is a guestimate  Coil Crate is 2’ X 18’  Fermi Tables 2’, but need to be anchored to floor  Do we need to ‘fasten’ cradles/rails to tables now?  Winch locations make Yoke-Shell assy line difficult to move  Discussion of this location most relevant—other lines can move relatively easily  Move closer to wall up front?  How much tooling can we afford on LARP  Return to this after schedule discussion…

54. Proposed Coil Production Rate  Schedule driven by Coil Part Procurement Milestones  Same for both Fermi and BNL, so double this rate  Production assembly start after 5-6 coils available for ‘matching’  Production rate ~50d/2 coils (not 34 as previously thought)  Need to finish 3 rd MQXF-L prior to production start  Lag (rate) drops to 25d mid 2019, then 19d mid 2020 50d Lag Production Start ~FY19

55. Observations  Production magnet assembly starts FY19—some additional tooling can be bought in FY19 to meet higher rate  Need to produce 3 MQXF-L prototypes beforehand  Propose to do all that with single set of tooling in 77A up until FY19 (‘dense’ configuration parallel assembly)  Consider moving Yoke-Shell sub-assy to 77  Need secondary bladder pressure unit and strain gage readout some months into FY19  Staging yoke-shell sub-assy in 77 end-bay likely better  LZ may be complete on FY19 timescale, but should not plan for that  Even if done, moving platform will take some time…  Plan to condense into 77A in 2020, perhaps add a second coil pack assembly line then…

56. Conclusion  ‘Dense’ configuration is consistent with initial production rates  Parallel short and long during ‘16-’17 possible with this configuration (maybe some impacts)  Use of surge space in 77 may be required  Estimate of tooling as function of year required  Inventory of tooling will be produced to facilitate estimate  Seems to be consistent with LARP Funding  Additional tooling procured late ‘19/early ‘20 to facilitate ramped production rates  Include cost estimate for this in inventory/estimate above