Structures and RF Components On behalf of the XBand Production team Joel Sauza Bedolla j oelsauzacernch ENMMEMA Index Introduction Accelerating Structures Baseline discs TD26R1CC ID: 1045021
Download Presentation The PPT/PDF document "Fabrication of CLIC Accelerating" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. Fabrication of CLIC Accelerating Structures and RF ComponentsOn behalf of the X-Band Production team Joel Sauza Bedollajoel.sauza@cern.chEN/MME/MA
2. IndexIntroductionAccelerating StructuresBaseline (discs) TD26R1CCSquare discs (alternative 1)Halves (alternative 2) ComponentsCatalogueOpen-source HardwareConclusions2
3. 1. IntroductionThe CLIC structures are designed to operate at X-Band, 2π/3 traveling wave mode with a loaded gradient of 100 MV/m The baseline design for accelerating structures is a damping waveguide disc The material is Oxygen Free Electronic (OFE) Copper Mechanical tolerances, at sub-micrometer level, are required to satisfy the RF design and beam dynamics constraintsUltra-precision diamond turning and milling is requiredThe disc are assembled by diffusion bonding. The rest of the components (waveguides, cooling blocks,…) by brazing3
4. 1. Introduction4T18 (TD18)T24TD24TD24 SiCTD26 CLEXTD26 R1 CC12SWV18026-01CSR1CC (TD26 R1 CC)12 GHz, damped, sealed, CLIAAS120245, 2ps assemblingØ83 mmMedical structures 3 GHzT24_PSI (brazing)12 GHz, undamped, sealed, CLIAAS120226, 1 pc tested, 1 pc is under test in Xbox 2Ø90 mmHG TW Proton LINAC3 GHz, sealed, CLIACBTW0021, 2 pcs assembledØ120 mmPROBE (Proton Booting extension for imaging)3 GHz, sealed, MELACCL30013, parts ready for assembly92 x 140 mmTD31 CC (for 380 GeV linac)CLIC HDS TD26 (EBW)12 GHz, damped, sealed, EBW version, under designTD26 R1 G* (CLIC G* bend WG)12 GHz, damped, sealed, assembled75 x 155 mm12 GHz, damped, sealed, CLIACTD310021, 4pcs under machiningØ83 mmTD31 R1 CCTD26 R05 CC12 GHz, damped, sealed, CLIAAS120084, 2ps assembled and tested
5. 1. Introduction76 structures have been built22 different designsDiscsQuartersHalvesDesign solutions Different diameterInterlocksDamped / undampedTank / sealedAssembly methodBonding / brazing / EB weldingH2 atmosphereTested at CERN, SLAC, KEK, PSI and Tsinghua5
6. 2. Accelerating structures6
7. 2.1 TD26R1CC“Nose” of the waveguide from an elliptical geometry to a 4-th order polynomial function. Optimized for surface magnetic fieldThe radius at the bottom of the RF waveguide was increased from 0.5 mm to 1 mm to allow the use of bigger milling cutter. Optimized for machining costDisc diameter was increased from 74 mm to 83 mm. Optimized for waveguides interface7
8. 2.1 TD26R1CC4 structures (118 discs). Biggest amount of discs ever produced in a single orderControl charts: to find production driftsRoughness and form: largely on toleranceProcess capability: to compare the output of the process w.r.t. the specificationsQuality assessment (visual inspection): evaluate the contribution of defects on the parts. A mix of structure 3 and 4 based on defects (scratches, machining marks) since dimensions were homogenous8
9. 2.1 TD26R1CCThe bonding of N1, N2 and N3 have been completed. Same supplier similar cycle N4 will be assembled next month Straightness after bondingN1: 22 µm N2: 70 µm N3: 35 µm In the 3 cases there was a diameter reduction (average) of 13 µmWhat is the effect of the misalignment?What happens to the internal RF volume?9Probably due to transport between assembly table and oven N1N2N3
10. 2.1 TD26R1CCWhat happens to the internal RF volume?(Un)fortunately, one disc of the N1 was mounted backwardsN1 will be cut and the internal volume will be thoroughly measured Bead-pull measurements before and after bondingCorrelation of b parameter with frequency deviationsWhat is the effect of the misalignment?Results after high power test10N1
11. 2.1 Disc handlingIt was observed oxidation on the film boxesChips coming from the tuning holes damaged one discReduce the handling of the discs to avoid scratchesOne frame per disc for all the process This transport method is now being employed for all structures11
12. 2.1 Tooling for Bead-pull measurementsNew tooling with 3 positions:30: To stack the discs0: To connect the waveguides90: RF Measurements12
13. 2.2 Rectangular DiscObjective: to reduce the number of components The possibility to have leaks The number of heating cycles Integrate SiC absorbers, pumping features and cooling circuits into the cell body. Prototype to evaluate the manufacturability and to test the bonding: 4 cells + 2 covers.13
14. 2.2 Rectangular Disc The manufacturing is feasibleThe bonding test failed. The structure leaks in several places The SiC absorbers were not dimensionally controlled and they may have interfered with the disc bondingPost-mortem analysis is under wayIn previous structures, it has been observed copper pollution on the SiC due to bonding14
15. 2.3 HalvesHard copper vs Soft copper: Lower breakdown rateBonding vs Brazing vs Electron Beam Welding: Hard copperPrecision Pins vs alignment features: iris shift of 70 µmManufacturing process: Milling-turning vs milling15T24 Open – SLAC designCERN designIaroslava Profatilova
16. 2.3 HalvesMechanical design completedTD26R1Assembly sequenceWaveguides brazingWaveguides + CoversHalves positioningLips for EB Welding Review of the alignment features 16
17. 2.3 HalvesCircular pockets placed on the vacuum input/output and on the sides of the structures. Precision discs that will deform due to Herzian contact Mock-up builtTooling to be built and tested17
18. 3. RF components18
19. 3.1 CatalogueRF components are used in the transmission and transformation of RF signalsThese devices operate between 10 and 120 MW peak power and up to 250 ns pulse length and ultra high vacuumThey have been tested in high powerA number of prototypes are installed at CERN in the testing facilities and work routinely19
20. 3.1 Catalogue20
21. 3.1 CatalogueSpiral Load: the actual design worksLast year, two spiral loads built in TitaniumSame machine, same process, same material: different resultsAnother test with Stainless SteelComplete re-design optimized for Rapid Prototyping No overhangingStackable design for series production21
22. 3.2 Open-source Hardware The design specifications of a physical object which are licensed in such a way that said object can be studied, modified, created, and distributed by anyone.The drawings are in a public repositoryStill, a registration with a lightweight account is necessaryWe would like to know who is working with this technology Evaluate its social impact22
23. 4. Conclusions23Main aspects for optimization Baseline (Disc)Rectangular DiscHalfNumber of external parts (manifolds, cooling, vacuum etc.)321Number of heating cycles321Assembly process231Machining123Stability of the process123
24. 4. Conclusions24StructureLast StepNext StepCommentsTD26R1CC N1Bonding failed Post-MortemMeasurement of the internal RF volumeTD26R1CC N2Brazing completedHigh power test TD26R1CC N3Brazing completedHigh power test TD26R1CC N4ManufacturingBonding Rectangular discBonding test failedPost-MortemRepeat bonding without SiC HalvesAlignment features under studyFabrication
25. 4. Conclusions25The discs for 4 TD31 structures will be finished by November Plans for the future:A full rectangular disc structure is being designed for manufacturing and power testing The Halves prototype looks promising. It has still to be validatedDrawings for a Klystron based structure are ready based on TD26R1CCBased on the results of the design alternatives, it is planned to build several superstructures for full CLIC modules. RF ComponentsAvailable technology Spiral load to be optimized for Rapid Prototyping
26. Ready for questions26Thank you for your attentionNuria Catalan Lasheras Anastasiya SolodkoKamil Tomasz SzypulaSerge LebetEnrique Rodriguez CastroSergio Gonzalez AntonJoel Sauza Bedolla
27.