1 / 26

DIPOLES FOR HIGH ENERGY LHC

CERN, 21 th February 2013 Snowmass at CERN. DIPOLES FOR HIGH ENERGY LHC. E. Todesco CERN, Geneva Switzerland Acknowledgements: B. Bordini , L. Bottura , G. De Rijk , L. Evans, P. Fessia , J. Fleiter , J. Nugteren , L. Rossi, F. Zimmermann. CONTENTS. Recall of Malta design

edena
Download Presentation

DIPOLES FOR HIGH ENERGY LHC

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CERN, 21th February 2013 Snowmass at CERN DIPOLES FOR HIGH ENERGY LHC E. Todesco CERN, Geneva Switzerland Acknowledgements: B. Bordini, L. Bottura, G. De Rijk, L. Evans, P. Fessia, J. Fleiter, J. Nugteren, L. Rossi, F. Zimmermann

  2. CONTENTS • Recall of Malta design • Where are wewithcurrentdensityrequirements? • Possible simplifications and associatedcosts • Optimization of celllength

  3. RECALL OF DESIGN • Field ~ currentdensityj  coilwidthw • Currentdensityis the main choiceof the magnet designer • Pioneeringwork of McIntyre • With380 A/mm2, one makes ~ 2.5 T each 10 mm of coil, sowe need 80 mm • Most acceleratormagnetsnot far fromthisj value • Lowcurrentdensitybringstwoadvantages • More margin for protection • Lowerstress • Other main choice: have a 20% marginon the loadline • So we must have a coilreaching 25 T at short sample! McIntyre Fresca2 Field versus coilwidth

  4. RECALL OF DESIGN • Malta layouthighlyoptimized to minimizeconductorcost • Hypothesis: cost Nb-Ti is one, Nb3Sn is 4 times, HTS is 16 times • Use Nb-Ti up to 8 T, Nb3Sn up to 13 T, HTS up to 20 T • Wealso use Nb3Sn withhalfcurrentdensityto have 2 more Tesla and reach 15 T, saving on HTS (seenext section) • Lowercost, at the price of complexity One quarter of a coil 20 T hybridmagnet: the Malta design [E. Todesco, L. Rossi]

  5. CONTENTS • Recall of Malta design • Where are wewith the currentdensityrequirements? • Possible simplifications and associatedcosts • Optimization of celllength

  6. WHERE ARE WE WITH NB3Sn? • Project is in midterm future sosomeoptimismisallowedto account for progress in technology • Nb3Sn performance has greatlyimproved (doubled in tenyears), sono spaceisassumed for furtheroptimization An historicalview on the improvement of Nb-Ti and Nb3Sn performance [L. Bottura, ASC 2012]

  7. WHERE ARE WE WITH NB3Sn? • Hypothesis • Copper to superconductor of 1.1 (as in recent LARP and 11 T magnets) • Insulation, voids (impregnation) bringdilution factor to 0.33 • So weaimat • 13/0.8=16.25 T wewant 380*3/0.8=1400 A/mm2 • 15/0.8=18.75 T wewant190*3/0.8=700 A/mm2 • Today best conductor (2500 A/mm2at 12 T, 4.2 K) providesthese values, with10% cabledegradation[B. Bordini, based on PIT and RRP data] • But thisis extrapolation of data atlowerfields, someasurements in the 15-18 T range wouldbewarmlywelcome

  8. WHERE ARE WE WITH NB3Sn? • Protection • For protection, we are at a level of energydensity in the coil of about 0.2 J/mm3 – thisis ~50% more whatwe have in Nb3Sn magnets • More coppercouldbeneeded, sofurtherincrease of currentdensity in the superconductorcouldbeuseful Energydensity in the coil versus currentdensity in the coil, with protection time margin [E. Todesco, ASC 2012]

  9. WHERE ARE WE WITH HTS? • Two options: YBCO and Bi-2212 • YBCO • Tape  • Very good currentdensity in paralleldirection  but stronganisotropy  • Stress resistant  • Bi-2212 • Cable • No anisotropy  • No stress resistant(reinforemcent in strandneeded)  • Severalactivitiesongoing in differentlabs[G. De Rijk, S. Prestemon, this workshop], wideexperiencewithsolenoids

  10. WHERE ARE WE WITH HTS? • Both YBCO and Bi-2122 are ~400 A/mm2, vs 480 A/mm2required • YBCO: Preliminaryanalysis of field direction in Malta coil: in HTS coil angle betweenfield and conductoris up to 30, so I thinkwe have to forget about YBCO parallelperformance Large improvement w.r.t. Malta workshop 2.5 yearsago ! (wehad ~200 A/mm2) HTS target Engineering currentdensity of YBCO and of Bi-2212 [P. Lee]

  11. CONTENTS • Recall of Malta design • Where are wewith HTS requirements? • Possible simplifications and associatedcosts • Optimization of celllength

  12. MAKING IT SIMPLER? • Malta design (slightly simplified) • Guideline: follow HD2 and Fresca2 mechanical structure, i.e. no supportingelements in the coil – preliminaryanalysis shows stress<200 MPa • One double pancake of HTS Revised Malta design for 20 T magnet, one quarter of coilshown

  13. MAKING IT SIMPLER? • Malta design (slightly simplified) • One double pancake of HTS • One double + one single pancake of low j Nb3Sn Revised Malta design for 20 T magnet, one quarter of coilshown

  14. MAKING IT SIMPLER? • Malta design (slightly simplified) • One double pancake of HTS • One double + one single pancake of low j Nb3Sn • One double + one single pancake of Nb3Sn Revised Malta design for 20 T magnet, one quarter of coilshown

  15. MAKING IT SIMPLER? • Malta design (slightly simplified) • One double pancake of HTS • One double + one single pancake of low j Nb3Sn • One double + one single pancake of Nb3Sn • One double pancake of Nb-Ti • Six coils to beassembled per pole, four withflared ends, two flat Revised Malta design for 20 T magnet, one quarter of coilshown

  16. MAKING IT SIMPLER? • Malta design, without Nb-Ti • One double pancake of HTS • One double + one single pancake of low j Nb3Sn • One double + one single pancake of Nb3Sn • Five coils to beassembled per pole • Cost: +15% Revised Malta design for 20 T magnet (no Nb-Ti), one quarter of coilshown

  17. MAKING IT SIMPLER? • Malta design without graded Nb3Sn • One double pancake of HTS • One double + one single pancake of Nb3Sn • Threecoils to beassembled per pole • Cost: +50% ( 1.5 times the Malta design) Revised Malta design for 20 T magnet (no Nb3Sn grading, no Nb-Ti), one quarter of coilshown

  18. THE 15 T CASE • The 15 T case, without Nb-Ti • I passed to 1 mm strand to avoidthreelayers – HD2-like layout • One double pancake of Nb3Sn usedatlowje = 190 A/mm2 • One double pancake of Nb3Sn, twocoils to beassembled per pole • Cost: 45% of Malta • 20% more (i.e. 55% of Malta) if no Nb-Ti grading Snowmass design for 15 T magnet, one quarter of coilshown

  19. BLOCK VERSUS COS THETA • Block design advantages • Fits the shape of the fieldallowinggrading for very large coils • Natural position of quenchheaters (midplaneand between double pancakes • Pursuingstudies on this design is important(Fresca2, HD2, HD3) Bologna city centre: block (left) vs cos theta (right)

  20. CONTENTS • Recall of Malta design • Where are wewith HTS requirements? • Possible simplifications and associatedcosts • Optimization of celllength

  21. OPTIMIZAZION OF CELL LENGTH • LHC has a semi-celllength (distance betweenquadrupoles) of L=50 m • Main scaling • Beta functionL • Beam sizeLso longer spacingrequireslarger aperture  • Longer spacingrequireslessquadrupoles  • Integrated gradient Gl1/Lso longer spacingrequireslowerintegrated gradient  • Example • HE-LHC needs 1600 T, 40 mm aperture sowith Nb3Sn weneed 3.5 m quads

  22. OPTIMIZAZION OF CELL LENGTH • If weget longer celllength, wewill have less force needed and lessquadrupoles but larger aperture • Hypothesis: 20 m max length of dipoles • Either 20 T fixed and largerenergy, or energyfixed and lowerfield FIRST CASE: FIXED FIELD OF 20T • Keeping a 20 T magnet, and makingitlargerwecan gain 0.9-1.3 TeVout of 16.5 TeV

  23. OPTIMIZAZION OF CELL LENGTH • If weget longer celllength, wewill have less force needed and lessquadrupoles but larger aperture • Either 20 T fixed and largerenergy, or energyfixed and lowerfield SECOND CASE: FIXED ENERGY OF 16.5 TeV • Keeping a 16.5 TeVenergy, wecanlower the field of 1-1.5 T

  24. OPTIMIZAZION OF CELL LENGTH • Whats the price? Rough estimatebased on sectorcoil • For a fixedenergy, a 66 m long cellallows go to 45 mm aperture with 19 T, saving 20% of HTS (10% of conductorcost in the cross-section, 5% on global costaccouting for higherfilling) • Looks marginal – and longer celldoes not help • For a fixedfield, 1 additionalTeVcosts ~10% • Thenitrapidly diverges

  25. FURTHER DEVELOPMENTS • Superconductors • Push (and measure) performance of Nb3Sn in the range 15-18 T • We are at Malta design values, but improvementcouldconsiderablyreducecosts • Get a 20% more on je in HTS, to reach 500 A/mm2at 25 T • Todaywe are not so far ! • Magnettechnology • Fresca2 and HD block designs are an essential steptowards the 20 T dipoles • Block design has the advantage of allowing a naturalway of grading the material, and saving money • Removable pole technology for Nb3Sn needed • Splicetechnologyto bewidelystudied • Flared ends are still a problemtoday for block design

  26. CONCLUSION • Simplifications • Wepresented successive simplification of Malta design, each one withprice tag • Seenfrom end to beginning, itcanindicate a roadmaptowards 20 T • The Nb-Ti allowssaving about 15% • Withoutgrading of Nb3Sn, price doubles • Stoppingat15 T, withgraded Nb3Sn priceis 55% • Optimizingcelllength • One could explore an option with longer celllength to have lessquadrupoles are largerfilling factor • Savingis not significantin this phase of the project

More Related