1 / 34

MAGNET TECHNOLOGIES FOR LHC UPGRADES episode II

CERN, 31 st July 2013 Summer student lectures. MAGNET TECHNOLOGIES FOR LHC UPGRADES episode II. E. Todesco Magnet, Superconductors and Cryostats Group Technology Department CERN, Geneva, Switzerland. CONTENTS. The challenge of superconducting cables

hilde
Download Presentation

MAGNET TECHNOLOGIES FOR LHC UPGRADES episode II

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, 31st July 2013 Summer student lectures MAGNET TECHNOLOGIES FOR LHC UPGRADESepisode II E. Todesco Magnet, Superconductors and Cryostats Group Technology Department CERN, Geneva, Switzerland

  2. CONTENTS • The challenge of superconductingcables • Towardslargerfields: Nb3Sn and HTS • Limitations to currentdensity: stress, protection • The Nb3Sn technology for the LHC luminosity upgrade • The High Energy LHC and HTS

  3. A bulk piece of superconductor does not work! It took a while to build superconducting magnets: cable need a complex geometry Firstly you need some stabilizer (Cu) Superconductor above the critical surface become insulator – current must go through a low resistivity material otherwise it burns everything Typically, at least half of the available volume Ok, but why such a complex structure? Looks like a fractal … or a honeycomb SUPERCONDUCTING CABLES Sketch of Rutherford superconducting cable and cross-section Superconducting cable made of strands Superconducting strand

  4. When field changes, the conductor gets magnetized Related to dB (hysteresis) Since it is superconductor, shielding current stay there forever Solution: fine filaments (~10 mm) Faraday law: change of field induces a current in loops Related to dB/dt, ramp rate effects Loops are large since magnets are long! Solution: twist the cable Complex cable geometry The cable is made of (20-50) strands of 1 mm diameter The strands are made of superconducting filaments of 5-50 mm diameter inside a copper matrix Both cable and strands are twisted SUPERCONDUCTING CABLES Sketch of Rutherford superconducting cable and cross-section Superconducting cable made of strands Strand deformation after cabling [D. Dietderich, L. Oberli]

  5. The manufacturing of a strand is an amazing technological process Cooking recipe in the hand of the firms SUPERCONDUCTING CABLES Superconducting cable made of strands

  6. CONTENTS • The challenge of superconductingcables • Towardslargerfields: Nb3Sn and HTS • Limitations to currentdensity: stress, protection • The Nb3Sn technology for the HL-LHC • The High Energy LHC and HTS

  7. TOWARDS LARGER FIELDS • Yesterdaywe have seen the Nb-Ti acceleratormagnets are limitedat 8 T • Critical surface – (cost and stability) • Nb3Sn has a widercritical surface: covers up to ~15 T SC Nb-Ti SC Nb3Sn Resistive Nb3Sn Nb-Ti

  8. TOWARDS LARGER FIELDS • Nb3Sn has a wider critical surface • But the material is more difficult to manufacture • It has never been used in accelerators, but tested successfully in short models and used in solenoids • With Nb3Sn one could go up to 15-18 T (-20%) • World record is 13.5 T (D20,HD2, Berkeley) Critical surface for Nb-Ti and Nb3Sn Field versus coil thickness for Nb-Ti and Nb3Sn at 1.9 K

  9. TOWARDS LARGER FIELDS • Summary of accelerator dipoles, Nb-Ti and Nb3Sn, built and planned Bore field in dipoles: operational for Nb-Ti and maximum achieved for Nb3Sn

  10. TOWARDS LARGER FIELDS • Nb3Sn islimitedat 15 T • HTS materials have the amazingfeature of having a critical surface withverylowslope dj/dB • Todaythey are at 300-400 A/mm2 in the range 15-40 T • Wejustneed 20% more and hugespaceswillbeopened! Nb-Ti Nb3Sn Resistive HTS Nb3Sn Nb-Ti HTS

  11. CONTENTS • The challenge of superconductingcables • Towardslargerfields: Nb3Sn and HTS • Limitations to currentdensity: stress, protection • The Nb3Sn technology for the HL-LHC • The High Energy LHC and HTS

  12. LIMITATIONS TO CURRENT DENSITY • What’s behind this plot ? • Electromagnet: • Why most of magnets have similar current density ? 400 A/mm2 Bore field in dipoles: operational for Nb-Ti and maximum achieved for Nb3Sn

  13. LIMITATIONS to CURRENT DENSITY • The Terminator-3 accelerator has 640 T with 200 mm (?) coil width→ 5000 A/mm2 overall j (~10 times larger than usual accelerator magnets) able to withstand 640 T • Surprise: even though we had a material with such a nice Bvsj performance we would not be able to exploit it • Going to very large current densities pose several issues (i) Stability (ii) Stress (iii) Protection T-3 LHC

  14. LIMITATIONS to CURRENT DENSITY: STRESS • Lorentz forces act on the coil • B j, so the force is  j2 • A prestress is given by the mechanical structure to compensate similar Lorentz forces and avoid movements • We want to avoid movements, but do we have to put a prestress to completely compensate Lorentz force? Hot topic for us • You can measure the pole unloading during magnet ramp ! Unloading Stress in the coil versus square of the current in TQ01, and unloading in all coils[LARP collaboration] Direction of forces in a dipolecoil (LHC)

  15. LIMITATIONS to CURRENT DENSITY: STRESS • What is the limit to stress in magnets? • 200 MPais a good optimistic guess • Insulation problems for Nb-Ti (polymide) • Degradation of superconductor with strain for Nb3Sn, at similar values • Force j2 • So the T3 magnet would have to withstand 100 larger stress (~10 GPa instead of ~100 MPa) • This really looks science fiction to me A. Godeke, “Performance boundaries in Nb3Sn superconductors”, PhD thesis, 2005 p. 43.

  16. LIMITATIONS to CURRENT DENSITY: PROTECTION • In case of an irreversible resistive transition (quench) we have a dangerous situation • Superconductors are very bad conductors above the transition • That’s one of the reasons for having copper in the cable • Joule effect (proportional to j2) heats the coil • If the current is not dumped rapidly enough the cable melts • Order of magnitude of time to dump the current: 0.1-0.5 s • It is a RL circuit, time constant t=L/R • Where R is the resistance of the magnet – we should make it quench everywhere to increase R and lower t

  17. LIMITATIONS to CURRENT DENSITY: PROTECTION • Best situation: all magnet quenches so energy is spread everywhere • Specific heat vary a lot in the range 2-300 K – very non linear problem • Enthalpy of a typical cable from 2 to 300 K is 0.6 J/mm3 • So energy density in the magnet must be smaller than this limit • If we want to make a magnet with given aperture and field, too small coil gives too large energy density → physical limit • Nb-Ti typically operate at 0.06 J/mm3 • This is 1/10, meaning that 3 times larger current density reaches the enthalpy limit • For Nb3Sntypically we are at twice this value (0.1 J/mm3) • With these numbers, one has to react in 50 – 100 ms • (first thing: switch power off)

  18. LIMITATIONS to CURRENT DENSITY • The Terminator-3 accelerator has 640 T over 50 mm aperture with 200 mm coil width • Exercise: rough estimate of energy density • B2/2m over a circle of radius (50+200)/2 mm  11.GJ/m of energy • Coil area: sector of thickness 200 mm over a radius of 25 mm  0.08 m2 • This gives 130 J/mm3l, 250 times larger then the enthalpy limit (perhaps the most sci-fi number of the movie)

  19. LIMITATIONS to CURRENT DENSITY: PROTECTION • It is a RL circuit, time constant t=L/R • The velocity of spreading the quench is ~10 m/s • 1 s for the quench to reach the end of a 10 m long magnet (too slow) • Solution: quench all the magnet asap through heating • Strips of stainless steel close to the coil – as soon as a resistive transition is detected, a capacitor is discharged, and the whole coil is brought above critical temperature • Protection sets a limit in current density Quenchheaters in LBL [H. Felice]

  20. CONTENTS • The challenge of superconductingcables • Towardslargerfields: Nb3Sn and HTS • Limitations to currentdensity: stress, protection • The Nb3Sn technology for the HL-LHC • The High Energy LHC and HTS

  21. HIGH LUMINOSITY LHC:tHE PATH TOWARDS 3000 FB-1 • CERN Project, EU funds for the design study, DR in 2014 • The target: 3000 fb-1of data over a decade • You need a lumi/year4-5 times largerthan LHC design • Peaklumi 1035 cm-2 s-1is not acceptable for the experiments (pile up) • A levellingisproposedat 51034 cm-2 s-1 • To have this the LHC must be able to reach a peaklumi21035cm-2 s-1 • You needpeaklumi 20 largerthan nominal: • Factor 5 from the beam • Factor 4 fromoptics (reducingb* and killing F) • How larger the triplet ? • Aperture of the triplet isb • b1/b* • Four times smallerb*impliestwice the aperture

  22. HL-LHC: MAGNET TECHNOLOGY FOR LOWER BETA* • Larger aperture can be obtained with lower gradients  optics needs longer magnets, but • Nb3Sn gives ~50% more gradient for the same aperture → more compact triplet → lower b* HL-LHC LHC HL-LHC Aperture versus gradient relation [L. Rossi, E. Todesco, Phys. Rev. STAB 9 (2006) 102401]

  23. HL-LHC: Nb3Sn TECHNOLOGY FOR LOWER BETA* • Special challenges of Nb3Sn • It has been discoveredbefore Nb-Ti, has widercritical surface, but still not used in accelerators • Difficultfeatures • It isformedthrough a heatingat more than 600 C, thatlastsseveraldays • Insulationmust beresistantto 600 C – glass fiberisused • After formation becomesverybrittle • Significantdegradationat 0.1% strain • Veryaccurate control of stress needed

  24. HL-LHC: Nb3Sn TECHNOLOGY FOR LOWER BETA* • Routinelyused in solenoids • To reach 15-20 T domain • What has been for accelerators • Short (1-m-long) models (many) • A few (two) long coils (3.4-m-long) • Fields from 10 to 13.5 T • Up to 16 T but withoutcore HQ (Nb3Sn LARP 120 mm magnet) [G. Sabbi, S. Caspi, et al.] LR (first long Nb3Sn racetrack, LARP) [G. Ambrosio, et al.]

  25. HL-LHC: Nb3Sn TECHNOLOGY FOR LOWER BETA* • Training mechanism • In many cases Nb3Sn exhibits a large training • 80% seems a bit tight, but 70% is there in most cases 90% of critical surface 80%, operationalcurrent LARP magnet HQ training at 1.9 K [Magnetbuilt by LARP collaboration, test by H. Bajas, J. Feuvrier, M. Bajko]

  26. CONTENTS • The challenge of superconductingcables • Towardslargerfields: Nb3Sn and HTS • Limitations to currentdensity: stress, protection • The Nb3Sn technology for the HL-LHC • The High Energy LHC and HTS

  27. THE HIGH ENERGY LHC • The idea • Installing a 16.5+16.5 TeV proton accelerator in the LEP tunnel • Main ingredient: 20 T operational field dipoles • CERN study: www.cern.ch/he-lhc « First thoughts on a Higher Energy LHC » CERN ATS-2010-177 «The High Energy LHC » CERN 2011-003 (Malta conference proceedings) • Motivations [J. Wells, CERN 2011-3] • The energy frontier is always extremely interesting and for many processes cannot be traded with more luminosity at lower energy “The results of the LHC will change everything, one way or another. There will be a new “theory of the day” at each major discovery, and the arguments will sharpen in some ways and become more divergent in other ways. Yet, the need to explore the high energy frontier will remain.”

  28. THE HIGH ENERGY LHC • Current density – if we keep same as LHC, 80 mm of coil width needed (still ok) • Working at 80%, we need 500 A/mm2 at 25 T Sketch of the double aperture magnet with the ironyoke – Coils are in blue Operationalfield versus coilwidth in acceleratormagnets

  29. THE HIGH ENERGY LHC • What materials are available today ? • YBCO and Bi-2212 • Both have current density 20% lower than target • Note than in 2010 were 50% lower • Bi-2212 • Only HTS existing in round strand • You can make Rutherford cable • BUT • Complex heat treatment at temperatures higher than Nb3Sn • Stabilizer is silver but is soft so no resistance to stress • High cost Critical surfaces of practicalsuperconductors[P. Lee]

  30. THE HIGH ENERGY LHC • What materials are available today ? • YBCO and Bi-2212 • Both have current density 20% lower than target • Note than in 2010 were 50% lower • YBCO • You can buy and wind • No need of heat treatment • BUT • No cable, only flat tape • How do you make a accelerator magnet with tape ? • Very high cost

  31. THE HIGH ENERGY LHC • What material can tolerate 380 A/mm2 and at what field ? • For Nb-Ti: LHC performances - up to 8 T • For Nb3Sn: 1500 A/mm2 at 15 T, 4.2 K – up to 12 T • With lower current density 190 A/mm2/m we can get to 15 T • Last 5 T made by HTS - we ask for having ~380 A/mm2 Nb3Sn coils HTS coils Nb-Ti coil Same plot, in linearscalewithloadlines and selectedmaterials Where to place our design in the j vs B plot

  32. THE HIGH ENERGY LHC • HTS opens the way to higher and higher fields • Used to build solenoids, proved to work in the 20-30 T range Nb3Sn coils HTS coils Nb-Ti coil

  33. A TENTATIVE SCHEDULE FOR NEXT 30 YEARS Using the LEP/LHC tunnel [L. Rossi, IPAC 2011]

  34. CONCLUSIONS • HL-LHC canprovide 3000 fb-1at the horizon of the 30’s • Enabling technologies: large aperture magnets and crabcavities • This couldbe the first application of Nb3Sn to accelerators, pushing the operationalfieldfrom 8 to 12 T • Materialisdifficult, but a lot of progress i n the past 20 years • CERN infrastructure and LEP tunnel isa preciousasset of ourlab • 20 T magnetswouldenable a 33 TeV hadron collider • No bottlenecks have been identifiedfrom the point of view of beamdynamics • Magnets are the main challenge • In particular the last 5 T thatshould come fromhightemperaturesuperconductors

More Related