Comments
Description
Transcript
Slides - Agenda
Stato del progetto MAGIX Giovanni Volpini a nome della collaborazione Frascati, riunione CSN V, 27 luglio 2015 MAGIX & INFN participation to HL-LHC CERN-INFN Collaboration Agreement MAGIX Design, construction and test of WP1 CORRAL the five prototypes of the corrector magnets for the HL CERN endorses MAGIX WP1 & WP2 deliverables and milestones, contributing with 527 k€, through the collaboration agreement KE2291/TE/HL-LHC interaction regions of HiLUMI WP2 PADS 2D & 3D engineering design of the D2 magnets Development of HTS coil for WP3 SCOW-2G application to detectors and 1 accelerators Low-loss SC development for WP4 SAFFO application to AC magnets MAGIX is a INFN-funded research project, (GrV, «Call») whose goal is to develop superconducting technologies for application to future accelerator magnets. It includes four WP’s, two of which are relevant to HL-LHC 2014-2017, 1 M€ + personnel funds 2 INFN already involved in FP7HiLumi (UE-HILUMI, GrV) WP2 beam dynamics, LNF WP3 magnets, MI-LASA WP6 cold powering, MI-LASA Economic Plan 93 m.u. (WP1) + 18 m.u. (WP2) (INFN personnel only, no University employees) Costs CERN contribution Personnel 521,900 € 50% 260,950 € Materials for magnet construction 113,000 € 75% 84,750 € Costruction & test 363,000 € 50% 181,500 € Travels, lab operation, tooling, R&D activities 305,000 € 0% TOTAL 0€ 527,200 € 781,000 € WP1 (679k€) + WP2 (102 k€) Payment linked to deliverables Giovanni Volpini CERN, 30 June 2015 Schedule HiLumi-MAGIX schedule 2014 v. February 2014 2 3 4 5 6 7 2015 8 9 10 11 12 1 2 3 4 5 6 7 2016 8 Project Management WP0 M 0.1 Feb 2014 Kick-off meeting with specification transfer M 0.2 Dec 2014 1st year activity monitoring M 0.3 Dec 2015 2nd year activity monitoring M 0.4 Dec 2016 3rd year activity monitoring M 0.5 Jun 2017 4th year activity monitoring M 1.1 M 1.2 D 1.3 D 1.4 Jul 2014 Dec 2014 Mar 2014 Mar 2015 Oct 2016 Dec 2015 Mar 2016 Jul 2016 Oct 2016 Apr 2015 July 2016 Feb 2017 Mar 2017 June 2017 M 2.1 D 2.1 June 2015 M 2.2 D 2.2 WP1 CORRAL D 1.1a D 1.1b D 1.2 M 1.3 M 1.4a M 1.4b M 1.5 M 1.6a M 1.6b M 1.6c WP2 * * ** *** *** **** **** **** Sextupol engineering design. Sextupol construction. Preliminary 2D design of the five magnet types Preliminary 3D design of the five magnet types Executive design of the five magnet types MgB2 quadrupole design. Octupole and decapole construction Quadrupole and dodecapole construction MgB2 quadrupole construction Test of the sextupole Test of the octupole and decapole Test of the dodecapole and quadrupole Corrector magnet test report Corrector magnets final check, packing and transport to CERN PADS 2D magnetic design to minimize the cross talk between the two dipoles. Dec 2015 2D mechanical design. M 2.3 Feb 2016 3D magnetic design including the coil ends. M 2.4 Apr 2016 Quench preliminary analysis. M 2.5 Jun 2016 3D mechanical design with the axial pre-load study. Dec 2016 Final Engineering design. M 2.6 Notes * ** *** **** 1 D 2.3 These two deliverables are grouped in one in the MAGIX project Note the change of scope wrt to the MAGIX project These two milestones are grouped in one in the MAGIX project These two milestones are grouped in one in the MAGIX project Explanation Activity Milestone Deliverable Giovanni Volpini, CERN 30 June 2015 9 10 11 12 1 2 3 4 5 6 7 2017 8 9 10 11 12 1 2 3 4 5 6 7 8 9 1 Milestones &Deliverables 50% June 2015 The INFN-CERN collaboration steering committee (F Bodry, L Rossi, E Nappi A Zoccoli) has endorsed the work done so far, authorized the transfer to INFN of 260 kE, in return of the deliverables completed so far. Giovanni Volpini, CERN 30 June 2015 WP1 WP1 CORRAL CORRAL CORrettori per le Regioni di interazione ad Alta luminosità di Lhc. Obiettivo: progettazione, costruzione e collaudo dei prototipi di cinque magneti correttori multipolari per le regioni di interazione ad alta luminosità di LHC, sulla base delle richieste del progetto HL-LHC dal CERN. Verrà verificata sia in fase di progetto che attraverso la costruzione di un prototipo, la fattibilità di soluzioni che utilizzino il MgB2 anziché il più convenzionale Nb-Ti. Giovanni Volpini, CSN V, Catania 23 luglio 2014 6 Corrector magnet inventory Iron yoke Iron yoke SC Coils SC Coils Mechanical support Mechanical support SC Coils 150 Mechanical support OD460 From 6-pole to 12-pole magnets exist in both normal and skew form (the latter is shown) 150 OD320 The superferric design was chosen for ease of construction, compact shape, modularity, following the good performance of earlier corrector prototype magnets developed by CIEMAT (Spain). Giovanni Volpini, CERN 30 June 2015 Sextupole layout Yoke Coil D320 Wedge CuBe TieRods 194 5.8 mm thick iron laminations, machined by EDM Yoke Bridge Flux-return plates Coil winding & impregnation tooling Insulation scheme: -wire w/ S2 glass 0.14 mm thick (on diameter) -ground insulation: G11, 2 mm thick plates on both sides of the coil, including the wire exits G11 thin, flexible layer on the inner wall of the coil; S2 tape on the outer wall Resin inlet/outlet Base plate mandrel Resin inlet/outlet Top plate Giovanni Volpini CERN, 26 February 2015 Closing cap (defines the impregnation chamber) Winding station Controlled wire tension 10 N (51 MPa) Telecentric camera system Single Coil Sample Holder Goals: 1) To test a coil in “realistic” conditions to identify major faults in the design/assembly; 2) To commission the “small” magnet test station, to be used to test sextupole, octupole and decapole y z x Test results First test at 4.2 K Current increased by steps at 0.3 A/s. Quench induced with heaters at 90, 160, 200 and 220 A. Ramp up to 260 A (no quench induced at this current value by choice). No spontaneous quench occurred. Test at subcooled LHe Significant heat load in the bath prevents from reaching a temperature lower than 2.5 K. Main reason is the thermal shield, whose temperature decreases very slowly. Current ramp up to quench. Four training quenches occurred at 295 A (2.56±0.04 K) or 80% of the s.s. at this T 318 A (2.60±0.04 K) or 87% “ 329 A (2.72±0.05 K) or 91% “ 325 A (2.85±0.06 K) or 91% “ The magnet operates at 40% on the load line Training at 4.2 K Current ramp up to quench at 0.3 A/s First quench at 280 A, then repeated increasing the ramp rate up to 5.7 A/s (limited by power supply in this configuration). In total 14 quenches at 280 A , or 95% of the s.s. limit. E.M. Forces A magnetic plate creates along the normal of the coil plane an e.m. force pattern more resembling to that experienced by a coil during its operation inside the magnet. Fx (normal to the coil plane, half coil) Fy (normal to long axis, half coil) Fz (normal to long axis, half coil) 2.9 kN @ Iop, 1.5 kN @ Iop, 0.6 kN @ Iop, here reached at about “ “ 300 A 250 A 180 A Green light to the magnet construction ! Sextupole assembly Iron laminations Cu traces for coil-to-coil junctions Duratron plate s.s. rings Giovanni Volpini CERN, 30 June 2015 Coil MgB2 development We are focusing the desing around an innovative solution. We call it Round Coil Superferric Magnet (RCSM) Simple, circular coil shape, cost effective. Expecially suited to strain-sensitive materials, like MgB2 We consider a sextupole configuration; different multipoles may be realized replacing the iron Preliminary design (milestone M1.3) almost completed. 14 Old & Older 15 A rule (symmetry) changer No matter how a sextupole magnet is done, it is invariant by a 120 degree rotation. A 60 degree exchanges the “north” and “south” poles; if we reverse the current direction as well, the field is globally unchanged! A RCSM is invariant by 120 degree rotation. A rotation by 60°, amounts to a “mirroring” w.r.t. a plane normal at z-axis, at z=0. No change in current. No overall mirror symmetry. This difference of the symmetries has profound consequences on the harmonics properties: a “traditional” layout has no even (“forbidden”) harmonics, and no net solenoidal field; a RCSM has also even harmonics, that vanish when integrated from -∞ to +∞, and a net solenoidal field. More complex configurations may suppress the latter, at the price of net even harmonics. G. Volpini, J. Rysti CERN 15 July 2015 Finanziamento 2016 Giovanni Volpini CSN V Frascati, 27 luglio 2015 17 Persone coinvolte + 1 tecnologo TD a partire da settembre 2015 Giovanni Volpini CSN V Frascati, 27 luglio 2015 18 WP1 Revised schedule to completion 2015 New dates M M M M M M M 1.3 1.4a 1.4b 1.5 1.6a 1.6b 1.6c D 1.2a Oct-15 Nov-15 D 1.2b Apr-16 Dec-15 Mar-16 May-16 Jul-16 Dec-16 Oct-16 Apr-15 Oct-15 Jul-16 Sep-16 Feb-17 May-17 D 1.3 Mar-17 Jun-17 D 1.4 Jun-17 1 Executive design of b3 b4 b5 Executive design of a2 and b6 MgB2 magnet design Octupole and Decapole Construction Dodecapole and Quadrupole Construction MgB2 magnet construction Test of the sextupole Test of the octupole and decapole Test of the quadrupole and dodecapole Corrector Magnet Test Report Corrector magnets Final Check & transport 2 3 4 5 6 7 2016 8 9 10 11 12 1 2 3 4 5 6 7 2017 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 3 month potential delay on the last deliverable Milestone Deliverable CERN-INFN Collaboration Agreement End Modified From the minutes of the CERN-INFN steering committee: “There is a potential 6 month delay (that so far has not impacted on Deliverables) due to longer design and R&D. It will appear as delay of next deliverable. However, G. Volpini is confident that part of the delay is recoverable, and the potential delay on the last deliverable could be of the order of three months. The CERN coordinator expresses an extremely positive judgement on the work done so far and supports the effort of INFN to recover the schedule without impacting of the quality of the work. However, the correctors are not on the critical path of HL LHC IR magnets schedule.” 19 Giovanni Volpini CSN V Frascati, 27 luglio 2015 WP2 PADS Progettazione dipoli di separazione (D2) per l’upgrade di luminosità di LHC. Trasparenze a cura di P. Fabbricatore Giovanni Volpini, CSN V, Catania 23 luglio 2014 20 The central critical point of D2 magnets Close to IR (ATLAS and CMS) two D2 magnets (one per side) have the same field orientation Multipole (units) 70 60 b2 105 50 b2 95 40 30 20 10 0 0 1 2 3 4 5 3 4 5 Field (T) 20 0 -20 Multipole (units) Preliminary studies (BNL): High harmonics due to cross talk!!!! -40 -60 -80 -100 b3 105 -120 -140 b3 95 -160 0 1 2 Field (T) A solution was found! Since the magnet suffers from magnetic cross talk between the two coils, we need to actively compensate this cross talk in a wide magnetic field range in the two apertures. This was done involving a strategy based on three pillars: a) No iron is placed in between the coils (so limiting saturation effects); b) Each coil is asymmetric in a way to cancel the magnetic cross talk each other. c) The yoke is suitably profiled for keeping constant the harmonic components . Based on these concepts a 2D magnetic optimization was carried out leading to an acceptable field quality with a limited variation of the multipoles as the magnet field is raised from the injection value to the maximum value (4.5T). Configuration INFN_3_3_6 2D Lay-out Rh Dy Dx R Main Characteristics Characteristics Aperture Units mm Number of apertures Value 105 2 Distance between apertures (cold/warm) mm 188.00/ 188.45 Cold mass outer diameter (min/max) mm 570/630 Magnetic length m 7.78 Bore field T 4.5 Peak field T 5.20 Current kA 12.050 Temperature K 1.9 Loadline margin (%) 35 Overall current density A/mm2 443 Differential inductance per meter mH/m 3.509 Stored energy MJ 2.18 Differential inductance mH 27.3 Superconductor Nb-Ti Field quality b2 and b3 optimised at a B field slightly lower than 4.5 T (4.3 T) 3 2 1 b2 0 b3 b4 -1 b5 b6 -2 b7 b8 b9 b10 -3 0 1 2 3 4 5 B (T) Coil end design is 90% done End with connections b3 (black) and a3 (red) components in the connection end Integrated harmonics still to be optimised (ex. b3 int. -5 unit) Mechanics: Plan B (single collar) Before the two-collar option, a solution with a single collar was completely studied. Not satisfactory because the complexity of the assembly and the differential thermal contraction. However this solution is still there as plan B in case of major problems with double collar option. Mechanical effects on b2 and be field harmonics (computed for a previous version) Presently we have engineering drawings of the 2D cross section Conclusions • The activities related to D2 development are progressing faster than foreseen • The 2 D magnetic and mechanical designs were completed • The 3D magnetic design is close to be completed • Still to do: 3 D mechanical design, 3D engineering and quench analyses Finanziamento 2016 Giovanni Volpini CSN V Frascati, 27 luglio 2015 32 Persone coinvolte Giovanni Volpini CSN V Frascati, 27 luglio 2015 33 WP3 SCOW-2G SuperCOnducting Windings - 2G tapes Obiettivo: Progettazione, realizzazione e caratterizzazione di avvolgimenti dimostrativi a doppio pancake realizzati in YBCO coated conductors, per verificare l’applicabilità di questa tecnologia alla realizzazione di magneti in alto campo, quali ad esempio quelli richiesti nei canali di raffreddamento muonici. Responsabile: Umberto Gambardella (SA) Giovanni Volpini, CSN V, Catania 23 luglio 2014 34 Attivita’ in corso 1. La completa caratterizzazione, in termini di corrente di trasporto in funzione del campo applicato, di nastri di diversa fabbricazione (SuperPower, SuNAM e SupeOx). 2. Analisi delle problematiche di impregnazione e di terminazioni dei primi avvolgimenti. 3. Il potenziamento delle stazioni di test a temperatura variabile con campo magnetico di background e cryogen free. Giovanni Volpini, CERN 14 January 2014 35 Apparato sperimentale Portacampioni Magnete cryogen-free 64.96 14 T=65 K B=0 T 64.94 Temperature (K) 12 Voltage (mV) Alle temperature di 77K, 70K e 65K, per ogni valore di campo magnetico è stata registrata la caratteristica I-V del nastro, ricavando da questa la corrente critica. Le misure sono state effettuate sia in flusso di gas che in bagno di azoto liquido pompato. La temperatura è misurata da un termometro Cernox. A destra è raffigurato un esempio di caratteristica I-V misurata, insieme all’andamento in temperatura La deriva termica è contenuta a 0.12 K. 10 8 6 4 2 0 200 64.92 64.9 64.88 64.86 64.84 64.82 250 300 350 Bias current(A) 400 450 64.8 200 250 300 350 Bias current(A) 400 36 450 Caratterizzazione dei nastri YBCO 2G 10 1 0 1 2 3 4 m H (T) 0 5 6 7 Ic @10mV/cm (A) 100 SuNAM H + tape surface SuperPower M4-217 H + tape surface Ic @10mV/cm (A) 65K 70K 75K 77K c I @ 10mV/cm (A) SuperOX H + tape surface 100 10 1 65K 70K 75K 77K 0 1 2 3 4 B (T) 5 6 7 8 65K 70K 75K 77K 100 10 1 0 1 2 3 4 m H (T) 5 6 7 0 I confronti delle prestazioni tra questi nastri individuano come miglior nastro a zero campo quelli SuNAM, mentre per le prestazioni in campo (ad es. a 2T) i nastri SuperPower risultano decisamente migliori. Comunque il rapporto qualità/prezzo (o più precisamente $/kA m) risulta a favore di SuNAM, ed è con questo tipo nastro che intendiamo proseguire le attività. Giovanni Volpini, CERN 14 January 2014 37 Test avvolgimento e impregnazione Per la realizzazione di avvolgimenti e’ necessario che le spire siano ben ferme, cosa che normalmente si ottiene impregnando i magneti con resine epossidiche. Nel caso dei nastri HTS 2G, gli stress termici tangenziali introdotti durante il raffredamento dalle contrazioni dell’impregnante sulla struttura esterna del nastro (cap layer in rame), facilmente degradano le prestazioni in corrente del nastro. Anche noi abbiamo riscontrato questo problema su diversi avvolgimenti di prova. Abbiamo quindi realizzato test coil da 5+5 spire impregnate con differenti soluzioni per la resina, ciclati più volte in bagno di LN2. Solo quando non si e’ manifestata degradazione alcuna in almeno 4 cicli termici siamo passati ad incrementare il numero di spire. L’ultima bobina di test è composta da 25+25 spire avvolte su un diametro di 40 mm e, con il tipo di resina messo a punto nelle precedenti prove, abbiamo riscontrato una sostanziale tenuta della corrente critica in autocampo a 77 K Ic=90 A, rispetto a 4 cicli termici. Giovanni Volpini, CERN 14 January 2014 38 Attrezzatura per il collaudo dei doppi pancake Per il collaudo dei doppi pancake è stato attivato il VTI a flusso (Variable Temperature Cryostat) in He gas inserito nel magnete cryofree da 12 T. Questo rappresenta un sistema economico e flessibile per testare piccoli avvolgimenti realizzati su diametro utile di 40-50 mm. Con avvolgimenti da 50+50 spire (massimo stimato come dimensione avvolta su 40 mm compatibile con il nostro VTI) si potrebbero raggiungere campi >2T a 40 K (con bias da 200 A). Una delle attrezzature in corso di realizzazione nel 2015 riguarda la supporteria per il test dei pancake nel VTI a flusso di He comprensiva di un discendente che fornisce contemporaneamente il sostegno meccanico ed i discendenti di corrente. Attualmente i componenti sono in fase di lavorazione Dopo la pausa estiva partira’ la realizzazione di un criostato in bagno di LN2 per il magnete cryofree, per alloggiare pancake con maggior numero di spire (a T= 65 K). Per questa attività esistono già disegni preliminari del contenitore in acciaio con diametro utile 80 mm. Restano da definire la supporteria ed i passanti di corrente per sostenere lo stack di pancake. Giovanni Volpini, CERN 14 January 2014 39 Avvolgimento cryogen-free E’ prevista la realizzazione di un avvolgimento compatibile con la modalità cryogen free. Per svolgere utilmente queste prove dovevamo incrementare le prestazioni dell’attrezzatura mostrata sopra (sx), innalzando la corrente dei passanti fino ad almeno 300 A. Ad oggi sono stati realizzati una coppia di adduttori di corrente HTS da 400 A con nastri ReBCO da 12 mm mostrati a dx. Nel 2016, quando sarà meglio definito il layout finale dello stack di pancake, ed i relativi former per la modalità cryogen free, si passerà alla progettazione dei componenti interni della camera Giovanni Volpini, CERN 14 January 2014 40 Nonostante un ritardo nell’avvio delle attività si ritiene che rispetto alla programmazione 2015 abbiamo raggiunto i seguenti risultati: 1. realizzata una stazione di test nel magnete cryogen free 12T con un inserto VTI a flusso forzato di He per misure su nastri HTS; 2. realizzato e testato uno dei componenti cruciali per la stazione di test cryogen free, cioè gli adduttori HTS da 400 A; 3. progettato il controcriostato in bagno LN2 per il magnete cryofree da 12T, adatto a contenere avvolgimenti di dimensioni fino a 80 mm; 4. progettato supporto per test pancake fino a 50 mm da inserire nel VTI in campo magnetico; 5. acquisito conoscenze sulle proprietà dei nastri di ultima generazione di fabbricanti diversi; 6. acquisito conoscenze per impregnare avvolgimenti senza degradazione delle prestazioni a seguito di cicli termici; 7. realizzate bobine “significative” (25+25 spire, con campo sul conduttore a ≈0.3 T in LN2) con primi approcci a terminazioni rigide. Giovanni Volpini, CERN 14 January 2014 41 Attività 2016 Nel 2016 il gruppo si concentrerà nella realizzazione dello stack a 10 pancake, da inserire nel VTI per raggiungere un campo da almeno 2T, e compatibilmente con le forze meccaniche, in presenza di un campo di background. Inoltre si avvierà la progettazione dello stack a 10 pancake in modalità cryogen free (raffreddato solo per conduzione) e la meccanica per l’inserimento nella camera cryogen free. Giovanni Volpini, CERN 14 January 2014 42 Finanziamento 2016 Giovanni Volpini CSN V Frascati, 27 luglio 2015 43 Persone coinvolte Giovanni Volpini CSN V Frascati, 27 luglio 2015 44 The End Infinite Stack Now, let’s consider the case with an infinite number of RCSMs stacked on top of each other with alternating orientation (RCSM∞). Closest to a 2D case for these magnets. For the radial field, same symmetries apply as for RCSM1. Longitudinal field Bz cancelled by successive magnets. G. Volpini, J. Rysti CERN 15 July 2015 46 Two coils • Two magnets with mirror orientation, and reversed current (RCSM2). z It possesses reflection symmetry w.r.t. a plane normal to z-axis, but –surprisingly- it has no other symmetry (apart from 120° for sextupole). Therefore +∞ 𝐵 𝑑𝑧 −∞ 𝑧 =0 so not net z-component But it turns out that its harmonic content is very high, lacking those symmetries which “cancel” specific harmonics. G. Volpini, J. Rysti CERN 15 July 2015 2D Mechanics LHC vs. HL-LHC corrector magnet comparison chart [A] [T.m] [H] 14 150 24.57 182 1.00 0.807 1.247 4.7 150 1.28 132 0.06 0.111 0.118 50 7.8 150 1.28 132 0.06 0.111 0.118 2 S 3 N MCSX 3 S MCSSX 70 4 N MCOX MCSOX 70 16 100 4.4 150 1.41 120 0.04 0.087 0.152 4 S 70 22 100 3.2 150 1.41 120 0.04 0.087 0.152 5 N 150 1.39 139 0.03 0.095 0.107 5 S 150 1.39 139 0.03 0.095 0.107 6 N 29.2 150 4.35 167 0.086 0.430 0.229 187/II [kJ] 6 S 150 0.92 163 0.017 0.089 0.052 MQSX MCSTX MCOSX MCTX MCSTX 70 2,116 550 [m] Differential Inductance @ Iop Magnetic Length Integrated field at r=50 mm Operating Current Stored energy Aperture [A] [mH] [mm] Rev 9 July 2014 [J] Inductance Order Type mm Operating Current HL-LHC Aperture Stored energy LHC 70 70 39 100 6 94 80 Giovanni Volpini, CERN 30 June 2015 Test Station 460 500 A current leads SC bus-bars Al-clad NbTi SC cable from Mu2e TS 300 λ-plate for subcooled LHe operation Single coil test stand (later 6-pole …) coil under test