Docstoc
EXCLUSIVE OFFER FOR DOCSTOC USERS
Try the all-new QuickBooks Online for FREE.  No credit card required.

KEK_CopperAddedSepCables

Document Sample
KEK_CopperAddedSepCables Powered By Docstoc
					  Fabrication and Performance of Nb3sn
  Rutherford-type Cable With Cu Added
        As a Separate Component
          (MT18 Paper 1c-a04)
                                 KEK Seminar Part 2
                                          October 30th, 2003

                                          Mirco Coccoli

20-24 October 2003                                     CERN
                     Superconducting Magnet Program
                                                               Mirco.Coccoli@cern.ch
   Slide 1/27
                                                Outline

1. Introduction
2. Mixed-strand Cable Manufacturing and Testing
3. MC Power Losses and Measurement
4. Heat Absorption and Spot Heater Test in MC
5. A more general approach: Copper Added as a Separate
   Component
6. Quench Propagation Performance Simulated
7. The Program: Fast Quenching - Radiation Proof Cable
8. Conclusions

20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
   Slide 2/27
                                         Introduction

 • Foreseeing future Hadron Colliders, such as a 35km
   radius VLHC, it is desirable to investigate low cost
   (conductor) solutions
 • From the stand point of overall conductor cost, it is
   desirable to minimize the amount of copper that is
   co-processed with the superconductor during strand
   fabrication.
 • A possible solution is to add a copper fraction at
   final, ie cabling stage

20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
   Slide 3/27
                                         Introduction

• A fraction of the copper added at the cabling stage opens some questions
   – Instability  to be investigated
   – High Jcu  investigate on Tmax
   – New Coupling Schemes  under investigation
• Typical parameters of this study investigation
   – Nb3Sn  open to study on other conductors
   – 60% Cu + 40% Sc cable
   – MC with 2/3 Sc strands (40% copper fraction)
   – 1/3 Cu strands (pure copper)  new degree of freedom
• Cost of the strand not depending on Sc content  33% reduction
• With larger Nb3Sn market  20% reduction
• Example VLHC: 30% of the machine cost is due to the conductor


20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
   Slide 4/27
                                MC Manufacturing




 • Two Mixed-strand Cables (MC) have been fabricated, winded
   in coils and tested
 • The first difficulty has been the matching of the elongation
   between the two types of strand, resulting in a mechanically
   instable cable (popped strands)
20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
   Slide 5/27
                        MC Electrodynamic Study


        49   10
  Pc Pc 8.8.4910
                                         
                       33 p s ,s w N sc  N
                          LL,pw2 2N s2,s2  N s s,sc2    B  
                                                               2
                                                                   • P~Pc(FO)
                                                          B   
                                             R
                                             Rcc
                                                        NsNs,sc
    Pc ,MC   N s2,sc,MC  N s,sc,MC    142  14   0.44
                                             2
                                              21  21 
   Pc ,TRAD N s ,sc ,TRAD  N s ,sc ,TRAD  
               2                                       
                                                       
- Less eddy current loops  Smaller power loss
- This feature has been measured (M. D. Sumption et al, “AC Loss of
   Nb3Sn-based Rutherford Cables with Internally and Externally Added
   Cu” ASC02) on the tested cable and in agreement with the formula
  20-24 October 2003                                      CERN
                       Superconducting Magnet Program
                                                                      Mirco.Coccoli@cern.ch
     Slide 6/27
                     MC Electrodynamic Study (2)


                              is , cab  C
                                                       
                                                 Lp , s N s2, sc  4 N s , sc   
                                                             Rc



                  is , cab, MC  Ns2, sc, MC  4 Ns, sc, MC   0.4
                                 2
                 is , cab,TRAD N s , sc ,TRAD  4 N s , sc ,TRAD 
                                 Faster eddy currents decay

20-24 October 2003                                          CERN
                      Superconducting Magnet Program
                                                                                    Mirco.Coccoli@cern.ch
   Slide 7/27
                                Losses Measurement

                                                                                    25


• Face On losses
                                                                                             Mixed
                                                                                             Control




                                       Total per-Cycle Loss, Qt (10 J/m of cable)
                                                                                    20
  dominant with
                                                                       3
  respect to Edge On                                                                15
                                                                                                                                          FO


                                                                   4
• MC has half of the
                                                                                    10
  power losses as
  expected by the                                                                    5
                                                                                                                                          EO


  formula
                                                                                     0
                                                                                         0       20       40              60        80           100
                                                                                                          Frequency, f (mHz)


  20-24 October 2003                                                                                   CERN
                       Superconducting Magnet Program
                                                                                                                         Mirco.Coccoli@cern.ch
     Slide 8/27
                      Subscale Magnet Test Facility
                               (SMTF) (1)

Common Coil Magnet
 x-section scale 1:1
F = 75 cm, L = 1 m)




  Subscale Magnet
 x-section scale 1:3



 20-24 October 2003                                     CERN
                       Superconducting Magnet Program
                                                               Mirco.Coccoli@cern.ch
    Slide 9/27
                     Subscale Magnet Test Facility
                              (SMTF) (2)




20-24 October 2003                                     CERN
                      Superconducting Magnet Program
                                                              Mirco.Coccoli@cern.ch
  Slide 10/27
                     Mixed Strand Test in SMTF (1)

 • The mechanical problem noticed during the cable
   fabrication have been shown by the tests performed in the
   Subscale Magnet Test Facility (SMTF) at Berkeley Lab
 • Two coils have been wound out of Mixed Strand cables
    – Simple mixed strand
    – Mixed strand cable with SS core for mechanical stability
 • Results of the Power tests have not been satisfactory (40%
   and 70% of ss limit) leading to the conclusion that a good
   compaction in these coils is too difficult to achieve


20-24 October 2003                                     CERN
                      Superconducting Magnet Program
                                                              Mirco.Coccoli@cern.ch
  Slide 11/27
                     Mixed Strand Test in SMTF (2)




20-24 October 2003                                     CERN
                      Superconducting Magnet Program
                                                              Mirco.Coccoli@cern.ch
  Slide 12/27
                                   Mixed Strand Test in SMTF (3)
         10000

              9000
                                                                           Theoric Values
              8000                                                            for MC
              7000
Iquench/Iss




              6000
                              Tested traditional
              5000            Rutherford Cable                                                                            SM01



              4000

              3000

              2000       VLHC
              1000

                 0
                     0         50        100           150 .         200        250         300       350       400
                                                           I (A/s)
              20-24 October 2003                                              CERN
                                    Superconducting Magnet Program
                                                                                                  Mirco.Coccoli@cern.ch
                Slide 13/27
                                   Mixed Strand Test in SMTF (4)
                 1

                0.9
                                                                                            • Short Sample
                0.8
                                                                                              limit reached
                0.7
                                                                                              in Nb3Sn coil
Iquench/Iss




                0.6

                0.5                                                                         • MC 1 reached
                0.4                                                                           ~40% ss
                0.3

                0.2
                                                                                            • MC 2 reached
                0.1
                                                                                              ~70% ss
                 0
                      0   0.05      0.1      0.15
                                                     . 0.2        0.25   0.3   0.35   0.4

                                                     B (T/s)

              20-24 October 2003                                           CERN
                                     Superconducting Magnet Program
                                                                                             Mirco.Coccoli@cern.ch
                Slide 14/27
                              Heat Absorption and
                              Spot Heater Test (1)
 • A source of heat from outside the cable




                          One Layer Schematically…




20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
  Slide 15/27
                              Heat Absorption and
                              Spot Heater Test (2)
 • In the MC there are preferential heat path (far from the sc)




                         One Layer Schematically…




20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
  Slide 16/27
                              Heat Absorption and
                              Spot Heater Test (3)




20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
  Slide 17/27
                              Heat Absorption and
                              Spot Heater Test (4)
 • The heat loads has to come from outside the conductor
   (epoxy impregnation cracking, stick slip, collective
   motions, resistive heating, ecc.)
 • The price paid is the intrinsic stability of the strand
 • It has been measured that the Rc is lower on the average for
   a MC because of the good quality of the copper
 • The better quality cable, i.e. the one with a SS core does
   not have this feature
 • Anyway, the results suggest a change in direction…


20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                             Mirco.Coccoli@cern.ch
  Slide 18/27
                     Copper Added as a Separate
                            Component
 • Add the copper as a core
    – Pure copper cored  poor quality
    – SS-Cu strip assembled in a “sandwich stile” ss-cu-ss

                                                             Ready a coil
                                                             to be tested in SMTF
      – Most likely alternative: cu-ss-cu  Good mechanical stability
                                       Good electro-dynamical stability?
                                                             Ready a coil
                                                             to be tested in SMTF


20-24 October 2003                                    CERN
                     Superconducting Magnet Program
                                                               Mirco.Coccoli@cern.ch
  Slide 19/27
                       Quench Speed Simulations:
                             MC vs Trad
• Copper Added
  separately as a core                                      1000
  means no advantage for
  the power losses (~)


                                            Vquench (m/s)
                                                             100
• Protection 
10 times faster quench
• Subject to verification                                     10

  in magnet tests (LBNL)
• No quench heaters in                                         1
                                                                                                               Imag/Iss
  Future Accel Magnets?                                        65%     70%          75%   80%     85%          90%
                                                                     Mixed Strand                   Baseline



  20-24 October 2003                                                    CERN
                       Superconducting Magnet Program
                                                                                          Mirco.Coccoli@cern.ch
    Slide 20/27
                       Quench Speed Simulations:
                                 Gel
• Two “wires” model 1400
• Simulated the         1200

  dependence from the 1000
  electrical conductance 800
  between the sc strand
  and the cu strip       600

• Quench speeds in the 400
  1km range              200

• To be measured in a 0
  test…                    65%                                       75%                 85%
                                                  Gel 200 SC-RRR40    Gel 100 SC-RRR40    Gel 10 SC-RRR40

  20-24 October 2003                                         CERN
                       Superconducting Magnet Program
                                                                                Mirco.Coccoli@cern.ch
    Slide 21/27
                       Quench Speed Simulations:
                          added copper RRR
• Added copper RRR                                          500
  effect on quench speed                                    450
• Small effect  one                                        400




                                            Vquench (m/s)
  degree of freedom added                                   350
  to the cable design
                                                            300
• High Copper RRR                                          250
  better heat conduction
                                                            200
  property  lower peak
  temperatures                                              150
                                                            100
• Simulations and practical                                   65%                 75%                 85%       Imag/Iss
  test are foreseen
                                                              Gel 100 Cu-RRR400   Gel 100 Cu-RRR200   Gel 100 Cu-RRR50


  20-24 October 2003                                                     CERN
                       Superconducting Magnet Program
                                                                                             Mirco.Coccoli@cern.ch
    Slide 22/27
                       Quench Speed Simulations:
                           SC strand RRR
• Effects of Sc strand RRR                      500
                                                450
• Good for low RRR
                                                400
• Of course there is a lower                    350
  limit
                                                300
• Typical choice ~ 40        250
  good enough for quench
                              200
  speed
                              150
• Stability must be addressed100
  as a next step of this study 65%                                         75%                 85%

                                                        Gel 100 SC-RRR40   Gel 100 SC-RRR300   Gel 100 SC-RRR150


  20-24 October 2003                                            CERN
                       Superconducting Magnet Program
                                                                                    Mirco.Coccoli@cern.ch
    Slide 23/27
                                  Quench Speed Model
                      Normal Zone                           Superconducting Zone


                 R (z)                                    I1(z0)           I1(z)
Nb3Sn+Cu

           Gel              Gel               Gel         Gel       Gel            Gel

 Cu
              R (z)
                                                          I2(z0)               I2(z)

                                                                                         z
      • The current redistributes in the nearby copper
        passing trough the Gel  initial hot spot
      • The current flows in the pure copper and
        redistributes back to the sc strand
 20-24 October 2003                                         CERN
                         Superconducting Magnet Program
                                                                           Mirco.Coccoli@cern.ch
   Slide 24/27
                                            Conclusions

• Several alternative methods for adding Cu to a Rutherford-type
  superconducting cable have been investigated
• This cabling technique begun as a cost effective approach to
  conductor/cabling in view of future HEP accelerator magnets (VLHC)
• The hypothesis of protection advantages related to quench propagation
  velocity is being investigated
• The thermal conduction channel represented by the separately added
  fraction of copper has been measured to be a mean of stability against
  “external” heat/energy sources  more test needed
• Good performance in actual magnet coils has not yet been
  demonstrated (<first half of 2004 at LBNL?)
• Testing in the Subscale Magnet Test Facility is neither a trivial nor an
  expensive task
  20-24 October 2003                                    CERN
                       Superconducting Magnet Program
                                                               Mirco.Coccoli@cern.ch
    Slide 25/27
                                                      End




20-24 October 2003                                      CERN
                     Superconducting Magnet Program
                                                               Mirco.Coccoli@cern.ch
  Slide 26/27
                                                 Auxiliary Page


                            rMC 
                                      2                   



                          
                                 
                                       L  N   4 N
                                    2 p              
                                        60 3 2 d s , MC
                                                                      3
                            rTRAD C 40, s 2 s , sc d s ,TRAD s , sc 2
                            is , cab
                                                          Rc
                                                                              
                                           L p, f B a                  1 Lp, f B a
      d III,MC                                                       
                   d ,s*cab, MC  2N s2,J c, eff  4 N3s , sc ,1  3 J c  eff
                     is ,MC
                               1
                                   MC sc MC   MC  0.4  3
     d III,TRAD
                   *
                    is ,TRAD           
                   d,scab,TRAD N s2, sc ,TRAD  4 N2s , sc ,TRAD
                                            
                                     L p, f B a
                                                                               
                                                                       3 L p, f B a
                                                                                    2
                               1                                 1
                                  2 TRAD J c  eff                   4 J c  eff




20-24 October 2003                                                   CERN
                          Superconducting Magnet Program
                                                                                        Mirco.Coccoli@cern.ch
  Slide 27/27

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:11
posted:7/15/2011
language:English
pages:27