View
2
Download
0
Category
Preview:
Citation preview
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 1
http://www.bnl.gov/magnets/staff/gupta
Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole
Ramesh Gupta and Bill SampsonSuperconducting Magnet DivisionBrookhaven National Laboratory
Upton, NY 11973 USA
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 2
Contents
Magnet Test
• Objectives
• Comparison between expectations and measurements
• Lessons learned, unanswered questions (see Bill’s presentation)
Bill Sampson will present
• Measurements (including important details)
• New Splice
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 3
Objectives
Objectives of this particular magnet test:
• Demonstration of HTS technology in an accelerator R&D magnet
• Demonstration of the ability of this design to tolerate high energy deposition
present in RIA (originally this simulation was slated for the next year)
• Demonstration of the conduction cooling (not original part of RIA program)
Thus, so far we are way ahead of what we were set out to do by now.
• We are using this opportunity to not only satisfy the RIA requirements but
also develop a program that opens up HTS technology for future magnets.
We successfully demonstrated the ability of this design to withstand RIA
type energy deposition and (b) the conduction cooling (Bill’s presentation).
• However, there are a few observations that are yet to be fully understood.
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 4
RIA HTS Quadrupole At Various Stages of Construction and Testing
Cold iron magnetic mirror test with six coils
Warm iron magnetic mirror test with twelve coils
HTS coil winding with SS tape insulator
HTS coils during magnet assembly
The RIA HTS model magnet has been successfully built and tested at BNL. Experiments of magnet operating with large energy depositions (tens of watts in 0.3 meter long magnet) have also been carried out.
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 5
Axial Scan of the Field Gradient
Design Gradient
Design Gradient
Mirror model with 6 coils (as required for simulation of full magnet with 12 coils).
Mirror model with twelve coils (2X of required)
@150 A135 A will do(@35 K)
@ 200 A(obtained
@5 K)
Higher gradient on two sides because the magnet is too short (end effects). However, higher peak field does not harm (HTS is great).
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 6
Ic Vs Temperature in Magnetic Mirror Model (when 2, 4, 6 or 12 coils are energized)
More coils create more field and hence would have lower Ic at the same temperature
0
50
100
150
200
250
300
0 10 20 30 40 50 60 70 80 90 100Tempratue (K)
Cur
rent
@ 0
.1 µ
V/cm
(A)
Two CoilsFour CoilsSix CoilsTwelve Coils
Comparison between measurements and calculations of Ic are much more involved in this HTS magnet than in conventional LTS magnet because (a) there is a significant difference between field parallel and field perpendicular Ic, with ratio being a strong function of temperature (b) requires 3-d analysis because the magnet is very short
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 7
Impact of Parallel and Perpendicular Field in the Ic of HTS (2223 Tape)
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0 1 2 3 4 5 6B(T)
Fiel
d pa
ra/F
ield
per
p R
atio
4.2 K20K35K
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
16.0
0 1 2 3 4 5 6
B(T)
Fiel
d pa
ra/F
ield
per
p R
atio
4.2 K20K35K50K
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 8
Field in RIA Warm Iron Design with 12 coils (175 turns each) at 200 A
Bmod (T) in iron and coil
Bmod (T) in coil (iron hidden)
B-parallel (T) in coil (iron hidden)
B-perpendicular (T) in coil (iron hidden)
Max 3.54 T
Max 3.54 T
Max 2.25 T
Max 3.79 T(in iron)
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 9
Coil Only Calculations(12 coils at 200A, 175 turns each)
• It turns out that the coil only calculations gives field within a few % level of actual field (in particular at high fields).• Coil only analysis is much faster and relative accuracy is much better.
Field parallel
Field perpendicular
Max 3.53 T
Max 2.31 T
We need to scan parallel and perpendicular field components to study
various scenarios
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 10
Perpendicular Component of the Field While Slicing through the Coil
2.31 T 1.81 T 1.42 T
Bottom Surface
0.99 T 0.64 T 0.0 T
Middle Surface
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 11
Parallel Component of the Field While Slicing through the Coil
2.40 T 2.87 T
Bottom Surface
3.11 T
3.30 T 3.41 T 3.49 T
Middle Surface
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 12
Variation of Field with Current for Different Number of Coils
200 A in coil (total amp-turns icreases with number of coils)0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0 2 4 6 8 10 12 14No. of Coils
Fiel
d pa
ralle
l & F
ield
Pe
rpee
ndic
ular
sqrt(Bx^2+Bz^2)
By(T)
Total Amp turn = 0.42 MA.turns
0
1
2
3
4
5
6
7
0 2 4 6 8 10 12 14No. of Coils
Fiel
d pa
ralle
l & F
ield
Pe
rpee
ndic
ular
sqrt(Bx^2+Bz^2)By(T)
Total Amp-turns is constant(normalized to 200 A in 12 coils)
Amp/turn is constant(200 A irrespective of
number of coils)
Parallel/Perpendicular ratio is 1.64 for 2 coils and ~1.53 for rest
0
50
100
150
200
250
300
0 10 20 30 40 50 60 70 80 90 100Tempratue (K)
Cur
rent
@ 0
.1 µ
V/cm
(A)
Two CoilsFour CoilsSix CoilsTwelve Coils
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0 1 2 3 4 5 6B(T)
Fiel
d pa
ra/F
ield
per
p R
atio
4.2 K20K35K
The ratio is ~1.6 for 4 to 20 K. And that’s about what we have !!!
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 13
Comparison Between Calculations and Measurements
0
50
100
150
200
250
300
0 10 20 30 40 50 60 70 80 90 100Tempratue (K)
Cur
rent
@ 0
.1 µ
V/cm
(A)
Two CoilsFour CoilsSix CoilsTwelve Coils
Field Perpendicular Scaling Factor at Various Field (T)
0
1
2
3
4
5
6
0 10 20 30 40 50 60 70 80
Temperature (K)
Scal
ing
Fact
or
0.10.20.30.40.5123
Field ParallelScaling Factor
at Various Field (T)
0
1
2
3
4
5
6
0 10 20 30 40 50 60 70 80Temperature (K)
Scal
ing
Fact
or
0.10.20.30.40.5123
• We specify critical current at 0.1 µV/cm whereas HTS industry specifies at 1 µV/cm.• The detailed calculations performed in select cases show that we are measuring 5-20% more current than expected. • At high temperature the current and field is low and one needs to do non-linear iron calculations.• Higher Ic may be due to re-distribution of currents. • There is a significant uncertainty in scaling also.• Bottom line: It meets the RIA requirements.
Superconducting Magnet Division
Ramesh Gupta, Test Results and Analysis of RIA HTS Magnetic Mirror Quadrupole, June 8, 2006. Slide No. 14
Summary of RIA HTS Magnet R&D Program
We designed RIA R&D program not only to build a magnet
• Most people would have started with a set objective of building magnet and would still be building components with no results to show yet.
However, we designed a flexible RIA R&D program to develop and prove HTS technology along the way with a number of test demonstrations
• We have done reasonably well in proving the technology despite the fact that we have not yet built the magnet. We and other people are already discussing what we have demonstrated and are making future programs based on that. In fact, building magnet now appears to be a routine task that must be done for the sake of completeness.
• This is due to a step by step program that we instituted. This included 77 K coil testing, magnetic mirror model with cold iron design, magnetic mirror model with warm iron design and now to the real magnet…
• We used a prudent combination of both conservatism (e.g., choosing conductor with stainless steel backing) and looked and grabbed new opportunities (for example, conduction cooling, energy deposition test).
Recommended