Template

1
BNL - FNAL - LBNL - SLAC
TQ Results LARP CM11 October 27-28, 2008 Gian
Luca Sabbi
2
TQC and TQS Designs
TQC (Collar)
TQS (Shell)

• Aluminum shell over iron yoke
• Assembly with bladders and keys
• Full preload after cool-down
• Rods/plates for high axial pre-load

• Stainless steel collars and skin
• Control spacers to limit pre-load
• Full pre-load at room temperature
• End support plates, low pre-load

3
TQ Short Sample Limit
MJR strand (TQ01 Models) Jc 1.9 kA/mm2 (12 T,
4.2 K)
RRP strand (TQ02 models) Jc 2.7 kA/mm2 (12 T,
4.2 K)
Magnet Top K Gss T/m Iss kA
S01a,b,c 4.4 216 12.1
S01a,b,c 1.9 234 13.2
C01a (b) 4.4 209 (201) 12.6 (12.1)
C01a (b) 1.9 228 (218) 13.7 (13.1)
Magnet Top K Gss T/m Iss kA
S02a,b,c 4.4 239 13.6
S02a,b,c 1.9 260 14.9
C02a,e 4.4 224 13.6
C02a,e 1.9 244 14.8

• TQC iron yoke is located further away from the
  coil than TQS (SS collar)
• - Causes shallower load line, lower short sample
  gradient
• From TQC01b, iron yoke was extended above the
  ends for mechanical reasons
• - End peak field increase causes further loss of
  short sample gradient

Same fraction of short sample results in 6
lower gradient for TQC than TQS
4
TQC Tests
Magnet Conductor Coils Island Temperature Test
TQC01a MJR 9,10,12,13 Bronze 4.4 K 1.9K August 2006 FNAL
TQC01b MJR 7,8,10,12 Bronze 4.4 K 1.9K August 2007 FNAL
TQC02e RRP 20, 21, 22, 23 Titanium 4.4 K 1.9 K October 2007 FNAL
TQC02a RRP 17, 19, 24, 27 Bronze 4.4 K 1.9 K January 2008 FNAL
TQC02b MJR/RRP 10, 12, 17, 19 Bronze 4.4 K 1.9 K Aug 2008 FNAL
New models virgin coils are in bold font.
Rebuilt models pre-trained coils are in italic
font.
5
TQS Tests
Magnet Conductor Coils Island Temperature Test
TQS01a MJR 5, 6, 7, 8 Bronze 4.4 K April 2006 LBNL
TQS01b MJR 14, 15, 7, 8 Bronze 4.4 K Nov. 2006 LBNL
TQS01c MJR 5, 15, 7, 8 Bronze 4.4 K 1.9 K March 2007 FNAL
TQS02a RRP 20, 21, 22, 23 Titanium 4.4 K 1.9 K June 2007 FNAL
TQS02b RRP 22, 23, 28, 29 Titanium 4.4 K 1.9 K March 2008 CERN
TQS02c RRP 22, 23, 28, 20 Titanium 4.4 K 1.9 K Jun Sep 08 CERN
New models virgin coils are in bold font.
Rebuilt models pre-trained coils are in italic
font.
6
TQ Quench Performance Summary
Model First Training at 4.4K First Training at 4.4K First Training at 4.4K First Training at 1.9K First Training at 1.9K First Training at 1.9K Highest Quench Highest Quench
GStart (T/m) GMax (T/m) Gmax/Gss () GStart (T/m) GMax (T/m) Gmax/Gss () GMax (T/m) GMax quench conditions
TQC01a 131 154 72 151 196 87 200 1.9K, 100A/s
TQC01b 142 178 86 179 200 90 200 1.9K
TQC02E 177 201 87 198 199 79 201 4.4K
TQC02a 124 157 68 145 164 65 169 1.9K, 50 A/s
TQC02b 141 173 85 158 173 78 175 3.6K, 50A/s
TQS01a 180 193 89 n/a n/a n/a 200 3.2K
TQS01b 168 182 84 n/a n/a n/a 182 4.4K
TQS01c 159 176 81 176 191 82 191 1.9K
TQS02a 182 219 92 214 221 85 222 2.2K
TQS02b 190 200 84 196 205 79 205 1.9K
TQS02c 216 222 93 205 209 80 231 2.7K
() Highest Quench value includes all Training,
Ramp Rate Intermediate Temperature Quenches
7
Discussion Points

• TQC-specific
• Are the coil stress levels known and understood?
• Is there sufficient tolerance to assembly
  parameters?
• TQS-specific
• Which factors determined the TQS plateau levels
  at 4.5K 1.9K?
• Overall results vs expectations
• Did the TQ program achieve its original
  objectives?
• Are there additional TQ objectives, and what is
  their priority?

8
TQC Coil Stress
Preload at inner pole in TQC models (MPa) TQC01b
data is listed separately for coils with
glued/non-glued outer poles.
9
TQC Models Analysis
TQC01a Viability of collaring process
demonstrated Low preload limited the quench
plateau in TQC01a at 4.5K and resulted in outer
layer mid-plane damage at 1.9K at the end of
magnet test. TQC01b Analysis incorporated
plastic coil behavior, resulting in appropriate
preload. Reached 88 of its short sample
gradient at 4.5K and 92 at 1.9K.
TQC02E Coils re-used from TQS02a, collared
without damage. Achieved 90 of its short sample
gradient at 4.5K and 81 at 1.9K After
disassembly, coils were re-used in TQS02b/c, with
no indication of degradation during assembly,
testing or disassembly of TQC structure.
10
TQC Models Analysis (2)
TQC02a TQC02a was first collared in March 2007
results from strain gauges as well as collar
deflection measurements were consistent with
analysis. The magnet was re-assembled in November
2007 with increased preload during the collaring
phase. The reassembled magnet included two new
coils, replacing two that may have been damaged
due to a misplaced mid-plane shim during the
initial assembly. Limited quench plateau in the
new coils indicates probable degradation during
construction due to higher collaring preload and
large increments during early collaring
steps. TQC02b Return to TQC01b preload
parameters and additional early collaring steps
in TQC02b will allow return to TQC02E/TQC01b
performance or better.
11
TQS01 Analysis of Plateau Quenches

• TQS01 plateau quenches are very repeatable (all
  segment dV/dt are overlays)
• Quench propagation velocity along pole turn can
  be correlated with local Ic
• Calculated gap Ic in coil 6 consistent with
  87 quench level (already in Q4)
• TQS01b TQS01c limit also consistent with Ic
  degradation (in different coils)

SQ02 Measured Quench Velocities
Iq/Iss
12
TQS01 Coil Stress near Pole Gaps

• 3D ANSYS calculations and TQS01b measurements
• indicate high longitudinal tension in coil
  across gaps,
• possibly leading to conductor degradation
• This effect depends on the interfaces between
  coil,
• pole (bronze or titanium) and outer support
  elements

Differential measurements of coil axial strain in
TQS01b
3. Training
4. Warm-up
2. Cool-down
1. Assembly
13
TQS02a Quench Analysis

• All plateau quenches in coil 21 (coil 21
  participates in initial training)
• Quenches start in the pole turn and the
  multi-turn of the outer layer
• Voltage signal overlays indicate same origin and
  evolution
• Proximity of quench onsets indicates start is
  near the pole turn
• FEA shows very high stress with Ti poles in
  outer pole turn
• Comparison of 1st and 2nd plateau shows further
  training occurred at 1.9K
• At 1.9 K, same limiting coil and same
  segments/sequence as 4.5K

14
TQS02b Quench Analysis

• All quenches (4.5K 1.9K) start in coil 29 at
  same location (pole turn)
• At 1.9K, 20A/s ramps resulted in lower quench
  current than at 4.5K
• At 1.9K, higher ramp rates resulted in higher
  quench currents
• Quench location correlates with kinks after
  un-winding/re-winding

15
TQS02c-TC1 Quench Analysis

• All Quenches are in coil 23 B6A10 (ramp between
  inner and outer layer)
• 4.5K start is at about 110 mm from the V tap A10
  (20A/s ramp rate)
• 1.9K start is at about 210 mm from the V tap A10
  (20A/s ramp rate)

• Exceptions
• the beginning of the training for q1-q2
  quenches are in coil 28
• ramp rate gt20A/s (4.2K) and ramp rate gt140A/s
  (1.9K) quenches are in coil 23 A2A4
• ramp rate lt20A/s (4.2K) when quenches are in coil
  23 B4B3

15
16
TQS02c-TC2 Quench Analysis
17
TQS Summary

• Structure performance
• Demonstrated accurate, repeatable control of
  warm and cold pre-load
• Delivered safe, fast assembly and disassembly
  for all 6 models
• Training starts high and improves rapidly to
  maximum no retraining
• All TQ02 models surpassed target gradient,
  achieved up to 230 T/m
• Coil Performance
• TQS01 Degradation due to high stress at pole
  gaps (several coils)
• TQS02a Degradation due to high stress in outer
  layer (coil 21)
• TQS02b Degradation due to unwinding/rewinding
  (coil 29)
• TQS02c Degradation in the inter-layer ramp
  region (coil 23)

18
Conductor Stability

• We expected that conductor instabilities may
  affect SQ/TQ/LR/LQ
• Conductor tests indicated limited margin at
  operating points
• Cabling and HT parameters were optimized to
  maximize margin
• Results?

• No instabilities in SQ02 and TQ01 models at 4.5
  K 1.9K (MJR)
• No stability limitations in SR/LR at 4.5 K (RRP)
• No stability limitations in TQ02 models at 4.5 K
  (RRP)
• Signs of instability in TQC02 models at 1.9K
  (both MJR RRP)
• Signs of instability in TQS02 models at 1.9K
  (RRP)

Question are instabilities observed in magnets
induced by conductor damage at some step in the
fabrication/test process?
19
TQ Objectives (2005 Tech Review)

• Phase 1 (TQ baseline) TQS01 TQC01
• Develop and verify the coil fabrication
  technology
• Perform a consistent comparison among mechanical
  support concepts
• Same conductor/cable and same coil design
• Same fabrication procedures for all coils
• Consistent instrumentation and documentation
• Lay groundwork for Lab collaboration
• Agree on coil module design and fabrication
• Share fabrication and instrumentation practices
• Integrate coil fabrication steps
• Integrate design, fabrication and testing teams

20
TQ Objectives (2005 Tech Review)

• Phase 2 (mechanical conductor studies) TQS02,
  TQC02, TQE01
• TQS02 reconfigure/optimize based on TQS01
  results
• Adjust pre-load conditions, assembly sequence
• If needed, replace one or more coils
• TQC02 first TQ coils made with baseline 54/61
  RRP strand
• No changes in the coil/cable design are expected
• Optimization of mechanical structure w/TQC01
  feedback
• TQE01 disassemble TQC01 coil and test in TQS01
  structure
• Compare coil performance in different structures
• (evaluate based on TQS01 and TQC01 results)
• Explore the feasibility of collared coil
  disassembly

21
TQ Objectives (2005 Tech Review)

• Phase 3 (optimized models) TQS03 and TQC03
• Demonstrate the TQ objectives
• consistently achieve Ggt200 T/m after training and
  thermal cycle
• evaluate the required design margins (fraction of
  short sample)
• characterize the mechanical performance of the
  two structures
• Support the follow-on model magnet RD
• LQ models
• Provide the optimized coil design for LQ
• Feedback on coil fabrication methods (integrate
  with LR)
• Input for structure selection (integrate with DS)
• HQ models
• Design methods/tools and coil fabrication
  technology

22
TQ Objectives Summary

• Optimize/characterize cable design and heat
  treatment cycle
• Evaluate conductor/cable performance and
  stability
• Develop and optimize coil fabrication/handling
  procedures
• Optimize and finalize the coil design for LQ
• Develop/calibrate FEA models
• Compare mechanical design concepts and support
  structures
• Compare test data with expected (design) values
• Evaluate and optimize quench protection
  parameters for LQ
• Develop instrumentation and diagnostic
  techniques
• Provide experimental feedback for LQ and HQ
  structure selection
• Consistently achive Ggt200 T/m after training and
  thermal cycle

Can we declare that these objectives have been
achieved?
23
Current Future TQ Objectives

• Evaluate new conductor on well-characterized
  coil/structure
• Investigate role of end pre-load in shell-based
  models
• Improve performance of collar-based models
• Other?
• Which priorities, in view of the main LARP
  milestones?