Quench

A. Verweij, TE-MPE. 2 Feb 2009, LHC Performance Workshop – Chamonix 2009

Training the dipolesTraining the dipoles

- Training during HWC- Training during HWC

- Estimate to reach 6, 6.5, and 7 TeV- Estimate to reach 6, 6.5, and 7 TeV

- Quench propagation- Quench propagation

Arjan Verweij, TE-MPEArjan Verweij, TE-MPE

Acknowledgment to everybody involved in the HWC


QuenchQuenchDipole Quench: Transition from the superconducting to the normal state

(usually due to local temperature rise), resulting in a detectable resistive

voltage, exceeding the threshold voltage for a duration larger than the

discrimination time.

Quench classification:

Heater induced/provoked quench

Natural (training) quench

Secondary quench (due to

increase of the bath temperature,

ramp rate, etc)

(Beam induced quench)

Calculations based on experiments, G. Willering, PhD thesis 2009

A local energy deposition of about

100 J is enough to trigger a quench

(i.e. a falling raindrop of 2 mm

diameter).

Circuit quench: Heater

induced/Natural quench +

secondary quenches.


Sector 1st training quench [A]

I_max

[A]# training quenches

Starting in:

# ALS # ANS # NOE

1-2 - 9310 0 0 0 0

2-3 - 9310 0 0 0 0

3-4 - 8715 (bus) 0 0 0 0

4-5 9789 10274 3 0 0 3

5-6 10004 11173 27 0 1 26

6-7 - 9310 0 0 0 0

7-8 8965 9310 1 0 1 0

8-1 - 9310 0 0 0 0

RB circuit quenches during HWCRB circuit quenches during HWC


Magnet distribution per sectorMagnet distribution per sector

4956 56

46

28

57 5464

96

60 65

46

42

36 40 24

9

38 33

62

84

61 60 66

0

20

40

60

80

100

120

140

160

S12 S23 S34 S45 S56 S67 S78 S81

3 (NOE)

2 (ANS)

1 (ALS)


Nr. of quenches in SM-18 to reach given currentNr. of quenches in SM-18 to reach given current

7000

8000

9000

10000

11000

12000

13000

10 100 1000 10000

Number of quenches

Cu

rren

t [A

]

All dipoles

Dipoles used in S56

x 11

x 6.8


Training during HWCTraining during HWC

8500

9000

9500

10000

10500

11000

11500

12000

12500

13000

1 10 100 1000

Quench number

Qu

ench

cu

rren

t [A

]

5.5 TeV

7 TeV

6.5 TeV

6 TeV

S45

S56 in SM-18

S56

S78

190


RB circuit S45: comparison with SM-18RB circuit S45: comparison with SM-18

7000

8000

9000

10000

11000

12000

13000

0 20 40 60 80 100 120 140

Magnet number (order by 1st quench in SM18)

Cu

rren

t [A

]

ALS: 1st training quench SM18

ANS: 1st training quench SM18

NOE: 1st training quench SM18

NOE: Quenches sector 45

7 TeV

6 TeV


RB circuit S56: comparison with SM-18RB circuit S56: comparison with SM-18

7000

8000

9000

10000

11000

12000

13000

0 20 40 60 80 100 120 140

Magnet number (ordered by 1st quench in SM18)

Cu

rren

t [A

]

ALS: 1st training quench SM18

ANS: 1st training quench SM18

NOE: 1st training quench SM18

ANS: Quenches sector 56

NOE: Quenches sector 56

7 TeV

6 TeV


Retesting of NOE magnet 3383 in SM-18Retesting of NOE magnet 3383 in SM-18

11691

1069110544

11304

12346

7000

8000

9000

10000

11000

12000

13000

Cu

rren

t [A

]

Quench 1st test (Aug 2005)

Quenches 2nd test (Oct 2008)

Stable operation at 12850 A

1st cold test (Aug 2005) 2nd cold test (Oct 2008)

Data from N. Catalan Lasheras

Note: 3383 suffered a transport accident, during which diode, IFS box and N-line were damaged


Sector

Number of magnets Number of quenches

ALS ANS NOE@ 6 TeV

(±2)

@ 6.5 TeV

(±30%)

1-2 49 96 9 00 44

2-3 56 60 38 11 88

3-4 56 65 33 11 88

4-5 46 46 62 22 1212

5-6 28 42 84 11 1515

6-7 57 36 61 22 1212

7-8 54 40 60 22 1212

8-1 64 24 66 22 1313

Total 154 154 154 1111 8484

Estimated dipole training to reach 6 and 6.5 TeVEstimated dipole training to reach 6 and 6.5 TeV

Est. 1: Based on 115 MB’s that have been submitted to a thermal cycle in SM-18 (2008 before HWC, P. Xydi and A. Siemko)

Est. 2: Extrapolation from sector 5-6 data + est. 1 for ALS & ANS

Est. 3: 2 quenches per NOE magnet + est. 1 for ALS & ANS



Sector

Number of magnets Number of quenches

ALS ANS NOE Est. 1 Est. 2 Est. 3 Est. 4

1-2 49 96 9 2222 4141 4040 4949

2-3 56 60 38 2323 9797 9292 130130

3-4 56 65 33 2121 8787 8383 116116

4-5 46 46 62 2222 145145 136136 198198

5-6 28 42 84 2121 190190 178178 262262

6-7 57 36 61 2020 142142 133133 194194

7-8 54 40 60 1414 140140 132132 192192

8-1 64 24 66 1919 151151 142142 208208

Total 154 154 154 162162 993993 936936 13491349

Estimated dipole training to reach 7 TeVEstimated dipole training to reach 7 TeV

Est. 1: Based on 115 MB’s that have been submitted to a thermal cycle in SM-18 (2008 before HWC, P. Xydi and A. Siemko)

Est. 2: Extrapolation from sector 5-6 data + est. 1 for ALS & ANS



The number of “circuit quenches” might be a bit smaller as compared to the number of “dipole quenches” due to “parallel training” of 2 or more dipoles during the same circuit quench.


RB quench propagationRB quench propagationCryogenic recovery time: see talk Serge ClaudetCryogenic recovery time: see talk Serge Claudet

Almost instantaneous (within 1 sec)

Observed 11 times on 8 different magnets. Caused by:

- an almost equal quench current

- an unbalance > 100 mV (triggering the QPS) due to very different inter-strand

coupling currents between the 2 apertures,

- traveling voltage waves due to the opening of the dump switches,

- other noise.

See also the talk of Karl Hubert Mess.

Through thermal propagation (typical delay 30-300 s)

Observed for all quenches > 6 kA.

Statistics for quenches > 9 kA: propagation to 1 dipole: 3

propagation to 2 dipoles: 9





Example thermal quench propagation (I_q=10274 A)Example thermal quench propagation (I_q=10274 A)LBBLA.27R4: 0 s, 10274 A

LBBLC.27R4: 123.5 s, 2844 A (355.7 s, 238 A)

LBALA.27R4: 46.6 s, 6464 ALBALB.26R4: 109.1 s, 3330 A

LBBLA.26R4: 167.4 s, 1748 A


Thermal quench propagationThermal quench propagation

0

50

100

150

200

250

300

350

0 2000 4000 6000 8000 10000 12000

Circuit quench current [A]

Del

ay s

eco

nd

ary

qu

ench

[s]

Same half-cell, 15.7 m apart


Next half-cell, 22.1 m apart




Thermal quench propagationThermal quench propagation

0

1000

2000

3000

4000

5000

6000

7000

8000

0 2000 4000 6000 8000 10000 12000

Circuit quench current [A]

Sec

on

dar

y q

uen

ch c

urr

ent

[A]







Quench at 6.9 kA

Quench at 7.0 kA

Quench at 9.0 kA

B30R7 0 0 0

C30R7 61.3 s 58.8 s 40.2 s

A30R7 111.1 s 108.4 s 47.0 s

Reproducibility of thermal quench propagationReproducibility of thermal quench propagation

Quench at 10.7 kA

Quench at 10.9 kA

C15R5 0 0

B15R5 45.1 s 38.3 s

A15R5 81.5 s 72.1 s

C14R5 281 s 270 s

Quench starting in B30R7at 3 different currents

Quench starting in C15R5at 2 different currents

Due to secondary quenching, each quench causes the firing of the heaters of about 4 magnets.

High-field-High-field-heater firingheater firing

SM-18 at warm 1150

SM-18 at cold, 1500 A 1150

SM-18 at cold, training 4100

HWC at cold, 0 A 1700

HWC at cold, I > 0 A 240


Conclusion: Dipole trainingConclusion: Dipole training

6 sectors reached 9310 A (5.5 TeV) without a quench. 1 sector showed 1

quench<9310 A, and in 1 sector a busbar joint opened/melted before reaching

9310 A.

Training of S56 showed a completely unexpected behaviour of the NOE

magnets, triggering quenches well below the respective 1st training quenches

in SM-18.

All training quenches in S56 happened (for the moment) on different

magnets, except in one case when a NOE magnet showed a detraining step

(10910 A 10720 A).

Retest in SM-18 of one NOE magnet (that was not installed in the tunnel and

previously tested in Aug 2005) showed a similar time-relaxation effect.

The expected number of RB circuit quenches needed to reach 6, 6.5, and 7

TeV is about 10, 80, and 900 respectively. Note that 900 is a rough estimate

since it is based on a large extrapolation of the S56 training curve and the re-

test of only one NOE magnet. Assuming training in all 8 sectors in parallel,

with 3 quenches per day would then require about 60 days to reach 7 TeV, and

about the same number of heater firings as the entire SM-18 test campaign.


Conclusion: Quench propagationConclusion: Quench propagation

Thermal quench propagation time is strongly affected by current level, cell

geometry, cooling conditions, etc.

Large variations in quench propagation time are observed for different

positions in the sector. Reproducibility seems however good.

High field (>7 T) dipole quenches usually cause 2-4 adjacent magnets to

quench while ramping down the circuit. Due to long propagation times (typically

30-100 s to the nearest neighbour) the total dissipated energy in a half cell is

about twice the stored energy of a single MB.

Several cases have been observed where the MB-MB propagation time was less

than 1 s, due to unbalanced coupling currents, traveling waves, noise and similar

quench current level.

Date post:	14-Jan-2016
Category:	Documents
Upload:	tahlia
View:	42 times
Download:	0 times

Quench

Documents