2nd Joint HiLumi LHC – LARP Annual MeetingINFN Frascati – November 14th to 16th 2012

LBNL: Helene Felice – Tiina Salmi – Ray Hafalia – Maxim MartchevskyFNAL: Guram Chlachidze

CERN: Ezio Todesco – Hugo Bajas

HQ Summary of quench protection studies

Work partially supported by the U. S. Department of Energy, under Contract No. DE-AC02-05CH11231

2nd Joint HiLumi LHC - LARP Annual meeting

2

MIITs curve and Time margin

2012/11/14

0

5

10

15

20

25

30

0 50 100 150 200 250 300

MIIT

s (10

^6*A

^2*s

)

Hotspot temperature (K)

T0 = 2 K

12 T, RRR = 6012 T, RRR = 1000 T, RRR = 1000 T, RRR = 60

• Well-known that material properties play an important role

• Example of HQ MIITS computation: NIST+cryocomp /cryocomp /quenchproAt 2K and 12 T RRR 60 = 20.6 / 21.6 / 19.5At 2K and 0 T RRR 60 = 25.2/ 26.3 / 23.5

Average: 2 K, 12 T and RRR 60 = 20.6 +/- 1 MIITs

Figure of merit: the time margin (E. Todesco)

(MIITS budget – MIITS decay)/I2 = time to quench

HQ time margin of the order of 27 ms

2nd Joint HiLumi LHC - LARP Annual meeting

3

HQ protection scheme

2012/11/14

Protection Heaters (PH)

• PH strip on both layers• 2 strips per layer• 4 strips per coil• Strip R: ~ 4.5 W at 4.2 K

• Powering scheme• Typical: 4 circuits

• all B02, all A02, all B01, all A01• PH hipot failure to coil can lead to replacement by a resistance• In some cases, enough power supplies to power a few strips individually

• Coverage of about 60 % (including propagation between station)

Dump resistor• 30 to 40 mW

Layer 2

Layer 1

A01

A02

B01

B02

2nd Joint HiLumi LHC - LARP Annual meeting

4

HQProtection studies overview

2012/11/14

• Several tests• Magnets: HQ01a to e• Mirror: HQM01 and HQM04

• Protection Heaters (PH) delay studies• versus Heater peak power• versus magnet margin• versus Kapton thickness

• Quench back • MIITS limits

2nd Joint HiLumi LHC - LARP Annual meeting

5

Protection Heater (PH)Delay time versus peak power

2012/11/14

0

10

20

30

40

50

60

70

80

90

100

40 60 80 100

Del

ay to

que

nch

onse

t (m

s)

Peak power (W/cm2)

Imag/ Iss tau = 24 mstau = 47 ms

29 %

29 %

42 %42 %

70 %

tau = 24 mstau = 47 ms

0

10

20

30

40

50

60

70

80

0 20 40 60 80

Tim

e de

lay

(ms)

Peak power (W/cm2)

HQ01d

30 % Iss, circ#1 (IL circ)

30 % Iss, circ#5 (OL strip)

60% Iss, circ#5 (OL strip)

HQ01d

HQM01

Tau = 31 ms

• Some saturation of the delay at high peak power

• Tau => total energy impacts the delay time• Strongly at low fraction of Iss

• Marginally at higher fraction of Iss • Some studies on optimization of energy deposition

2nd Joint HiLumi LHC - LARP Annual meeting

6

0

10

20

30

40

50

60

0 20 40 60 80 100

Del

ay to

que

nch

onse

t (m

s)

Imag/iss (%)

HQM04, 4.6 K

HQM01, 4.6 K

HQ01e, 4.4 K

Impact of Kapton thickness on PH delay

2012/11/14

75 mm

25 mm

50 mm

Kapton thickness

H. Bajas et al., “Test Results of the LARP HQ01 Nb3Sn quadrupole magnet at 1.9 K”, presented at ASC2012T. Salmi et al.: “Quench protection challenges in long Nb3Sn accelerator magnets”, AIP Conf. Proc. 1434, 656 (2012)

HQ01e at CERN: Pw0 = 50 W/cm2, tau 40 ms: Iss = 17.3 kA @ 4.4 K; 19.1 kA @ 1.9 KHQM04 at FNAL: Pw0 = 45 W/cm2, tau 46 ms: Iss = 16.2 kA @ 4.6 K; 18.2 kA @ 2.2 KHQM01 at FNAL: Pw0 = 47 W/cm2, tau 46 ms: Iss = 17.0 kA @ 4.6 K

• Large increase of PH delay with Kapton thickness

• From 60 to 150 % from HQ01e to HQM04

• 75 mm: choice made for HQ02to ensure electrical integrity

2nd Joint HiLumi LHC - LARP Annual meeting

7

0

20

40

60

80

20 40 60 80 100

PH

del

ay t

ime

(ms)

Imag/Iss (%)

OL - 4.4 K

IL - 4.4 K

Il versus OL

HQ01e at CERN – coil 9 tested(25 mm Kapton)

40 -49 W/cm2

55 W/cm2

HQM04 at FNAL (75 mm kapton)

• In HQ01e: OL PH delay shorter than IL • In HQM04: IL PH delay shorter than OL• Possible differences between HQ01e and HQM04

=> Cooling of the bore?=> high compression due to mechanical preload in HQ01e: better thermal contact?

2012/11/14

0

10

20

30

40

50

60

0.0 20.0 40.0 60.0 80.0 100.0

Del

ay to

que

nch

onse

t (m

s)

Imag/iss (%)

VPH = 250 V

45 W/cm^2 -OL - 4.6 K

49 W/cm^2 - IL - 4.6 K

2nd Joint HiLumi LHC - LARP Annual meeting

8

Bath Temperature

0

20

40

60

80

100

20 30 40 50 60 70 80 90

PH

del

ay t

ime

(ms)

Imag/Iss (%)

OL - 1.9 K OL - 4.4 KIL - 1.9 K IL - 4.4 K

2012/11/14

HQ01e at CERN

• From HQ01e and HQM04: No effect of the bath temperature observed on the PH delay

0

10

20

30

40

50

60

0.0 20.0 40.0 60.0 80.0 100.0

Del

ay to

que

nch

onse

t (m

s)Imag/iss (%)

VPH = 250 V - OL

HQM04: 45 W/cm^2 -4.6 K

HQM04: 45 W/cm^2 - 2.2 K

HQM01: 47 W/cm^2 - 4.6 K

HQ01e: 50 W/cm^2 - 4.4 K

HQ01e: 50 W/cm^2 - 1.9 K

HQ01e at CERN – HQM04 at FNAL

2nd Joint HiLumi LHC - LARP Annual meeting

9

Next steps on PH studies

2012/11/14

• Development of a 2D thermal model (Tiina Salmi, LBNL) • Simulation of heat transfer from PH to coil for design optimization

• Heater longitudinal layout• PH to coil insulation layout• PH powering

• Optimization of energy deposition: Square pulse, Truncated capacitance discharge

• Minimizing PH Temperature and Voltage• Comparison with experimental data showing good agreement

HQM04 simulationThermal system: half PH period

2nd Joint HiLumi LHC - LARP Annual meeting

10

• Magnet sitting at a constant current: from 5 to 13 kA – (NO quench)• Discharge in the 40 m W dump resistor without PH

• Does the magnet quench from eddy current generation in the cable (form of quench-back)?

• From the current decay:

• At 5 and 10 kA: no sign of quench• A 13 kA: signs of quench• At 15 kA: fraction of the magnet is quenching• Last 15 kA test with PH: no clear impact

HQ01e – Quench-Back

2012/11/14

𝑅𝑚𝑎𝑔 (𝑡 )=− 𝐿𝑑𝑑𝑡

ln ( 𝐼 (𝑡 )𝐼 0 )−𝑅𝑑𝑢𝑚𝑝

𝐼 (𝑡 )=𝐼 0𝑒−𝑅 𝑡𝐿

H. Bajas et al., “Test Results of the LARP HQ01 Nb3Sn quadrupole magnet at 1.9 K”, presented at ASC2012

HQ01e at CERN

2nd Joint HiLumi LHC - LARP Annual meeting

11

Edump= I(t) * Vdump(t) dt = 0.425 MJ10 ms

300 ms

2) Find total energy in the magnet:Magnet inductance is L=7.5 mH; Then, at I = 14819 A, Emag = LI2/2 = 0.823 MJ

4) Find magnet resistance: A = I2(t) dt = 7.13 106 =>

=> Rav = Ecryo/A = 0.055 W - average magnet resistance during extraction (10-300 ms)10 ms

300 ms

1) Find energy dissipated in the dump resistor

X

3) Find energy dissipated in the cryostat: Ecryo = 0.823-0.425 = 0.398 MJ

A36

Estimate of HQ01 resistance during extraction

2012/11/14

2nd Joint HiLumi LHC - LARP Annual meeting

12

HQ01e MIITS limit

2012/11/14

• Explored high MIITS by removing the dump resistor and inner layer PH (spontaneous quenches)

• HQ01e-3 at LBL will perform validation quench to check for degradation

H. Bajas et al., “Test Results of the LARP HQ01 Nb3Sn quadrupole magnet at 1.9 K”, presented at ASC2012

18.3 Miits

13.2 Miits 16.9 Miits

Time (s)

2nd Joint HiLumi LHC - LARP Annual meeting

13

PH and Trace revision

2012/11/14

• Avoid overlapping PH strip with metallic end parts: Rev C

• first implemented in HQ16

• Fit the “LHQ style” extension (Rev D): will be first implemented in HQ20

HQ02: - coil 15 OL: rev B - coil 17 OL: rev C -IL unchanged- coil 16 OL: rev C - coil 20 OL: rev D

HQ02 coils will have 75 microns kapton between PH strips and coil

2nd Joint HiLumi LHC - LARP Annual meeting

14

Summary

2012/11/14

• Protection heaters:• A wide mapping of the PH delay has been performed with the HQ01 series and

the mirrors• Good start to benchmark models

• From HQ01e: OL PH seem more efficient than IL PH• PH efficiency seems to be independent of temperature

• HQ01e test at CERN explored High MIITs regime• The dump resistor as well as the IL PH were deactivated

• HQ01e test at CERN exposed quench-back in the cable for current above 13 kA• HQ01e coils do not have a core

• Necessity to reproduce these tests with HQ02• cored cable• 75 microns Kapton between coil and PH strip