Damage Levels - Comparison of Experiment and Simulation V. Kain AB/Co

1/21/2005 Verena Kain, AB-CO 1

Damage Levels - Comparison of Experiment and Simulation

V. Kain AB/CoContents …

• Introduction• Assumed damage levels – TT40 accident• Controlled damage test• FLUKA results• Comparison with experiment• Conclusion

Scope …• Comparison of predicted damage limits with experiment• Damage limits for the LHC will not be defined

1/21/2005 Verena Kain, AB-CO 2

Introduction

10 cm~25cm long hole in chamberof QTRF in TT40. BothChamber and magnet had tobe exchanged.

Inside, damage visible over ~1m (melted steel)

• design of machine protection procedures• design of protection elements (robustness,…)• settings of protection systems• thresholds of monitoring systems (BLMs,…)• “safe” beam condition• design of operational procedures• …

The damage limit of equipment has to be known for …

• especially for LHC energies and intensities• mostly static energy deposition calculated (FLUKA,…) • dynamic effects (shock waves,…)?• simulation vs. experiment?

•simulation vs. experiment!

Knowledge on damage levels based on simulations…

1/21/2005 Verena Kain, AB-CO 3

Assumed damage levels • 450 GeV

• ~ 2x1012 protons ≈ 5% of full ultimate batch

• 7 TeV• ~1010 protons

Damage limit at 450 GeV:1 full nominal batch » damage limit

10 cm~25cm long hole in chamber

25th of October: MSE trip during high intensity extraction. Damage of QTRF pipe and magnet.

1/21/2005 Verena Kain, AB-CO 4

Reconstitution of Scenario and Comparison with Simulation

• Analysis of logging data:– MSE tripped due to EMC of

LHC beam 11ms before extraction

– in 11ms field changes by 5%– number of extracted protons:

3.4x1013@450GeV

Reconstructed trajectory according to screenshots.

Input for FLUKA simulation (x, x’).

max. Temp. 1350°CThe energy deposition result is verysensitive to accurate input parameters.Difficult to reproduce the observed damage.

The melting point of 314L: 1400°C.Slit due to heat + stress, rather than melting?

stainless steel: 314L

Meaningful comparison with simulation: CONTROLLED EXPERIMENT

1/21/2005 Verena Kain, AB-CO 5

Controlled Damage Test

• Low-tech target, no extra instrumentation (no temperature sensors,…)

• Simple target geometry:– Stack of high-Z metal

plates

• 4 predefined 450GeV- beam intensities (A, B, C, D)– Intensities chosen to see

certain effect on plate: melting/not melting

30 cm

108 plates

6 cm

6 cm

Double Confinement

Screen

Motor

1/21/2005 Verena Kain, AB-CO 6

Target Description• Materials:

– Zn, Cu, Stainless Steel (316L, INCONEL)

• Special order of materials:– packages of 3 materials– Zn, Cu, 316L (INCONEL),

Zn, Cu, 316L (INCONEL),…

• plates 6cm x 6cm x 2mm

• place-holders between plates (0.5mm) → avoiding molten plates sticking together

FLUKA model of target geometry

Every plate has unique longitudinal position:→ number on each plate→ 3D experiment: longitudinal position + damaged area on plate

1/21/2005 Verena Kain, AB-CO 7

Intensity protons [1012]

Melting starting in plate (nr.)

Zn Cu 316L INCONEL

A 1.3 - - - -

B 2.6 17 - - -

C 5.3 9 18 - -

D 7.9 6 12 23 -

FLUKAResults…

CuHeat of fusionnot taken into account…

Results taking heat offusion into account…

plate 11

plate 15

¼ of a nominal batch

Predicted observations…

1/21/2005 Verena Kain, AB-CO 8

Damage test in TT40TED – TT40

Ti entrance window

• Target was installed in TT40 in front of TED in air.

• Target irradiation: 8th of November.• 4 shots: intensities A,

B, C, D

Screen

• Opened box: 5th of January → no full analysis yet! • Took pictures of some plates (will take pictures of

ALL plates)

1/21/2005 Verena Kain, AB-CO 9

Results …

Observations when we opened the box…– outer confinement: no damage– first plates no damage as expected– after 6 packages more and more

damage– no stress related damage (buckling,

twisting, cracks, …), mostly very clean results…

1/21/2005 Verena Kain, AB-CO 10

Results…

Only in this part shows melting …

beam direction

1/21/2005 Verena Kain, AB-CO 11

Results…

A B D C☼ ☼ ☼ ☼

Plate 1, Zn

Impact locationsfor the differentintensities…

1/21/2005 Verena Kain, AB-CO 12

Results…Plate 7, Zn

Marks of meltingfor intensity D…

Prediction: melting @ D

A B D C

1/21/2005 Verena Kain, AB-CO 13

Results…Plate 8, Zn

Marks of meltingfor intensity D only…

Prediction:melting @ D

A B D C

1/21/2005 Verena Kain, AB-CO 14

Results…Plate 9, Zn

Marks of meltingfor intensity D & C

Prediction: melting @ D & C

A B D C

1/21/2005 Verena Kain, AB-CO 15

Results…Plate 20, Zn

Meltingfor intensity D & C & B

A B D C

1/21/2005 Verena Kain, AB-CO 16

Results…Plate 10, Cu

Marks of heatingfor intensity D

Prediction:no melting

A B D C

1/21/2005 Verena Kain, AB-CO 17

Results…Plate 12, Cu

Marks of meltingfor intensity D

Prediction: melting @ D

A B D C

1/21/2005 Verena Kain, AB-CO 18

Results…Plate 17, Cu

Marks of meltingfor intensity D & C

Prediction: melting @ D(plate nr. 18:melting @ D & C)

A B D C

1/21/2005 Verena Kain, AB-CO 19

Results…Plate 36, Cu

Meltingfor intensity D & C

Marks of heatingfor intensity B

A B D C

1/21/2005 Verena Kain, AB-CO 20

Results…

None of the stainless steel plates shows holes.

•INCONEL: as predicted

•316L: predicted melting from plate 23… not observed.

Needs further investigation…

1/21/2005 Verena Kain, AB-CO 21

Conclusions• First examinations show good agreement with

simulations– Zn, Cu, INCONEL as predicted – 316L still unclear

• Melting as predicted by FLUKA– transverse size of damage still to be checked

• Gives confidence that damage limits could be adequately simulated…

• geometry has large effect• sensitive to details of beam impact • full modelling on case-by-case basis is essential • no simple scaling

• Establishing generic damage limits seems to be difficult…

1/21/2005 Verena Kain, AB-CO 22

Thanks to …

K. Vorderwinkler, J. Ramillon, F. Loprete, G. Ferioli, R. Schmidt, B.Goddard, J. Lettry, F. Decorvet, J. Vo Duy, R. Harrison, W. Weterings, S. Sgobba, S. Calatroni, A. Dorsival, H. Vincke, M. Mueller, D. Forkel-Wirth, A. Desirelli