AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

1

MTTR & Spare Policy for the LHC Injectors

Magnets for the PS Complex

T. Zickler

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

2

IntroductionScopeMagnet typesTypical problems and failuresInterventions in 2007Risk analysis – MethodRisk analysis – ResultsRisk analysis – DetailsConclusion and Future

Thanks to A. Newborough, D. Bodart, A. Hue for their contribution

Outline

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

3

PS- Complex: structure grown during 50 years

Linac2, Linac3, Booster, PS, LEIR, AD, Isolde, Experimental Areas, beam lines …

More than 1200 magnets, more than 200 different types

In the past: responsibility of machine superintendent No central data base

Documentation kept in ‘private’ archives

Spare components distributed all over CERN

2003: responsibility of all nc magnets successively transferred to AT-MEL

2007: start to set up nc magnet database

Unique naming system

Inventory of installed and spare magnets

Upload magnet characteristics

Gather and scan related documents (drawings, specifications, measurement reports...)

Link to layout database

Introduction

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

4

Scope: the LHC-injector

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

5

All types: Bending, combined function, corrector, dipole, multipole, octupole, quadrupole, sextupole, solenoid, water cooled, indirect water cooled, air cooled, iron-less, PFW, pulsed, continuous, etc…

Installed units Magnet types

Linac 2 + TL 52 14

Booster 253 15

Booster TL 70 16

PS 243 24

TT2 49 9

Linac 3 40 19

Ion Beam Lines 47 20

LEIR 44 7

Total 798 108

Magnet types

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

6

LHC injector (PS-Complex): 800 Magnets of 108 different types

Policy:

– Systematic refurbishment of PS main magnets (106 units)

– All other: keep sufficient spare magnets and spare components (coils)

Scope: the LHC-injector

0

5

10

15

20

25

30

35

40

45

50

installed units spare units

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

7

Typical problems and failures which occur on magnets due to aging, radiation and fatigue, which lead to repair interventions or magnet replacement:

– Water leaks in cooling circuits

– Electrical short circuits to ground

– Obstructed cooling ducts

– Degradation of coil shimming

– Broken cable insulation

– Inter-turn short circuits

Typical problems and failures

Pictures courtesy of A. Newborough and D. Bodart

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

8

Interventions in 2007

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

9

Risk analysis - Method

Availability of spares

1 1 new; fabrication < 10 y Complete spare units

2 2 old; fabrication > 10 y Spare coils

3 3 old; indication of degradation Spare yokes

1 1

2 2

3 3

4 4

Not significant

Low risk

Medium risk

High risk

3 ≤ S ≤ 6

7 ≤ S ≤ 10

10 ≤ S ≤ 16

Acceptance

Action required

catastrophic: failure to meet scientific objectives

Risk Score S = P x I

high; air cooled; low power;

medium; water cooled; medium power

low; high power; high voltage; high flow

Reliability (design)

X

Time to repair and/or replace magnet

Time to re-establish vacuum (0 < t <10 d)

Severity of Impacts [I]

insignificant: loss of phyiscs < 1 day

moderate: less of physics < 1 week

major: physics loss < 1 month

exceptional

rare

probable

frequent

X

S ≤ 2

State (age)

Probability of Failure [P]

←no

Impact on LHC Operation

yes

MTTR (Mean Time To Recover)

Cooling down time (0 < t < 2 d)

Based on CERN Risk Management System (EDMS No. 832542)

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

10

Risk analysis - Results

LINAC2 + TL (0/52)

Booster (109/253)

Booster TL (2/70)

PS (3/243)

TT2 (7/49)

LINAC3 (13/40)

Ion Beam Lines (15/47)

LEIR (26/44)

Total (175/798)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Insignificant

Small

Medium

High

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

11

Risk analysis: Booster

Medium RiskMultipoles 11 ONO/XNO/QSK (Type I) 4 ONO/XNO/QNO (Type II) 1 OSK/XSK/QNO (Type III B)4 ONO/OSK/XNO/XSK (Type A) 4 OSK/XSK/DVT/DHZ (Type B)

Mitigation: produce spare units of each type

Ressources needed: 400 kCHF, 0.8 FTE*y, 1.5y delay

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

12

Risk analysis: Booster Transfer Lines

Medium Risk: BI.DVT

Mitigation: new magnet foreseen for H- injection from Linac 4

Medium Risk: BT.BHZ10

Mitigation: find or produce spare coils

Ressources needed: 40 kCHF, 0.3 FTE*y, 1y delay

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

13

Risk analysis: Linac 3

Medium Risk Solenoid (2)

Bruker quads (4)

Spectrometer (2)

Doublet B (2)

Triplet (3)

Bending 106º (2)

Proposed Mitigation

Produce spare coils

Produce spare coils

Produce spare coils

Produce spare coils

Produce spare coils

Produce spare coils

Costs [kCHF]

35 25 40 25 40 50

Manpower[FTE*y]

0.2 0.2 0.2 0.2 0.2 0.3

Delay[months]

12 12 12 12 12 18

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

14

Risk analysis: LEIR

Ion pump 400 l/s

Ion pump 60 l/s

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

TiTi

Ti

Ti

Ti

TiTi

Ti

Ti

Ti

Ti Ti

Ti

TiTi

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

Ti

TiTiTiTi

Ti

Ti

Ti

Ti

TiTi

Ti

Ti

Ti Titanium sublimation pump

Pumping group (with RGA + gauges)

VV

S 2

1

VVS 11

VVS

VV

S

VVS Sector valve

VV

S 2

2

VVS 32

VV

S 4

1

Bayard-Alpert gauge

B

BB

B

B

B

B

B

B

B

B

B

B

B

B

B

BB

B

NEG’s (cartridge or coating)only shown for electron cooler

R Residual Gas Analyser

RB

Ti

R

B

B

R

R

R

R

R

Ti

BRB

B

AT-VAC, E. Page, 10.2.2004

LEIR layoutM. Chanel, E. Mahner

10.02.2004

1 m

Pictures courtesy of E. Page

Medium Risk: MC 100 (3)

Mitigation: find possible replacement magnets at CERN

Medium Risk: Main Bending (4)

Mitigation: produce spare coils

Ressources needed: 100 kCHF, 0.5 FTE*y, 1.5 y delay

Medium Risk: Main Quads (20)

Mitigation: produce spare coils

Ressources needed: 50 kCHF, 0.3 FTE*y, 1.5 y delay

Medium Risk: Skew Quads (2)

Mitigation: find spares at CERN

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

15

Situation OK for:

Linac 2, PS, TT2

Spare situation to be improved for:

Booster, Linac 3, Ion Beam TL, LEIR

Total required: 800 kCHF, 3 FTE*y, 2 years delay

Extend magnet inventory to other machines and beam lines

CTF3, AD, Isolde, East Hall EA, n-TOF, SPS, North Area

Complete data base

Upload all relevant documents

Link to layout data base to ease maintenance and traceability

Central storage for PS Complex magnets

Regroup all spare magnets and magnet components in 150

Conclusion and Future

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

16

Spare Slides

AT

C /

AB

OC

Day

s –

Ses

sion

4

21-2

3 Ja

nuar

y 20

08.

T

h. Z

ickl

er A

T-M

CS

-MN

C

17

Risk analysis – Results (types)

LINAC2 + TL (0/14)

Booster (5/14)

Booster TL (2/7)

PS (1/24)

TT2 (3/9)

LINAC3 (6/19)

Ion Beam Lines (7/20)

LEIR (3/7)

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Insignificant

Small

Medium

High