New LHC collimator installation schedule impact on 450 by veteranlives

VIEWS: 42 PAGES: 36

									New LHC collimator installation schedule & impact on 450 GeV & 7 TeV run

R. Assmann, AB

MAC

December 8th, 2006

RWA, 8/12/2006

1

Situation MAC June 06
• Situation was looking promising in June at the last MAC:
– First ring collimators delivered (transfer line collimators also). – First 2 ring collimators were conform and accepted for installation.

– First ring collimator installed.
– Production was expected to take off for installation of 70 collimators by Summer 2007.

•

Major break-down of production starting in July 06:
– July: New problem of brazing feed-throughs for a number of collimators. – August: First signs of oil pollution inside collimators. – September:
• Proof of serious oil pollution inside all collimators starting with #3. • Much of prepared material oil polluted from machining with oil (even silicon-oils) and crosspollution in non-adequate UHV cleaning. CERCA did not respect own quality standards at subcontractors. • Pollution of expensive vacuum equipment which could not be recovered.

– October:
• Rejection of 6 fully completed collimators due to oil pollution (4 already at CERN). RWA, 8/12/2006
2

Hardware Procurement...
Major contracts:
Collimators (FR)
Motors and sensors (D) Position sensors (FR/China) CFC (J) Supports (Bulgaria) Vacuum pumping ports (Italy) Material/screws (Switzerland) Glidcop material (US)
Motor drivers + supplies (Italy)

RWA, 8/12/2006

3

Ring Collimators Installed

RWA, 8/12/2006

4

Transfer Line Collimators Installed
QRL

TCDI TCDI

Dipole

TCDI

RWA, 8/12/2006

Injection line

Beam 2

Beam 1

5

CERCA Technical Quality Problem
• There is no fundamental design problem:
– No major problems in design and drawings. – CERCA has built several conform collimators.

– Major acceptance criteria are checked and fulfilled:
• Jaw flatness: 40 mm tolerance fulfilled over 1.2 m long jaws. Each checked. • Heat conductivity: Brazing qualified for heat conductivity (Jan 2006). • Vacuum spectra: Looked OK for first two ring collimators. Each checked. • Mechanical precision: 20 mm mechanical play, micron alignment. Each checked.

•

Problem are the delays due to very bad technical quality control at CERCA:
– Delay evident in March 2006  Top management meeting requested by us. – May 17th: CERCA DG at CERN. Commits on new schedule. Additional resources. A few technical issues resolved in weeks after meeting. – Agreement with CERCA: No fundamental technical issues but many problems in technical quality in CERCA production! – Severe quality problems proven by CERN in September (decay of quality when trying to increase production speed, for example machining with oil).

RWA, 8/12/2006

6

TCT jaw repair September 2006:

Ni coating badly scratched. Limited flow of brazing material.

RWA, 8/12/2006

7

TCT repair September 2006:
• Ni coating badly scratched. • Coloration on FT (SolidSealing type 2).

RWA, 8/12/2006

8

CERCA brazing test September 14, 2006: • Lack of brazing material at FT. • New FT„s from CERAMASEAL.

RWA, 8/12/2006

9

CERCA production brazing September 30, 2006: • Lack of brazing material at cooling pipes. Leak tight but not at CERN standards! Repair requested. • New FT„s from CERAMASEAL. Analysis with microscope needed?

RWA, 8/12/2006

10

Comparison Brazing Cycles CERN - CERCA
900

CERN: 4 May 2004
800

CERCA: 15 Sep 2006

700

600

Temperature [deg C]

500

CE RN: 200 mbar Argon not cooled

400 CE RCA:Over-pressure gas. Cooled and strong circulation! 300 AGAINST CE advice and verbal agreement! RN 200

100

0 0 2 4 6 8 10 12 14 16 18

Time [h]

RWA, 8/12/2006

11

Double Coating
The Ni coating for brazing was done twice at CERCA!?

Another example of an avoidable problem and lack of quality control!

RWA, 8/12/2006

12

TCS010 after bake-out (8 Sep 2006)
1.00E-07

Extreme Pollution (HydroCarbon) of UHV parts:
1.

Leak
9.00E-08

CERN procedure not followed: check spectrum at cold before start of bake-out to prevent major pollution. Explicit advice from CERN vacuum expert ignored on September 18th. Of last 3 delivered collimators 2 are polluted. CERN tries to recover vacuum. Failed.

2. 3.

8.00E-08

Partial Pressure [uncalibrated]

7.00E-08

6.00E-08

“Oil Pollution“ in Sep/Oct
5.00E-08 4.00E-08

3.00E-08

2.00E-08

1.00E-08

0.00E+00 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105

RWA, 8/12/2006

Mass

13

Commitment and Reality
12 Commitment 10 Delivered (TL) Delivered (ring)

Delivered Collimators

8 6 4 2 0 Jan-06
Last 4 ring collimators to be sent back to CERCA:

Vacuum too bad for installation into the LHC (oil pollution)!

Apr-06

Jul-06

Oct-06 Time

Jan-07

Apr-07

The schedule presented by H. Sztark (CERCA DG) at CERN on May 17th is completely obsolete! As a consequence our CERN collimation planning is!
RWA, 8/12/2006
14

CERN Actions
• Try to push on three different aspects to improve collimator production:
– Help for CERCA:
• • AB and TS review of critical production steps. TS review and specification of all production procedures. TS help in critical production steps (e.g. Ni coating of all feed-throughs).

–

Control of CERCA production:
• • • Existing AB production and quality control reinforced by additional resources and help from senior CERN experts. Senior TS expert (J.P. Bacher) now in charge at CERCA to sort out manufacturing process and quality control. Senior AT expert (N. Hilleret) got involved to “teach” CERCA the UHV procedures.

–

Pressure on CERCA management:
• • Top management meeting with CERCA on October 13th. CERCA top management accepted quality issues and reassured us that quality is THE highest priority for the AREVA company in France. Quality is now taken VERY serious!

•

Note: New role of TS department in collimation production:
– – TS has worked out the mechanical design and built very successfully 3 LHC prototype collimators in-house. Unexpected quality problems at CERCA: many hidden and only recognizable with TS resources.

–

Production experts (TS/MME) brought in for sorting out severe quality issues at CERCA, in collaboration with the existing AB team!

•

However, delays cannot be recovered: Plan for collimator installation must be changed.
15

RWA, 8/12/2006

Needs 450 GeV Run 2007
Collimators are an intensity-driven system: LHC can be started at low intensities without any collimators. However, once intensity or energy is increased a high number is quickly needed!
Intensity Primary collimation (TCP) and scraper (TCHS) Secondary collimation (TCS) Absorbers for collimation debris (TCLA) IR (exp) collimators (TCT, TCLP, TCLI) Dump protection (TCDQ +TCS) Total number (two beams)

< 5 × 1011 < 1 × 1013 < 3 × 1013 Full system
(ring only)

0 8 8 14

0 0 18 30

0 0 0 18

0 0 0 28

0 2 2 2

0 10 28 92

• 160 times less stored energy than at 7 TeV and nominal intensity! • Quench limit ~50 times higher than at 7 TeV! • Factor 8000 relaxed to nom 7 TeV case! RWA, 8/12/2006

TCT not required if triplet is not aperture bottle-neck (as foreseen)  b* ≥ 11m!

2007 run goals
16

Effect on Cleaning Efficiency
• General: No tungsten absorbers in IR3 and IR7. Showers lost in SC arc. Momentum cleaning: all primary and secondary collimators installed. Losses during start of ramp due to uncaptured beam. Will allow test ramp. Betatron cleaning: only partially installed.
– All primary collimators installed, but only 5 secondary collimators per beam for cleaning of horizontal, vertical and skew losses! Loss in effciency quite accetable. Most lost due to missing tungsten absorbers before the SC arc!

• Quench limit for 10% of intensity Fewer secondary coll. •

Missing absorbers

–

RWA, 8/12/2006

17

Effect on Machine Protection
• Fewer injection protection elements, including one transfer line:
– Main ring protection (TDI) against injection errors is installed. – Restrict injection to safe beam (10-20 bunches), to exclude damage in transfer line.

– Total number of bunches is then restricted to ~150 full intensity bunches, or a total intensity of 2e13 p (1.1e11 p per bunch).
– Fully compatible with maximum goal of 156 bunch operation with bunch intensity of 0.4e11 p per bunch.

•

Passive interception of losses:
– All primary collimators are installed to define the standard aperture bottle-necks at 450 GeV.
– Due to fewer secondary collimators more holes in phase space. However, also present with full scheme. Collimation is designed as cleaning system and not as system with full passive protection (design choice).

•
•

Dump protection: No effect, as fully installed.
Must rely on various channels of machine protection (as always).

RWA, 8/12/2006

18

New Plan for Collimation
• A reduced set of collimators is sufficient for the start-up in 2007:
– – – Leave already installed collimators in (2 in IR8), even if not needed. Install all infrastructure and collimator quick plug-in supports. Keep IR6 fully installed. Do not install tertiary collimators (IR1/2/5) and absorbers (IR 3/7). Do not install sub-set of secondary collimators in IR7. Do not install further transfer line collimators as part of minimal system.

• • •

No impact on 2007 parameters for 450 GeV run. LTC and MARIC approved. After minimization: Minimal collimation system requires 28 collimators. Plus already installed collimators gives 35 collimators to be installed in minimal system for LHC start-up:
– 7 already installed. 5 at CERN/CERCA completed but vacuum is polluted. – 28 to be produced.

•

Minimal system was matched with the installation planning.

RWA, 8/12/2006

19

Present Preliminary Production Plan
• • All still preliminary, as we are re-establishing production. January 9th,2007:
– – Delivery of 2 ring collimators. Installation in IR6 at beginning of February.

•

January 2007:
– – Delivery of another 4 ring collimators. Series production established.

• •

April 2007:
– Completion of in total 28 collimators (minimal system).

May-November 2007:
– – – Production of 71 missing phase 1 collimators. Required rate is ~10/collimators per month  challenging at CERN for final assembly on supports. Goal: Last phase 1 collimator is installation-ready by December 2007.

•

December 2007 – February 2008:
– Production of 19 spare collimators for phase 1
20

RWA, 8/12/2006

Summary Production Plan Minimal System (Preliminary)
Delivery date 09 Jan 2007 16 Jan 2007 Number of collimators (ring) 2 2 Installation readiness 08 Feb 2007 16 Feb 2007 Total 2 4 Destination 6L (1), 6R (1) 3R (2)

24 Jan 2007
06 Feb 2007 22 Feb 2007 21 Mar 2007 28 Mar 2007

2
4 4 1 3

28 Feb 2007
15 Mar 2007 03 Apr 2007 17 Apr 2007 01 May 2007

6
10 14 15 18

3R (2)
3R (1), 3L (3) 3L (2), 7R (2) 7R (1) 7R (3)

04 Apr 2007
12 Apr 2007 20 Apr 2007

4
4 4

15 May 2007
29 May 2007 12 Jun 2007

22
26 30

7R (2), 7L (2)
7L (4) 7L (2)

Note: Significant work at CERN must be performed for delivered collimator: reception tests, mounting of motors and sensors, mounting on collimator support with correct orientation (upper quick-plugin ), final bake-out, internal transport. RWA, 8/12/2006
21

Installation
• Preliminary dates for installation, as agreed with installation (Dec 7th):
LSS Number of installed collimators Ready for installation (last collimator) Final deadline (start bakeout)

6L 6R 3R 3L 7R 7L

1 1 5 5 8 8

08/02/07 08/02/07 15/03/07 03/04/07 15/05/07 12/06/07

19/02/07 19/02/07 14/04/07 20/04/07 13/06/07 25/06/07

1L
5L 5R 2R 2L 1R
RWA, 8/12/2006

0
0 0 0 3? 3?

x
x x x x x

30/10/06
27/11/06 23/04/07 21/05/07 09/07/07 06/08/07
22

Impacts for LSS Installation
• • All standard collimator supports (110) and external connections (quick plug-ins) installed and aligned as foreseen (dummy collimator for alignment). No impact. All infrastructure (cables, cooling, …) installed and connected as foreseen. No impact.

•

Replacement chambers (108) in place of collimators.
Production delays with mobile pumping ports (AT/VAC) shall not affect the installation. However, if required it can be treated with a longer replacement chamber + flexible bellow (total length 2.52m)  AT/VAC takes responsibility for all missing replacement chambers, allowing flexible optimization with availability of pumping ports (minimizing number of replacement chambers).

•

Summary tables for collimation-related vacuum and support installations see below. Detailed lists defined.
Vacuum installation Installed ring collimators (no transfer line) Standard replacement chambers stainless steel Standard replacement chambers copper (coated) Special replacement chambers Collimator DB placeholders (phase 4) Total N 30 52 Support installation Standard quick-plugin supports (low version) Standard base supports (low version) Standard quick-plugin supports (high version) Special supports No supports (phase 4) Total N 66 30

30
26 22 160

14
28 22 160

RWA, 8/12/2006

23

Beyond Minimal System
• At the end of the production for the minimal system:
– 28 ring collimators (minimal) – 2 ring collimators (installed before crisis) – 6 transfer line collimators (already produced)
After May 2007 First shutdown Upgrade

Installation
LSS planning

Number collimators

Performance reach (ideal) ~0.07 × Inom at 450 GEV Fill sectors 0.4 × Inom tbd

30 + 6 ? ≤ 70* 32

•

Missing are 70 collimators for the complete phase 1 system. Use production after May 2007

of 4 TCLP in IR1 and IR5 for high luminosity not defined: Conflict with Roman Pots!

*Installation

•

to fill vacuum sectors for as long as possible.

RWA, 8/12/2006

24

Impact on 7 TeV Run
• • The 7 TeV run will take place in 2008. Stored energy might already be demanding:
– For 10% of nominal intensity we have 20 times more stored energy than Tevatron and HERA.

–

The ideal performance reach of the phase 1 collimation system is predicted to be 40% of design intensity (clearly big uncertainties in this number).

• • •

Baseline: Install all missing phase 1 collimators during the first shutdown (2007/8)! In this case: No limitation for the 7 TeV run predicted up to full phase 1 performance. If we must compromise on this goal then we have possible important consequences:
– – – Limitations on intensity due to limited cleaning efficiency! A few % of nominal. Limitations on b* due to missing triplet protection. Triplets in the shadow of the arc for b* of 6 m. More difficult collimator installation with already activated cleaning insertions.

• •

Consequences can be minimized with another minimal system for 7 TeV. Reach depends on the number of available collimators! Too early to speculate on this scenario. We need experience on production progress.
25

RWA, 8/12/2006

Baseline Installation in Shutdown
• Up to 70 instead of 30 collimators to be installed during first shutdown:
– Transport: ~ 4 collimators per day. – Installation/connection of collimator: 0.5 h per collimator. – Installation of pumping ports (if missing): 0.7 h per collimator. – Hardware commissioning (HWC) in tunnel: 3 h per collimator. – Alignment: Not required but at least point checks.

– Total time: ~ 4 weeks (installation), ~ 5 weeks (HWC, 1 team).
– Total time is about 5 weeks, as collimator hardware commissioning will be done in parallel to ongoing installation. Compatibility with other activities…

• Additional important work load for vacuum group (bake-out).
– Discussions with Miguel Jimenez.
– First preliminary analysis performed.
RWA, 8/12/2006
26

Time Estimate for Shutdown Installation (Preliminary)
Bake-out scenario Full Reduced: Limited bakeout in IR3 and IR7 Length [m] 1720.2 905.0 Nsector 32 18

Optimal: Reduced bakeout in IR3/IR7 and filling
additional vacuum sectors.

570.6

12

•

Optimal scenario about 4 weeks installation and 5 weeks bake-out.

•
•

Reduced scenario about 4 weeks installation and 8 weeks bake-out.
Bake-out can start after quick plug-in installation of collimators in first sector (1 week). Afterwards parallel work. Total time required: 9 weeks (reduced scenario)

• •

Compatible with foreseen length of first shut-down (3 months).

RWA, 8/12/2006

27

Other Work: Top Level Controls
S. Redaelli et al

Successful test of LHC collimator control architecture with SPS beam (low, middle, top level) RWA, 8/12/2006
28

90 80 70 60 90 50 80 40 70 30 60 20 50 10 40 0 30
20 10 0 0
Time [ms]

Other Work: Middle and Low Level (PXI) Controls Speed
Movement times (0 - 40um)
M. Sobczak et al

Movement times (0 - 40um)

Tmin [ms] Tmax [ms] Tavg [ms]
Movement times (full range)
18000 Tmin [ms] 16000 Tmax [ms]

Time [ms]

Tavg [ms]
12000

14000

0

0.01

0.02

0.03

0.04

0.05

Time [ms]

10000 8000 6000 4000 2000

Delta [mm]

0.01

0.02

0.03

0.04

0.05

0 0 5 10 15 20 25 30 35 Delta [mm]

Delta [mm]

Collimator response time (request – notification of completed movement):
RWA, 8/12/2006

10-80 ms!
29

Other Work: LVDT Position Sensors
< 20 µm
1.01 1.005 1 0.995

LVDT Calibration Repeatability test (TT40) 36 repetitions

Normalized position [mm]

0.99 0.985 0.98 0.975 0.97 0.965 0.96 0.955 0.95 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 distance [mm]

~ 25 µm mechanical play

•

Measured during test in TT40 (Oct. 31st) in remote!!!!

R. Losito et al

RWA, 8/12/2006

30

Other Work: Temperature Response
TT40 collimator temperature probes
65.0

60.0

55.0

Spikes correspond to 2 MJ beam shock impact: Possibility to detect accidental beam impact!

Temperature [°C]

50.0

Temp Upstream [°C] Temp Downstream [°C] Temp Water Cooling [°C] Temp Downstream [°C] Temp Water Cooling [°C] Temp Upstream [°C]

45.0

40.0

35.0

30.0

25.0

20.0 0:00

2:24

4:48

7:12

9:36

12:00

14:24

16:48

Time Elapsed [h:min]
P. Gandner et al

RWA, 8/12/2006

31

Conclusion
• Collimator production is on the critical path and we had a number of bad surprises recently. Problem is technical production quality and resulting delays! Several conform collimators were built by the CERCA company and have been installed into the LHC tunnel!
Deteriorating quality as production ramps up!

• • • • • •

CERCA production is accumulating delays which cannot be recovered. Minimal collimation system with 28 collimators for 2007 (plus 7 installed)! Requires up to ~ 6 collimators per month produced instead of 12 per month (CERCA production goal)! All standard infrastructure will be installed: 110 standard collimator supports, including electrical and water connections and alignment. No impact on beam operation during the first 450 GeV run in 2007. Minor impact on LSS installation (replacement chambers). Impact on work during first shut-down. Collimators can be installed in 4 weeks but long time required for bake-out (AT/VAC)! Optimized bake-out scenarios compatible with length of first shutdown. Partial bake-out and installation of some collimators beyond minimal system after February 2007. Potentially important impact on 2008 run at 7 TeV, if new production goals are missed!
32

•

RWA, 8/12/2006

Other Work: Loss Tails with Echo

12 s

C. Bracco, T. Weiler et al

Beam tails with long decay times (several 10 s) already shown in 2004! This time observation of “echo” in beam loss tails…
33

RWA, 8/12/2006

Other Work: Shock Wave Measurements on Collimator Jaw
• Test of collimator robustness with shock beam impact. • About 2 MJ in 1mm2 spot. • Set-up of laser vibrometer. • Measurement of shock waves (velocity) versus time.

H. Richter, R. Wilfinger et al 34

RWA, 8/12/2006

Other Work: New Impedance Measurement
Improved controls in 2006:

• Possibility of automatic scan in collimator position.
• Much more accurate and complete data set in 2006 than in 2004!

R. Steinhagen et al

RWA, 8/12/2006

35

Other Work: Beam Loss Tails

Qx=0.125 Qy=0.21

Qx=0.18

Qx=0.20

Qx close to 0.3

Qx=0.125 Qy=0.21

Qx=0.24

Qx=0.3

1

2

Measurement for different tunes and coupling values!BEAM DUMP!!
C. Bracco et al

RWA, 8/12/2006

36


								
To top