Status of Collimation System

1
Status of Collimation System Commissioning Plans

• R. Assmann, CERN/AB
• 15/07/2008
• for the Collimation Project
• LHCCWG

Slides, data and input provided by O. Aberle, A.
Bertarelli, C. Bracco, F. Caspers, J. Coupard,
A. Dallocchio, W. Hoefle, Y. Kadi, L. Lari, R.
Losito, A. Masi, E. Metral, R. Perret, S.
Perrolaz, V. Previtali, S. Redaelli, T. Weiler,
AB/BDI (R. Jones et al),
2
1) Production and Installation
All collimator locations under vacuum. Good end
of a long race
3
Consequence of Reduced Installation

• Not all collimators were ready for installation
  by March 2008. Some collimators had to be
  delayed.
• Most important impact Two-beam collimators in
  IR2 and IR8 (TCTVB)
• Ready collimators (CERN production) showed vacuum
  problem ? need to reopen, disassemble, clean and
  reassemble. Not ready in time.
• No vertical protection and cleaning of the IR2
  and IR8 triplets against incoming beam.
• Consequence for operation No squeeze of IR2 and
  IR8 in 2008 triplets then in shadow of the arc
  aperture (bgt6m).
• Accepted by ICC. Presented to LHCCWG and LPC.
• Phase 1 collimator installation to be completed
  in first shutdown.

4
Radiation-Optimized IR7
5
Secondary IR7 Collimator
6
Three Primary Collimators
7
Special Device
8
IR1 Tertiary Collimation
ATLAS
Very efficient but delicate protection W jaws!
9
Jaw Flatness (Ring TL)
Total 148 jaws
Flatness better than many feared. Out of
tolerance collimators were placed in locations
with more relaxed tolerances, meaning larger beta
(limited sorting). Enough collimators for
tightest places (40 mm).
10
Minimum Collimation Gap (Ring)
Total 32 TCSG, 30 TCT
TCSG(fiber-reinforced graphite)
TCT (tungsten)
High precision collimators produced adequate for
LHC conditions! Note No time to discuss here
production problems with a few CERN
collimators. Important Readiness for 2008 run
and parameters is ensured.
11
2) Hardware Commissioning and Residual Problems

• Collimator beam commissioning is only efficient
  if a thorough and successful hardware
  commissioning is done before (sort out as many
  issues without beam as possible). Therefore
  summarized here
• Hardware commissioning procedure published and
  approved.
• MTF system is set up.
• Collimator HWC comes after vacuum commissioning
  (bakeout). Request from vacuum group to perform
  bakeout without water in collimator.
• 95 of collimators have completed hardware
  commissioning. Now working on getting all data
  into MTF.
• Fully integrated into LHC schedule. Frequent
  shifts and adjustments to access conditions
  (cooldown has priority)
• Present scheduled end of collimator HWC July
  21st.

12
HWC Procedure MTF
13
New Issue Lifetime of Rail System
Two rails per axis. Pressed together to support
jaw weight (pre-load). Without grease. Outside
of vacuum.
14
New Issue Lifetime of Rail System
Cage creeps too fast in collimators from early
series production. Random effect. Not seen in
tests for prototype collimator (32,000
cycles). Stop not in all collimators ? cage can
get damaged. Seen in lab during controls tests.
Cage velocity v/2
Connected to movable jaw
Rail velocity v
Rail fixed
creeping
Rollers
15
Example Measurement on Series Collimator
Reached gt22,000 cycles without damage.
Similar but faster movement observed with Inox
cages (vertical collimator). Difference
Aluminium stopped by end pin without damage
(stiffer cage).
RWA 21/5/08
16
Collimator Creeping
Collimator Cage type Orientation Creeping expected mm/km Maximum creeping observed mm/km Creeping observed mm/cycle
TCS Inox Vertical 5 278 16.7
TCS Inox Horizontal 5 20 1.2
TCS Aluminum Vertical 5 53 3.2
TCT Aluminum Vertical 5 10 0.6
1 cycle 60 mm
Statistical behavior and reason for increased
creeping not understood. Studies ongoing in CERN
mechanical engineering group (TS/MME). Creeping
not an issue for collimator lifetime if cage is
adequately stopped (demonstrated 22,000 cycles,
all but 28 first collimators equipped following
production issues).
17
Anti-creeping tool ACT used to retrofit sliding
tables of collimators and to improve lifetime
creeping problem of inox cages (28 collimators
from early series production)
Inox cage in sliding table clamped by
retrofitted ACT
Proper functioning confirmed with 50 cycles after
installation.
18
Tunnel installation of ACT by AB/TS teams
Completed 4 collimators Time required 6h per
collimator Number of teams 3 teams (2 TS, 1
AB)Total time for 28 collimators 2 weeks
19
Roller Cage Summary

• No problems in 2008 run to be expected based on
  lifetime data available presently. Several
  collimators now used for cycling tests
  (originally restricted to 50 cycles to keep
  lifetime) and reached gt 20,000 cycles.
• 10 collimators (early production) with original
  lifetime of about 1-2 nominal years. With ACT 3
  times longer. Still want to fix this at some
  point (1st shutdown).
• 15 collimators (early production) with lifetime
  of about 10 nominal years. With ACT 20 years,
  based on tests of 4 cages (to be continued).
• All other collimators conform with specified
  lifetime of gt 20 years (gt 22,000 cycles).
• We are very lucky that we found this problem in
  the lab and not in the tunnel with beam.
  Retrofitting hook on all inox collimators
  (significant improvement).
• Anyway good idea to adjust operational procedures
  to maximize collimator lifetime minimum number
  of cycles, no cycling for fun,
• Operational impact Ensure reasonable usage of
  collimators.

20
3) Controls Readiness and Remote Commissioning

• Collimators are fully operational after HWC
  (single collimator). Lifetime problem not an
  issue for the first year and will anyway be fixed
  at latest in the shutdown.
• Low, middle and top level software is operational
  with most required features (see Stefano).
• Allows to generate movement functions versus time
  for any collimator
• Specify setting in number of sigma.
• From database information (later including
  beam-based calibration) generate settings in mm
  (take into account local beta, emittance, local
  orbit, calibration data).
• As the first collimators have become available,
  functionality has been remotely commissioned for
  ensembles of collimators. Many single collimators
  must work as a coherent system!

21
For Fun
Example Remote, function-driven movement of
collimator jaw (10 min). Remote reading of jaw
position in tank reference system with
LVDTs. What does this read???
? Stefano
Initially in 2008 Control 88 collimators for the
two beams ( 400 DOF)!
22
First 7 Collimators in IR3 Remotely Executing 30
min Ramp Function
Gap
Position
Realistic LHC ramp functions generated for 7
collimators in IR3. Executed synchronously
(software trigger used timing signal not
available). Different absolute settings due to
local beta function and different types of
collimators (families). Automatically generated
at top level application. Jaw and gap positions
surveyed independently of motor drivers with 10
sensors per collimator.
23
Jaw Position Error Requested - Measured
7 IR3 collimators 14 jaws with 28
sensors. Remotely controlled from CERN control
room.
Example interlock level
width of human hair
Demonstrated (1) Accurate jaw positioning. (2)
Synchronous movement. (3) Precise position
readout and survey (collimator and machine
protection).
24
Collimator Interlocks

• Two checks agreed (10 position sensors/collimator)
  
• Measured position and gap at a given time checked
  versus allowed position versus time (already
  implemented into low level controls).
• Measured gap at a given energy (beta) versus
  allowed gap for this energy (beta). Keep this as
  simple as possible.
• Energy and beta function from timing signal.
• Work ongoing for implementation by end of June.
• Working on MP check procedure for commissioning
  this (run special automatic cycle to trigger
  interlocks).
• Note temperature, state-driven and BLM interlocks
  in and close to collimators have also been
  defined.

25
Position Interlock Test TCDI2905
Example showing that interlock logic works (low
level controls). Once interlock (dump limit) is
reached jaw movement is stopped. At the same time
interlock generated ? any beam would be
dumped. Can decide later to relax (e.g. just
stop movement but do not dump).
? Stefano
26
Position Interlock Check Procedure(automatic
verification of position interlocks)
Interlocks generated
27
Collimator Temperature Interlocks

• Jaw measurement
• Warning 45 ºC
• Interlock 50 ºC
• Measurement of cooling water temperature
• Warning 30 ºC
• Interlock 35 ºC
• Thresholds are for TCP, TCS (C jaws) but put the
  same everywhere for the moment.

28
BLM-based Interlocks at Collimators

• In cleaning insertions protect collimators and
  downstream equipment against damage from
  excessive beam loss.
• Idea Dump the beam when measured beam loss is
  excessive.
• Specified from AP side. Under conversion into
  energy deposition and BLM thresholds.
• Defined on April 28. Work ongoing.

29
BLMs
30
Table 1 Estimated settings of damage interlock
limits for various collimator types in the
cleaning insertions. Power refers to nominal
intensity.
Device Location Energy Condition 1 Condition 2 Condition 3
TCP IR3 450 GeV dN/dt gt 1.2e12 p/s for T gt 10 s (87 kW) dN/dt gt 6e12 p/s for 1 s lt T lt 10 s (430 kW) dN/dt gt 1.5e13 p/s for T lt 1 s (1.1 MW)
TCP IR7 450 GeV dN/dt gt 1.2e12 p/s for T gt 10 s (87 kW) dN/dt gt 6e12 p/s for T lt 10 s (430 kW)
TCP IR3, IR7 7 TeV dN/dt gt 0.8e11 p/s for T gt 10 s (90 kW) dN/dt gt 4e11 p/s for T lt 10 s (449 kW)
TCSG IR3 450 GeV dN/dt gt 1.2e11 p/s for T gt 10 s (9 kW) dN/dt gt 6e11 p/s for 1 s lt T lt 10 s (43 kW) dN/dt gt 1.5e12 p/s for T lt 1 s (110 kW)
TCSG IR7 450 GeV dN/dt gt 1.2e11 p/s for T gt 10 s (9 kW) dN/dt gt 6e11 p/s for T lt 10 s (43 kW)
TCSG IR3, IR7 7 TeV dN/dt gt 0.8e10 p/s for T gt 10 s (9 kW) dN/dt gt 4e10 p/s for T lt 10 s (45 kW)
TCLA IR3 450 GeV dN/dt gt 6e8 p/s for T gt 10 s (45 W) dN/dt gt 3e9 p/s for 1 s lt T lt 10 s (215 W) dN/dt gt 7.5e9 p/s for T lt 1 s (550 W)
TCLA IR7 450 GeV dN/dt gt 6e8 p/s for T gt 10 s (45 W) dN/dt gt 3e9 p/s for T lt 10 s (215 W)
TCLA IR3, IR7 7 TeV dN/dt gt 4e7 p/s for T gt 10 s (45 W) dN/dt gt 2e8 p/s for T lt 10 s (225 W)
31
Table 2 Estimated settings of damage interlock
limits for various collimator types outside of
cleaning insertions. Power refers to nominal
intensity.
Device Location Energy Condition 1 Condition 2 Condition 3
TCTH, TCTVA, TCTVB IR1, IR2, IR5, IR8 450 GeV dN/dt gt 6e8 p/s for T gt 10 s (45 W) dN/dt gt 3e9 p/s for T lt 10 s (215 W)
TCTH, TCTVA, TCTVB IR1, IR2, IR5, IR8 7 TeV dN/dt gt 4e7 p/s for T gt 10 s (45 W) dN/dt gt 2e8 p/s for T lt 10 s (225 W)
TCL, TCLP IR1, IR5 450 GeV dN/dt gt 6e9 p/s for T gt 10 s (450 W) dN/dt gt 3e10 p/s for T lt 10 s (2.2 kW)
TCL, TCLP IR1, IR5 7 TeV dN/dt gt 4e8 p/s for T gt 10 s (450 W) dN/dt gt 2e9 p/s for T lt 10 s (2.2 kW)
TCLIA, TCLIB, TCSG IR2, IR6, IR8 450 GeV dN/dt gt 1.2e11 p/s for T gt 10 s (9 kW) dN/dt gt 6e11 p/s for T lt 10 s (43 kW)
TCLIA, TCLIB, TCSG IR2, IR6, IR8 7 TeV dN/dt gt 0.8e10 p/s for T gt 10 s (9 kW) dN/dt gt 4e10 p/s for T lt 10 s (45 kW)
32
4) Plan for Beam Commissioning

• By start of beam operation
• Collimators fully operational.
• Controls system ready for driving collimators
  following any function.
• Controls system ready for setting any warning and
  interlock levels safely (MCS).
• MP functionality fully working by commissioning
  jaw position interlocks already without beam.
• Collimator safety (closest devices to beam)
  ensured by temperature interlocks and interlocks
  on BLM rates next to collimators.
• Other systems operational (BIC, MCS, BLM, Timing
  incl. E signal), especially relative reading of
  BLMs close to collimators. BLMs already
  working.
• Beam commissioning plan then means
• Decide what collimators to use when and at what
  settings.
• Beam-based calibration of collimators (input to
  functions).

33
Reminder Beam-Based Calibration of Collimators

• Not reported in detail since reported many times
  before
• Method has been worked out (based on
  Tevatron/RHIC/HERA experience).
• Method has been tested and realistic accuracy
  established with the LHC prototype collimator in
  the SPS (see past reports).
• Machine conditions for calibration defined (up to
  few nominal bunches at top energy).
• Full set-up is lengthy 6 shifts of about 8
  hours. Faster if only updating, checking
  calibration.
• Agreed goal is to have automatic beam-based
  calibration (as TEVATRON) available for 2009 run
• Hardware connections prepared (collimators
  BLMs).
• Wok package in AB/CO group. Further SPS tests
  this year.
• No more details here this time to avoid
  repetition Note big expected impact from
  collimation upgrade in phase 2 (much better)!

34
Collimator Operational Modes

• For simplification, several modes defined
• Mode 1 Primary collimators and protection
  collimators only.
• Mode 2 Primary collimators, absorbers and
  protection collimators.
• Mode 3 Full 2008 system (5 instead of 11
  secondary collimators in IR7 per beam).
• Mode 4 Full phase 1 system, as installed for
  2009.
• Settings, performance reach and tolerances
  defined for each mode.
• Realistic assumptions for efficiency
• Factor 2-3 margin for BLM thresholds and
  uncertainties in FLUKA.
• Factor 10 reduction in cleaning efficiency with
  realistic imperfections (shown in PhD thesis of
  Chiara Bracco).
• Master table for collimator commissioning,
  defining settings for various operational stages.
  A few examples in next slides

? C. Bracco should report after end of her PhD
thesis (Sep)
35
Commissioning Table Settings
Here listed for nominal settings (b0.55m)
tightest gaps and tolerances. Settings in
normalized sigmas for every collimator
family. Converted into mm with the tools
described before.
? Chiara
36
Commissioning Table Performance
Here listed for nominal settings tightest gaps
and tolerances
? Chiara
37
Comm. Table Optimized Ramp Settings
Just one out of 3 fully calculated ramp cases
shown here. Used also for b. For each case
settings, performance reach and tolerances
calculated (not fully shown). Our basis for
placing LHC collimators. Too many numbers ? some
figures for visualization
? Chiara
38
Collimators Closed with Relaxed Tolerances During
Energy Ramp
with imperfections
Transient b beat 10 40 Orbit 0.3 s 1.2
s Collimator 0.4 s 1.6 s
39
Collimators Closed with Tight Tolerances During
Ramp (Nominal)
with imperfections
Transient b beat 10 10 Orbit 0.3 s 0.3
s Collimator 0.4 s 0.4 s
40
Performance Reach 7 TeV(Nominal Settings)
Work on imperfections to approach ideal
performance reach (40 of nominal intensity)!
Here assume peak loss rate of 0.1 per second. If
lower, higher intensity can be reached!
? Chiara
41
TCT Required Settings 5 TeV(preliminary)
5 TeV flat
with separation IR n1 s for b m n1 s for b m n1 s for b m n1 s for b m n1 s for b m
IR 0.55 1.00 2.00 10.00 11.00
1 7.98 --- 16.43 --- 33.29
5 8.02 --- 16.43 --- 33.32
2 --- --- --- 36.97 n/a
8 n/a --- --- 34.69 n/a
IP2 b 0.5m

5 TeV cossing
IR n1 s for b m n1 s for b m n1 s for b m n1 s for b m n1 s for b m
IR 0.55 1.00 2.00 10.00 11.00
1 6.02 --- 13.87 --- 31.15
5 5.91 --- 13.72 --- 31.23
2 --- --- --- 30.40 n/a
8 n/a --- --- 30.47 n/a
IP2 b 0.5m
2008 settings derived from this
? T. Weiler
42
TCT Required Settings 7 TeV(preliminary)
7 TeV flat
with separation IR n1 s for b m n1 s for b m n1 s for b m n1 s for b m n1 s for b m
IR 0.55 1.00 2.00 10.00 11.00
1 7.98 --- 16.45 --- 33.29
5 8.02 --- 16.45 --- 33.32
2 7.83 --- --- 31.26 n/a
8 n/a 11.23 16.44 30.65 n/a
IP2 b 0.5m

7 TeV crossing
IR n1 s for b m n1 s for b m n1 s for b m n1 s for b m n1 s for b m
IR 0.55 1.00 2.00 10.00 11.00
1 7.13 --- 16.41 --- 31.15
5 6.99 --- 16.23 --- 31.33
2 7.71 --- --- 30.40 n/a
8 n/a 9.38 12.13 30.58 n/a
IP2 b 0.5m
? T. Weiler
2009 settings derived from this
43
The Collimation Plan on One Slide

• Pilot bunch
• Start with primary collimator only and protection
  elements.
• Keep open during first ramps and close as we
  learn about the ramp.
• Bring in additional collimators and test ramp
  functions stability.
• End of ramp compatible with b 11 m.
• 43 bunches and/or squeeze to b 2 m
• Use full installed system, according to
  collimator hierarchy.
• Close during ramp with optimized ramp settings
  (maximum tolerances).
• Squeeze without closing collimators (end of ramp
  compatible with b 2 m).
• 156 bunches and/or squeeze to b 0.55 m
• Close collimators to nominal settings and tight
  tolerances.
• Squeeze with tight collimator settings.
  Compatible with b 0.55 m.
• 75 ns running (push intensity and luminosity)
• Reduce imperfections (machine collimators) and
  improve machine stability.

44
Milestones Ahead

• July
• Install ACT tools on all 22 ring collimators plus
  maybe 6 transfer line collimators (tunnel work
  finished by end of next week!?).
• Approval of all collimators for operation (HWC
  data into MTF). Delay due to ACT installation
• Dry run move all collimators together,
  remote-controlled from the CCC. Log jaw positions
  and position measurements.
• August
• Close IR7 collimators for injection test (easy).
• Establish full machine protection ? automatic
  checking of generating interlocks when violating
  allowed position window. Demonstrate also for
  energy-dependent limits.
• Prepare functions settings, warnings, errors
  (well advanced formal check).
• Take temperature measurements and interlocks into
  operation.
• Ready for stored beam

45
Slides Assumed Shown by Stefano
46
Problem Magnetic Interference on LVDTs of TI2
Collimators (TCDIV.29012 and TCDIH.29050)
See last MAC!
The resistive magnets (5KA peak current) current
cables in TI2 generate a magnetic field that
perturbs the LVDTs collimator nominal behavior
The position drift follows the magnets current
cycle. Drift of up to 150 um have been
experienced. Reproduced in lab. Shields tested
47
Countermeasures Applied for TCDIV.29012 and
TCDIH.29050
60 reduction of the LVDTs excitation voltage as
best trade-off between magnetic interference
reduction and reading accuracy reduction (the
standard deviation of most LVDTs is still below
1 um)
Installation of a magnetic shielding to reduce
the external magnetic field on the LVDTs
The LVDT deviation is now below 50 um (apart from
one LVDT. A better shielding strategy is under
study).
Magnetic Interferences summary for collimator
TCDIH.29050
Magnetic Interferences summary for collimator
TCDIV.29012
48
Problem Overshoot on the Position Reading after
Large Displacement
As consequence of the long cables capacitance
between LVDTs and conditioning electronic a
phenomenon of reading overshoot has been
remarked. The overshoot can reach more than 100
um after a displacement of 40 mm. The recovery
time can reach several minutes.
Problem 0.3 overshoot in reading with decay
over minutes! Impact on operation Run
collimators slowly. Anyway good for protection.
No reason for cm scale fast movements with
beam! No major problem with foreseen ramp and
squeeze functions. Nevertheless, ATB found
problem and is fixing it (electronics card).
The phenomenon has been reproduced (even if not
yet understood..) in the lab and different
scenarios have been tested. The figure refers to
a 70 mm displacement
Countermeasure Again at half excitation voltage
the overshoot is reduced by a factor 4.