Commissioning%20of%20accelerator%20systems%20-%20magnets

1
Commissioning of accelerator systems - magnets
LHCCWG Meeting, 8 May 2007

• Presented by W. Venturini Delsolaro
• On behalf of many people
• (in HWC, FIDEL, MPP)
• Particular thanks to Adriaan, Andrzej,
  Jean-Pierre, Frank, Jörg, Luca, Massimo, Rob,
  Mike, Roberto, and Rüdiger

2
Outline

• (Brief) overview of the magnet system(s)
• Hardware commissioning enterprise
• Questions for beam commissioning
• Conclusion, open issues

3
Overview of the magnet system

• Chapters 7,8, and 9 of the LHC design report
• 28 powering sub sectors (EDMS 361532)

4
Main magnets in the arcs MB and MQ
sextupole corrector (MCS)
quadrupole MQF
orbit corrector
quadrupole MQD
quadrupole MQF
MCB
MCB
main dipole MB
main dipole MB
main dipole MB
main dipole MB
main dipole MB
main dipole MB
lattice sextupole (MS)
decapole octupole corrector (MCDO)
lattice sextupole (MS)
lattice sextupole (MS)
special corrector (MQS)
special corrector (MO)
special corrector (MO)
Picture stolen from Rüdigers talk
5
Correction magnets
6
SSS circuits (EDMS doc 104157)
7
Matching sections and dispersion suppressors
8
Separation regions and final focus
9
Hardware commissioning

• HWC was long prepared and documented within the
  Hardware Commissioning Working Group, now HWC
  coordination http//hcc.web.cern.ch/hcc/
• Strategy for hardware commissioning (EDMS 382004)
  
• Individual System Tests
• QPS, EE, PIC, ELQA
• Hardware Commissioning Activities
• SC, PIC1, PIC2, PCS, PLI, PNO, PAC
• LHC-D-HCP-0001(EDMS 477145) General Procedure
  for the commissioning of the electrical circuits
  of a sector
• LHC-D-HCP-0003(EDMS 519716) Powering of the
  superconducting circuits of a sector up to
  nominal current
• Detailed procedures prepared by MPP
• Training day on 29 March 2007 all presentations
  can be found on http//indico.cern.ch/conference
  TimeTable.py?confId13360

10
The magnet system shaping the LHC beam is the
innermost hand of a complex body behind
it Hardware commissioning is now checking this up
completely for example heat runs to validate the
functionality of cooling, ventilation, power
controls, etc
Magnet system
Networks
Electric Power
Cooling
Ventilation
11
Questions to be addressed

• What will be the status of the system when we
  start beam commissioning?
• Which signals are available in the control room?
• What needs to be measured with beam?
• What is the commissioning plan with beam (if
  any). Is any special beam time required to
  commission the system?
• How can the system affect beam operation?

12
What will be the status of the system when we
start beam commissioning?

• The goal of HWC is to release all systems ready
  to be operated at their nominal performance
  level, OR, more realistically, with a complete
  list of NON CONFORMITIES
• First experience with sector 7-8 shows that we
  may have to accept some limitations, translating
  in current ratings lower than nominal for some
  circuits
• However, sector 7-8 is (hopefully) not fully
  representative it is used as a playground to
  validate and tune HWC procedures
• Moreover it will be reopened, allowing repairs to
  be done before commissioning with beam
• A list of magnet circuits not needed for the
  first phases of beam commissioning was presented
  at Divonne 2006 (M. Giovannozzi)
• Expected values for the number of quenches during
  HWC have been given by P. Pugnat in the same
  workshop
• Hardware commissioning is completed by a dry run
  (powering all circuits at unison through the
  whole machine cycle)

13
MPP recommendations for 600A circuits in sector
7-8, first run
14
Magnet performance panel

• The MPP ( the CPP) are expected to actively
  contribute to the Hardware Commissioning effort
  by
• participating to the definition of the
  procedures,
• analyzing the data gathered,
• supplying all the information necessary and
  timely validating the continuation of the
  commissioning procedures interpreting the data
  and depending on it allowing/refusing the
  execution of the following step of the test
  procedure
• in case of failure, the MPP/CPP is expected to
  perform in-depth diagnostics to unblock the
  situation or escalate the problem via Hardware
  Commissioning Coordination
• R. Saban at the HWC training day

15
Non conformities can affect machine performance,
but they may be tolerated for the first stages of
beam commissioning depending on the circuits
concerned During series tests of magnets, this
was eventually handled by the MEB. How do we
deal with NC during the HWC, when correcting
actions might still be possible? Escalation of
Non Conformities (simplified, see details in
http//indico.cern.ch/sessionDisplay.py?sessionId
1slotId0confId133602007-03-29)
Very BIG Troubles
!
LHCCWG ?
16

• So, what will be the status of the system at the
  start of beam commissioning ?
• It depends on the type of non conformities
  found during HWC and on the their outcome
• This can have an influence on the procedures of
  beam commissioning
• ? the LHCCWG needs up-to-date forecasts of the
  expected performance of circuits

17
Which signals are available in the control room?

• Fixed displays for Power Converters status (LSA)
• GPMA application for voltages, currents, logic
  states, etc (Labview)
• Circuits that do not quench will not produce any
  QPS PM data

Courtesy A. Rijllart
18
What needs to be measured with beam?

• Polarities
• Calibration of orbit correctors
• Matching of MB fields over the octants
• Gross tune errors
• Beta Beat
• Alignment errors
• Calibration of lattice correctors and MCS spool
  pieces
• MB/MQ tracking
• Transfer functions of some critical elements (D1,
  D2) at 450 GeV and 7 TeV
• Quench levels for BLM thresholds (see Laurettes
  talk )

19
What is the commissioning plan with beam Is any
special beam time required to commission the
system? (1)

• SETTINGS AND TRIMS of MAGNETIC FIELDS
• Involves almost all phases of beam commissioning
• Pre cycling, MB, MQ, MCB, MCBC, MCBY transfer
  functions (A-1, A-2)
• Stability and reproducibility (A-4)
• MB and MQ errors, MQS, MQT, MQTL, MS, MCS
  transfer functions (A-4)
• MCD transfer functions? (A-5)
• Dynamic field model, refined MCS transfer
  functions (A-8)
• Transfer functions of D1, D2 (A-9)
• MQM, MQY hysteresis models (A-11)
• Probably more input needed for stages B, C, D
• FIRST SETTINGS WILL COME FROM FIDEL

20
Corrections based on Beam Measurements

• Friday 4/4/2007 first SPS turn based on
  trajectory measurement, the current in the MB was
  changed by some 1.6 A, and the beam went from a
  few turns to a few millions! Then the new
  injection level was joined by interpolation to
  the higher energy points, to allow the power
  converters ramping the magnets quick and
  efficient.
• But LHC Beam based corrections will have to be
  integrated in the magnetic model (see Luca and
  Mikes talks at Divonne 2006)
• Cumbersome? The complexity is such that we cannot
  just trim around, there needs to be a coherent
  frame

21
Field Description for LHC project
http//fidel.web.cern.ch/fidel/
How will this work?
22
The problem is how to adjust the model parameters
based on beam measurements If the model
incorporates blindly all trims that turn out to
work in the CCC, it will rapidly loose physical
sense becoming totally ad hoc. On the other
hand, the model must be open to revisions based
on experience. But it has to remain
physicalpossible to extrapolate We need to
understand the beam and the magnets at the same
time, as the magnet system has enormous influence
on the beam Issues 1. Automatic feedback
corrections superimposed through real time
channels should be taken into account for a
correct analysis 2. Trims based on beam
measurements come as field/current increments,
not as Fidel components. They cannot be directly
incorporated into the model, but need to be
analyzed to be correctly decomposed.
23
What is the commissioning plan with beam Is any
special beam time required to commission the
system? (2)

• POWER PERFORMANCE quenches are of 3 types
• Natural hopefully most during HWC. Consequence
  will be unscheduled down time, GPMA should give
  green light to go ahead. No need to plan for any
  beam time
• False triggers estimate rates are available in
  the LHC report. Diagnosis based on timing of
  logical states, but causes might be difficult to
  find out. Consistency check foreseen before
  powering. No need to plan for any beam time
• Beam induced may lead to reconsider optics,
  collimator settings, BLM thresholds
• the level of beam loss induced quenches will be
  assessed either via quench and learn to set BLM
  thresholds, or parasitically.
• At least one good argument supports the first
  option if the damage level and the quench level
  are close at high energy for some type of losses,
  and if energy scaling is reliable, then quench
  and learn at 450 GeV will validly protect the
  machine. (See Laurettes talk)
• Foreseen at the beginning of the phase A1.5
  (increasing intensity)

24
How do we deal with quenches during beam operation

• PM analysis
• Manual and GPMA
• Circuit release
• Automatic, operation, MPP
• Recycling prescriptions
• Normally never beyond a powering sub sector
• Events that do not give rise to recycling in
  general, quench of magnets with no decay

25
Magnet pre cycling depends on the level of
performance reached during Hardware
commissioning? Prescriptions being prepared by
R. Wolf with FIDEL

• Example simple cycling against
• degaussing of correctors

26
Time lags for correctors with parallel resistors
L(I)

• t will depend on current (L varies up to 30)
• To be taken into account when tuning the PID
• And in the settings generation for the ramp, for
  example for snapback compensation

Rpar
27
Conclusions

• The status of the magnet system at the end of
  hardware commissioning will have direct impact on
  machine energy, optics and tunability, and on
  operation procedures. Keep an eye on expected
  performance.
• Software for automatic PM analysis aims at
  minimizing the need of expert support to release
  powering after quenches
• Pre cycling prescriptions being prepared
• Tracking test in SM18 is in preparation
• Corrections based on beam measurements how to
  integrate them in the magnetic model?