Eric Prebys, Fermilab Director, US LHC Accelerator Research Program (LARP) Google welcome screen from September 10, /5/2010

 Overview of the LHC  2008 Startup and “The Incident”  The Response  Startup and current commissioning status  2009/2010 Run plans  The future (as time permits) Note: This talk is a monument to plagiarism. I’ll give specific acknowledgements and “further reading” at the end. 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 8 crossing interaction points (IP’s)  Accelerator sectors labeled by which points they go between  ie, sector 3-4 goes from point 3 to point 4 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Huge, general purpose experiments:  “Medium” special purpose experiments: Compact Muon Solenoid (CMS) A Toroidal LHC ApparatuS (ATLAS) A Large Ion Collider Experiment (ALICE) B physics at the LHC (LHCb) 1/5/ Eric Prebys - LHC Talk, Aspen 2010

ParameterTevatron“nominal” LHC Circumference6.28 km (2*PI)27 km Beam Energy980 GeV 7 TeV Number of bunches Protons/bunch275x x10 9 pBar/bunch80x Stored beam energy MJ MJ* Peak luminosity3.3x10 32 cm -2 s x10 34 cm -2 s -1 Main Dipoles Bend Field4.2 T8.3 T Main Quadrupoles~200~600 Operating temperature4.2 K (liquid He)1.9K (superfluid He) *2.1 MJ ≡ “stick of dynamite”  very scary numbers 1.0x10 34 cm -2 s -1 ~ fb -1 /yr 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 1994:  The CERN Council formally approves the LHC  1995:  LHC Technical Design Report complete  2000:  LEP completes its final run  2002:  Magnet production fully transferred to industry  2005  Civil engineering complete (CMS cavern)  First dipole lowered into tunnel  2007  Last magnet delivered  All interconnections completed  2008  Accelerator complete  Last public access  Ring cold and under vacuum 1/5/ Eric Prebys - LHC Talk, Aspen 2010

For these reasons, the initial energy target was reduced to 5+5 TeV well before the start of the 2008 run.  Magnet de-training  ALL magnets were trained to achieve 7+ TeV after a thermal cycle.  After being installed in the tunnel, it was discovered that the magnets supplied by one of the three vendors “forgot” their training, and would need to be retrained to reach 7 TeV.  Symmetric Quenches  The original LHC quench protection system subtracted the inductive voltage drop by taking the difference between the voltage drop across the two apertures.  It was discovered in tests that when quenches propagate from one dipole to the next, they often do so symmetrically, rendering the system dangerously insensitive at high current. 1 st quench in tunnel 1 st Training quench above ground 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 W (M W =80 GeV)  Z (M Z =91 GeV) 200 pb -1 at 5 TeV+5 TeV ~5 fb -1 at 1 TeV+ 1 TeV Only HEP slide in this talk 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 9:35 – First beam injected  9:58 – beam past CMS to point 6 dump  10:15 – beam to point 1 (ATLAS)  10:26 – First turn!  …and there was much rejoicing Commissioning proceeded smoothly and rapidly until September 19 th, when something very bad happened 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Italian newspapers were very poetic (at least as translated by “Babel Fish”): "the black cloud of the bitterness still has not been dissolved on the small forest in which they are dipped the candid buildings of the CERN" “Lyn Evans, head of the plan, support that it was better to wait for before igniting the machine and making the verifications of the parts.“*  Or you could Google “What really happened at CERN”: * “Big Bang, il test bloccato fino all primavera 2009”, Corriere dela Sera, Sept. 24, 2008 ** ** 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Sector 3-4 was being ramped to 9.3 kA, the equivalent of 5.5 TeV  All other sectors had already been ramped to this level  Sector 3-4 had previously only been ramped to 7 kA (4.1 TeV)  At 11:18AM, a quench developed in the splice between dipole C24 and quadrupole Q24  Not initially detected by quench protection circuit  Power supply tripped at.46 sec  Discharge switches activated at.86 sec  Within the first second, an arc formed at the site of the quench  The heat of the arc caused Helium to boil.  The pressure rose beyond.13 MPa and ruptured into the insulation vacuum.  Vacuum also degraded in the beam pipe  The pressure at the vacuum barrier reached ~10 bar (design value 1.5 bar). The force was transferred to the magnet stands, which broke. *Official talk by Philippe LeBrun, Chamonix, Jan /5/ Eric Prebys - LHC Talk, Aspen 2010

Vacuum 1/3 load on cold mass (and support post) ~23 kN 1/3 load on barrier ~46 kN Pressure 1 bar Total load on 1 jack ~70 kN V. Parma 1/5/ Eric Prebys - LHC Talk, Aspen 2010

QQBI.27R3 1/5/ Eric Prebys - LHC Talk, Aspen 2010

QQBI.27R3 V2 line QQBI.27R3 N line 1/5/ Eric Prebys - LHC Talk, Aspen 2010

QBBI.B31R3 Extension by 73 mm QBQI.27R3 Bellows torn open 1/5/ Eric Prebys - LHC Talk, Aspen 2010

QQBI.27R3 M3 line QBBI.B31R3 M3 line 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

LSS3 LSS4 OK Debris MLI Soot The beam pipes were polluted with thousands of pieces of MLI and soot, from one extremity to the other of the sector clean MLIsoot Arc burned through beam vacuum pipe 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 15 Quadrupoles (MQ)  1 not removed (Q19)  14 removed 8 cold mass revamped (old CM, partial de-cryostating for cleaning and careful inspection of supports and other components) 6 new cold masses Some additional old cold masses salvageable  42 Dipoles (MBs)  3 not removed (A209,B20,C20)  39 removed 9 Re-used (old cold mass, no decryostating –except one?) 30 new cold masses New cold masses are much faster to prepare than rescuing doubtful dipoles) Many old cold masses salvageable. 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Why did the joint fail?  Inherent problems with joint design No clamps Details of joint design Solder used  Quality control problems  Why wasn’t it detected in time?  There was indirect (calorimetric) evidence of an ohmic heat loss, but these data were not routinely monitored  The bus quench protection circuit had a threshold of 1V, a factor of >1000 too high to detect the quench in time.  Why did it do so much damage?  The pressure relief system was designed around an MCI Helium release of 2 kg/s, a factor of ten below what occurred. 1/5/ Eric Prebys - LHC Talk, Aspen 2010

Theory: A resistive joint of about 220 n  with bad electrical and thermal contacts with the stabilizer No electrical contact between wedge and U- profile with the bus on at least 1 side of the joint No bonding at joint with the U-profile and the wedge A. Verweij Loss of clamping pressure on the joint, and between joint and stabilizer Degradation of transverse contact between superconducting cable and stabilizer Interruption of longitudinal electrical continuity in stabilizer Problem: this is where the evidence used to be 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Old quench protection circuit triggered at 1V on bus.  New QPS triggers at.3 mV  Factor of 3000  Should be sensitive down to 25 nOhms (thermal runaway at 7 TeV)  Can measure resistances to <1 nOhm  Concurrently installing improved quench protection for “symmetric quenches”  A problem found before September 19 th  Worrisome at >4 TeV *See talks by Arjan Verveij and Reiner Denz, Chamonix /5/ Eric Prebys - LHC Talk, Aspen 2010

New configuration on four cold sectors: Turn several existing flanges into pressure reliefs (while cold). Also reinforce stands to hold ~3 bar New configuration on four warm sectors: new flanges (12 200mm relief flanges) (DP: Design Pressure) L. Tavian *Vittorio Parma and Ofelia Capatina, Chamonix /5/ Eric Prebys - LHC Talk, Aspen 2010

 With new quench protection, it was determined that joints would only fail if they had bad thermal and bad electrical contact, and how likely is that?  Very, unfortunately  must verify copper joint  Have to warm up to at least 80K to measure Copper integrity. Solder used to solder joint had the same melting temperature as solder used to pot cable in stablizer  Solder wicked away from cable 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Electrical splice measurements everywhere while cold (measuring nΩ) Q Had to warm up sectors  Electrical stabilizer measurements everywhere while warm or at 80K (measuring µΩ) Q1 Q Had to warm up sector 45  Major new protection system based on electrical measurements Q1 – Q (nQPS)  Pressure relief valves installed everywhere possible Q1 – Q (dipoles have to be warm)  Reinforcement of floor anchors everywhere Q1 – Q Q4 2008Q1 2009Q2 2009Q3 2009Q ColdCold  WarmWarmWarm  ColdCold 23< 100K < 100K  ColdCold  80K  Cold Cold 34Warm Warm  ColdCold 45< 100K 80K  WarmWarm  ColdCold 56ColdCold  WarmWarmWarm  ColdCold 67ColdCold  WarmWarmWarm  ColdCold 78Cold< 100K< 100K  80K80K  ColdCold 81Cold< 100K< 100K  80K80K  ColdCold Q4 2008Q1 2009Q2 2009Q3 2009Q Sector 34 repairRestart 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Decision to limit energy to 1.2 TeV based on need for final shakedown of new quench protection system.  Somewhat ahead of this schedule *Taken from slides by Roger Bailey, shown at LARP meeting 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Total time: 1:43  Then things began to move with dizzying speed… 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Sunday, November 29 th  Both beams accelerated to 1.18 TeV simultaneously  Sunday, December 6 th  Stable 4x4 collisions at 450 GeV  Tuesday, December 8 th  2x2 accelerated to 1.18 TeV  First collisions seen in ATLAS before beam lost!  Monday, December 14 th  Stable 2x2 at 1.18 TeV  Collisions in all four experiments  16x16 at 450 GeV  Wednesday, December 16 th  4x4 to 1.18 TeV  Squeeze to 7m  Collisions in all four experiments  18:00 – 2009 run ended >1 million events at 450x450 GeV 50,000 events at 1.18x1.18 TeV Merry Christmas – shutdown until Feb to commission quench protection LHC Highest energy accelerator LHC Highest energy collider Should be good to 3.5 TeV after restart 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 RF  Excellent apart from some controls & procedural issues  Measurement and control of key beam parameters  Orbit, tune, chromaticity, coupling, dispersion  lifetime optimization: tune, chromaticity, orbit  energy matching  aperture  Optics checks  beating & correction  polarity checks of correctors and BPMs *Courtesy Mike Lamont 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Ramp  2 beams to 1.2 TeV Feedback excellent, feed forward show good reproducibility  Squeeze  Some work required here but impressive nonetheless  Collisions  steering, scans  Two beam operation – with and without bumps  Experiments’ magnets  Solenoids – brought on without fuss and corrected  Dipoles – brought on at 450 GeV – issues with transfer functions 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Beam dump  extensive program of tests with beam  Inject & dump, circulate & dump  Beam based alignment of TCDQ and TCS  Aperture scans  Extraction tests  Synchronization with abort gap  Asynchronous beam dump tests with de-bunched beam  Collimation  Full program of beam based positioning,  hierarchy established and respected in tests  collimation setup remained valid over 6 days, relying on orbit reproducibility and optics stability  Even the Roman pots got a run out 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Beam Position Monitors (BPM’s)  looking very good, FIFO as per injection tests  capture mode commissioned – enabling multi-turn acquisition and analysis  Beam Loss Monitors (BLM’s)  magnificent following full deployment during injection tests – a close to full operational tool  issues with SEMs, some thresholds to be adjusted, some still masked  Beam Current Measurement (DBCT, FBCT, lifetime)  commissioned and operational  controls issues  Wire scanners  operational, calibrated and giving reasonable numbers  Coupling  measured and corrected  Abort Gap Monitor  cleaning attempted with transverse damper 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Tune  BBQ FFT from day 1 – used in feedback during ramp  horizontal and vertical MKQA tune kicker for B1 and B2 operational  PLL – good progress, feedback to be tested  radial modulation tested  issues with the hump, tune stability, 8 kHz  Chromaticity  Standard method  Semi-automatic BBQ peak analysis  Radial modulation  Synchrotron light monitor  B2: undulator commissioned, SLM operational at 450 GeV and 1.2 TeV  B1: undulator not commissioned, SLM operational at 1.2 1/5/ Eric Prebys - LHC Talk, Aspen 2010

LHC Beam Commissioning Team Commissioning slides from talk by R. Assmann and F. Schmidt at recent Tevatron studies workshop 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

Kickers sweep bunches to “dilute” intensity on dump 1/5/ Eric Prebys - LHC Talk, Aspen 2010

Could get to design intensity (at injection energy) 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

B1: Q x = 0.293, Q y = 0.269; lifetime = 26h B2: Q x = 0.297, Q y = 0.267; lifetime = 5h B1: Q x = 0.293, Q y = 0.269; lifetime = 25h B2: Q x = 0.312, Q y = 0.305; lifetime = 12h 3/11 2/7 3/10 1/3 LHC Beam Commissioning Team 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Automated feedbacks seem to be working, but not quite yet standard operations.  Bottom line: things look good! Position control Bump introduced Removed by feedback loop Tune feedback Feel happy that yellow line and pink line add up to blue line 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Decision whether to go above 3.5 TeV will be made next week at Chamonix  Based on “confidence in thermal model”  50/50 according to Mike Lamont 1 month pilot & commissioning 3 month 3.5 TeV 1 month to go up in energy (maybe) 5 month 5 TeV 1 month ions 1/5/ Eric Prebys - LHC Talk, Aspen 2010

Total beam current. Limited by: Uncontrolled beam loss!! E-cloud and other instabilities  *, limited by magnet technology chromatic effects Brightness, limited by Injector chain Max tune-shift Geometric factor, related to crossing angle… *see, eg, F. Zimmermann, “CERN Upgrade Plans”, EPS-HEP 09, Krakow, for a thorough discussion of luminosity factors. If n b >156, must turn on crossing angle 1/5/ Eric Prebys - LHC Talk, Aspen 2010

Maybe. Otherwise, push luminosity at 3.5 TeV 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

Comment Energy (TeV) Max Bunches Protons/ bunch % nom. Intensity Min.  * (m) Peak Lum. (cm- 2 s -1 ) Int. Lum. (pb -1 ) Pilot Physics, Partial Squeeze, Gentle increase in bunch int x x x x10 30 ~1 Max. bunches with no angle x x10 31 ~9 Push bunch intensity x x10 31 ~ x x10 31 ~36 Increase energy to 4-5 TeV, as deemed prudent Would aim to first provide a period of physics at the higher energy without crossing angle, this could be followed by a move to 50 ns with a limited number of bunches x x10 31 ~26 Introduce 50 ns bunch trains and crossing angle! x x10 31 ~23 Push n b and N b to limit of machine safety x x10 31 ~ x x10 32 ~ x *22.1x10 32 ~110 *limited by collimation system 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Going beyond a few percent of the design luminosity depends on how far they are willing to push the existing collimation system.  Won’t really know about this until after significant running experience  Getting anywhere near requires the Phase II collimation system  Details and schedule still being worked out  Expect some guidance from Chamonix Projection assuming Phase II collimation and Phase I upgrade done in 2013/2014 shutdown* *R. Assmann, “Cassandra Talk” 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Note, at high field, max 2-3 quenches/day/sector  Sectors can be done in parallel/day/sector (can be done in parallel)  No decision yet, but it will be a while *my summary of data from A. Verveij, talk at Chamonix, Jan /5/ Eric Prebys - LHC Talk, Aspen 2010

 Initial operation  Ramp up to 1x10 34 cm -2 s -1  Phase I upgrade  After ~2 years of operation (~2014)  Replace 70 mm triplet quads with 120 mm quads   * goes from 50->30 cm  Linac4 to increase PSB injection energy to reduce space charge effects  Luminosity goes to 2-3x10 34 cm -2 s -1  Phase II upgrade  Second half of next decade (nominally 2020)  Luminosity goal: 1x10 35  Details still under study New technology for larger aperture quads (Nb 3 Sn) crab cavities? Improved injector chain (PS2 + SPL)? No major changes to optics or IR’s Possible Significant Changes 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 This talk represents the work of an almost countless number of people.  I have incorporated significant material from:  Numerous talks given at the 2009 Chamonix session regarding “The Incident”  Mirko Pojer’s talk at the US LHC Users’ Organization meeting at LBNL in September,  Oliver Bruening’s talks at the LARP collaboration meeting in November (taken from Roger Bailey)  Commissioning status slides from Mike Lamont, and also significant material shown by Ralph Assmann and Frank Schmidt at the recent Tevatron Studies Workshop  Luminosity considerations and upgrade plans, Frank Zimmermann’s talk to EPS- HEP, Krakow  All things collimation (in particular, R. Assmann “Cassandra Talk”) 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Twitter feed (big news):   Commissioning log (more technical detail):   E-logbook (very technical, but good plots):   Only visible inside CERN network (if you have a CERN account, you can use remote desktop or VPN from US). 1/5/ Eric Prebys - LHC Talk, Aspen 2010

1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Based on discussions at Chamonix 2009  Decided to warm up in 12 and 67 to replace faulty magnets  Decided to warm up sector 56 in parallel for other reasons  Warming up means  3 weeks to get to 300K  Repair work  ELQA and other issues  6 weeks to get back to 2K Q4 2008Q ColdCold  Warm 23< 100K 34Warm 45< 100K 56ColdCold  Warm 67ColdCold  Warm 78Cold< 100K 81Cold< 100K Q4 2008Q1 2009Q2 2009Q3 2009Q Sector 34 repairRestart Talk by O. Bruning, LARP CM13 meeting, November, /5/ Eric Prebys - LHC Talk, Aspen 2010

J.Ph. Tock 1/5/ Eric Prebys - LHC Talk, Aspen 2010

 Electrical measurements while warm on sectors  Confirms new problem with the copper stabilizers  Non-invasive electrical measurements to show suspicious regions Several bad regions found  Open and make precise local electrical measurements Several bad stabilizers found (30µΩ to 50µΩ) and fixed  Measured other 4 sectors at 80K (noisy but gives limits) 1/5/ Eric Prebys - LHC Talk, Aspen 2010

ParameterSymbolInitialPhase I Phase II Options Early Sep. Full CrabLow Emit. Large Piw. Ang. transverse emittance  [  m] protons per bunch N b [10 11 ] bunch spacing  t [ns] beam current I [A] longitudinal profile Gauss Flat rms bunch length  z [cm] beta* at IP1&5  [m] full crossing angle  c [  rad] Piwinski parameter  c  z /(2*  x *) peak luminosity L [10 34 cm -2 s -1 ] peak events/crossing initial lumi lifetime  L [h] Luminous region  l [cm] *excerpted from F. Zimmermann, “LHC Upgrades”, EPS-HEP 09, Krakow, July /5/ Eric Prebys - LHC Talk, Aspen 2010

Collimation at tightest settings throughout ramp and squeeze Somewhat more relaxed collimation settings *Ralph Assmann, “Cassandra Talk” 1/5/ Eric Prebys - LHC Talk, Aspen 2010