Operation of the LHC Cryogenics system and interface with beam & machine operation

Transcription

1 Operation of the LHC Cryogenics system and interface with beam & machine operation S. Claudet (CERN, Geneva) on behalf of the Cryogenics Group Technology Department

2 Outline Introduction to LHC Cryogenics Operation, organisation and results Availability and interaction with beam operations Summary 2/30

3 LHC accelerator p-p collision cm -2.s -1, 14 TeV, 0.5 GJ stored energy Machine operation Technology 24 km of superconducting K, 8.33 T 3/30

4 Layout of LHC cryogenics Pt 5 Pt 4 Pt 6 Pt 3 8 x 4.5 K sc magnets 24 Cryoplant km & 20 Distribution 1.8 K Present Version K 130 t He inventory Pt 7 LHC cryogenics is the largest, the longest and the most complex cryogenic system worldwide Pt 2 Pt 1.8 Cryogenic plant Pt 1 Magnets Distribution Pt 8 4/30

5 How does it compare? LHC, ATLAS, CMS LHC Before LHC: existing experience for design, safety, controls, operation, availability, ITER OMEGA, BEBC, ISR Low-Beta LEP2 ALEPH, DELPHI, LEP Low-Beta LEP LHC: 144 kw Tevatron, RHIC, Jlab, SNS, HERA, Tristan, Year We did not start from scratch! 5/30

6 LHC compressor station (x8) 4.2MW input power Bldg: 15m x 25m Oil/Helium Coolers Compressors Motors 6/30

7 K Refrigerators (x8) K to 75 K K to 20 K - 41 g/s liquefaction LHe: l/h 4m diam, 20m long, 100tons 7/30

8 1.8K Units with cold compressors (x8) 300 K under atmosphere Active magnetic bearings Cold Compressor 3-phase induction Electrical motor (rotational speed: 200 to Hz) Hz) 125 g/s GHe from 15 mbar to P atm with 3 or 4 stages Cold under vacuum Outlet Pressure ratio 2 to 4 Inlet Fixed-vane diffuser Spiral volute Axial-centrifugal Impeller (3D) 8/30

9 Electrical feed boxes for current leads 48 Boxes, 1200 leads LSSL2 of the LHC 9/30

10 One LHC sector: production-distribution-magnets Total for 8 sectors: Compressors: 64 Turbines: 74 Cold Comp.: 28 Leads: I/O signals: PID loops: Extremely large installed cooling capacity Complexity associated with 1.8K units Extremely large distribution system => Recovery from failures can last from few minutes to 20 hrs, exceptionally 2-3 days x 13.5 From LHC Magnet String test 3.3km 10/30

11 Interfaces: follow-up electrical perturbations EL perturbations and their impact on our LHC Cryo system Voltage change [%] Electrical systems recover in ms Cooling systems recover in min Cryo systems recover in hrs => A big incentive to be as tolerante to glitches as possible Duration [ms] Typical tolerance envelope 11/30

12 Main reasons to superconducting For accelerators in high energy physics Compactness through higher fields E beam 0.3. B. r E beam E. L [Gev] [T] [m] [Gev] [MV/m] [m] Be sure that at design stage, working at higher temperature was considered, but not selected to maximise LHC beam energy => Cryogenic systems takes longer to recover from failures than conventional ones (but we work on it!) Saving operating energy Electromagnets: Acceleration cavities Resistive: P input E beam P input Rs.L.E 2 /w Superconducting: P input Pref R s R BCS + R o R BCS (1/T) exp(-bt c /T) 12/30

13 Interactions between LHC systems Powering OK or interlock Static and Dynamic heat loads Beam related Dynamic heat loads 13/30

14 Outline Introduction to LHC Cryogenics Operation, organisation and results Availability and interaction with beam operations Summary 14/30

15 Key factors for operation Equipment architecture: Central liquefier to intermediate buffer, distribution decoupled Cooling capacity production in line with demands Type of operation Transients (cool-down / warm-up) or various recovery Alarm monitoring, simple reset actions, calling for experts Detection of process degradation and curing action HW checks and preventive treatment of slow evolving problems Frequency of required actions: Once per month, once per week Once per 1-2 days On-call adapted LHC: A huge and complex system without significant buffer and frequent operator actions required Dedicated 24/7 required so far!!! 15/30

16 Structure - Coordination - Outils Mechanics Management Electricity-Controls Instrum-Cryolab Methods - Logistics Operation Accel. Operation Detect. Coordinations: Team Leaders + Management (1/wk) Performance panel (1/2wks) Operation / Maintenance panel Methods & Tools panel Tools (web interface DB oracle): e-logboog operation for any change of configuration (wanted or not) or observation and diagnostic request Diagnostic tables, work-orders, intervention reports Asset & spares management, intervention procedures Maintenance plan Scheduling 16/30

17 Staff & team evolution People should be able to quit, newcomers should be integrated High level requirements for recruitment (Bachelor & Masters) Formalised induction process: Academic training - On the job training - Shadow shifts => Certification after 10 months as shift operator (alone!) Senior operator (>3 yrs): Able with all sub-systems, ability to optimise production-needs-time Certification diploma: Written - Site - Simulator - Improvement study (report + presentation) If selected for indefinite contract: Operation for 5 to 10 years Ability to become production Eng. as site responsible Ability to switch to support teams or another activity at Cern 17/30

18 Cryo operator in Cern Central Control room Shift 24/7 Fixed displays Tendancy curves (summary) Process synoptics and orders 18/30

19 Operation, indicators Alarms Powering Efficiency Global availability 19/30

20 Outline Introduction to LHC Cryogenics Operation, organisation and results Availability and interaction with beam operations Summary 20/30

21 CM CS SP CM CS Availability: a signal Yes/No is required T2 = Achieved up time during required time / Required time x 100 (operational availability) Cryo Maintain: Few important conditions checking integrity of HW, with slow power abort in case this signal is lost (leading to beam dump!) set-point Cryo Start: set of conditions to allow powering of concerned sub-sector, with no action if powering started (illustrates good stability of process) T2 indicator w.r.t EN CM CS CM Sum CM 8 sectors: Global availability Possibility to treat thousands of channels in a structured way to match at best the LHC powering sub-sectorisation and the cryo sub-sectorisation 21/30

22 LHCCryo global availability Target % 70 Availability xP8 CCs L7 TT891 SEU? P8+R5 P18 P4 oil ice P4 CCs - Excellent 1st part to TechSTop - Heavy works done during Technical Stop #1, and cabling weakness caused difficult recovery - Very moderate impact from High Luminosity operation in Feb 26-Mar 23-Apr 21-May 18-Jun 16-Jul 13-Aug 10-Sep 8-Oct 5-Nov 3-Dec Scheduled Stops Daily 2012 Weekly 2012 Between TS /30

23 Performance and origin of downtine LHCCryo - Average of 8 sectors (Between TS) Global availability as seen by LHC during beam operation periods Others according to relative ratio of their average for the 8 sectors Percent [%] Supply (EL, CV, IT) Cryo Cryo SEU Users Global availability Evolution: (260 days) (271 days) (137/290) (Full days, Mondays & Fridays of Technical Stops not counted here) : Correcting early Cryo bugs : Adapting to SEU : So far rewarding!!! 23/30

24 Availability: from global to single plant 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Considering 8 independent sectors % 60% 70% 80% 90% 100% Individual Single sector and cryoplant 24/30

25 Indicators: recovery categories & tendency Nb stops Better global control/tuning (operation, instrumentation) -! SEU! Less Cryo induced failures,. (but 3 VERY LONG ones!) - More Supply (EL) failures -! SEU! No longer recurrent Cold Compressors particular issues (Leaks, electronics) Cryo SEU Cryo down short(<8h) Medium(8h-30h) Long (>30h) From the books: Immediate effect of (good!) practice Annoying if frequent, to be kept low with moderate efforts Serious cases requiring specific monitoring and significant efforts Nice tendency, promising for 2012 or new surprises to come up? 25/30

26 Operation structure & approach 2007/2008 cool-down & HWC: Per site, one experienced engineer with agreed minimum protocol to guide a local team of operators, with help of support teams (instrumentation, experts, controls) Since 2009 and operation with beam: One operator in shift 24h/7d, more transverse structure site/cerncontrolcenter, procedures & operation tools For machine controls (temperature, level, pressure): Basic interlocks and simple PID loops with generic tools for fast orders, now completed with automated sequences & procedures Indicators: From temperature stability to daily availability on-line cool-down curves to on-line cryo-status Control rooms: site - CCC- office 26/30

27 Power Consumption for LHC Cryogenics Power [MW] input Jul Installed power Cool down HWC 700 h Stop Cryoplant Operation with 8 plants Tests LHC physics 6200 h Cryo optimized power LHC physics 6500 h Net gain 50 GW.h per year (3 MCHF / year!!!) 91% Cryo Availability 90% Cryo Availability Cryo Cryo unavailability 3 Utilities breakdown Users or Beam 4 02-Oct Jan Apr Jul Oct Dec Apr Jul Sep Gain 8MW (20% of installed power) 27/30

28 Helium inventory follow-up Total masse He 2012 Livraison camion He Remplissage DEWAR Transfert SM18 Transfert CAST Transfert CMS Helium invenrory [tons] en tonne TS T Technical Stop Now < 30kg/day With 50kg/day in 2011 and better tendency to be CHF/kg - 50kg/day # CHF/day t # 7.5 MCHF v.-12 v.-12 v.-12 v.-12 v.-12 vr.-12 vr.-12 vr.-12 vr.-12 rs-12 rs-12 rs-12 rs-12 vr.-12 vr.-12 vr.-12 vr.-12 vr.-12 ai-12 ai-12 ai-12 ai-12 in-12 in-12 in-12 in-12 uil.-12 uil.-12 uil.-12 uil.-12 uil.-12 ût-12 ût-12 ût-12 ût-12 pt.-12 pt.-12 pt.-12 pt.-12 ct.-12 ct.-12 ct.-12 ct.-12 ct.-12 v.-12 v.-12 v.-12 v.-12 c.-12 c.-12 c.-12 c.-12 c.-12 Helium Losses [tons] Xmas Tech Stop Operation en tonne /30

29 Interfaces with Beam-OP HW signals: Cryo Start and Cryo Maintain towards Powering Interlock module SW panels: Cryo web page People in Control Room (LHC): 1 Eng in charge + 1 operator 1 Cryo operator 1 operator for technical infrastructure Possible evolutions? Closer discussions with Eng. In charge in case of cryo problem Other operators involved to help diagnostics/recovery No longer cryo operators at night (on call only) Text zone 29/30

30 Summary LHC cryogenics is the largest, the longest and the most complex cryogenic system worldwide. We could achieve a reasonable availablity (> 90 %) so far with beams. This demonstrates that there are no big issues in concept, technology or global approach for operation. Despite all our efforts, we had very hard time and lengthy commissioning to learn how to tune all these sub-systems together while permanently consolidating what was not conform. Experience has been converted into automatism, procedures, tools, training Cryogenics operation is well integrated in central control room with LHC main systems, but operated/supported independently (about 50 people) Maintenance is as well reaching an efficient preventive/corrective ratio, with efforts to be made for non-standard cases. We have to prepare for higher energies and intensities with continued gain in reliability! 30/30