Radiation To Electronics Reality or Fata Morgana Mirage

Radiation To Electronics: Reality or Fata Morgana (Mirage). … thanks Ralph … Radiation 2 Electronics Chamonix 2011 M. Brugger for the R 2 E Project !!! Many Thanks To Everybody !!! www. cern. ch/r 2 e Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)

Next Shutdown 2012/13/ 14/15/… Chamonix 2011 2 R 2 E Reality or Fata Morgana (Mirage)

Radiation Issues - Effects Thanks to Y. Thurel Total Ionizing Dose (TID) + Displacement Damage (DD) Cumulative effect (Effect will be seen after given time, giving some freedom to react) Low stress level (50 Gy max in 2030 when “standard component” can survive 20 -30 Gy) 0 LHC Tunnel: 2009 Critical Areas: ( after shielding ) 2009 Component Stress Conditions Easy Medium 100 Hard 10 100% now 1% 2010 50 time Condition severity vs time some 2030 Single Event Effects [SEE] Stochastic Effect (“Events scale with number of affected components”) Very High stress level (Failures observed <1 x 106 cm-2) 0 Component Stress Conditions Easy 1 E 6 Medium 1 E 8 Hard 100% now 0. 1% 2010 1 E 9 Critical Areas (as today): 2009 LHC Tunnel (well tested and “safe”): Chamonix 2011 2030 Condition severity vs time 2012 -15 up to 2030 2009 3 R 2 E Reality or Fata Morgana (Mirage)

Radiation Issues – Failure Observation TID + Displacement Damage Thanks to Y. Thurel Devices get slowly out of tolerance (final failure can often be anticipated; access not immediately required) No ‘early’ failures (due to radiation) Failures Possible Scenario: Anomaly Observed Expert/CCC Informed 2000 Correction/ Steering Access at later Stage 2030 Failure Rates Single Event Effects Failures will appear and rapidly increase in frequency (destructive failures possible; access often required) ‘Early Operation ’ problem (observation might falsify reality) Possible Scenario: Failures Chamonix 2011 Sudden Problem Expert/CCC Informed Unclear Situation Access Required 4 2010 2030 Failure Rates R 2 E Reality or Fata Morgana (Mirage)

Confidence: CNRAD Update Tested Equipment: PLC-S 7 -200 (CV) [profibus lost, reset needed]: 24 V DC Power Supply (CV) [burned]: PLC-S 7 -300 (CV) [blocked, reset needed]: PLC-Schneider (CV) [PS burned]: WIC Rack [beam dump and access]: PLC S 7 -300 + FM 352 -5 Siemens Fire Detectors ASD [power cycle]: Ethernet Switch [blocked, reset needed]: … Failure x. Section: 1. 8 x 10 -7 cm 2 1. 1 x 10 -8 cm 2 ~200 failures in a nominal year in UJ 14/16 7. 8 x 10 -8 cm 2 1. 1 x 10 -7 cm 2 1. 0 x 10 -9 cm 2 1. 4 x 10 -8 cm 2 “Reset Req. ”: 1 every 106 -109 cm-2 “Damaged”: 1 every 107 -1010 cm-2 Uncertainty: up to one order of magnitude (but both directions!) © D. Kramer Chamonix 2011 5 R 2 E Reality or Fata Morgana (Mirage)

Confidence: FLUKA Calculations Location Monitors CERF, CNGS, … TI 2/8 UX/US 85 Source Rad. Mon, RAMSES, Beam on target TLDs Rad. Mon Controlled loss on TED and TCDI Rad. Mon, LHCb collissions RAMSES IR 7/UJ 76/R R 77 Rad. Mon Losses on Collimators IR 1/5 Rad. Mon Collisions Agreement Comments Within 10 -20% Benchmark setup Within 30% Source term well controlled Within 30 -50% Detector update required Mostly within a factor of two Very sensitive on loss distribution Within a Only QUALITATIVE factor of 2 -3 check Uncertainty: Dominated by the source term and the considered details! Chamonix 2011 6 R 2 E Reality or Fata Morgana (Mirage)

IR 7 FLUKA Application Benchmark © K. Roeed Assumptions FLUKA Calculations Loss-Term (Six. Track [R. Assmann et al. ]) TCP TCAP MBW TCS MQW TCLA Normalisation >500 m Summary (Protons) RR 73 UJ 76 LM 01 LM 02 Rad. Mon Dcum 6 L 7. 7 LM 03 S 19846 5 L 7. 7 LM 02 S 19904 4 L 7. 7 LM 01 S 19991 4 R 7. 7 RM 03 S 20045 5 R 7. 7 RM 04 S 20133 6 R 7. 7 RM 05 S 20208 RR 77. 7 RM 06 S 20241 Chamonix 2011 Rth 5 3 5 31 1 LM 03 SEU (3 V) Measured 14401 5253 2689 950 18727 303 13 RR 77 RM 03 RM 04 RM 05 RM 06 Beam SEU (FLUKA) contribution Expected B 1 15015 B 1 9765 B 1+B 2 3116 B 1+B 2 401 B 2 13032 B 2 962 B 1+B 2 17 Error [%] Stat. only 3 3 6 6 4 8 22 Exp. /Mes. 1. 04 1. 86 1. 16 0. 42 0. 70 3. 17 1. 33 In 6. 02 E+15 Dumped 5. 82 E+15 Lost in Machine 1. 99 E+14 Of Lost protons Collisions 2. 33 E+13 Elsewhere 1. 76 E+14 96. 70% 3. 30% 11. 73% 88. 27% BLM ratio IR 7 / IR 3 Ratio TCSG. A 6 L 7. B 1 / TCSG. 5 L 3. B 1 3. 1 TCSG. A 6 R 7. B 2 / TCSG. 5 R 3. B 2 5. 6 % Loss in IR 7 76 85 R 2 E Reality or Fata Morgana (Mirage)

Radiation Levels 2010 Source terms, operational conditions as well as monitor readings have to be carefully evaluated Tracking/B eam. Gas/etc. FLUKA Prediction s Operation Assumption s Rad-Level Estimate Monitor Readings Calibration Assumption s Final Values Threshold of monitors is ~106! (lower values only by ‘trick’ [big uncertainties!]) VERY GOOD AGREEMENT (given the underlying uncertainties) Chamonix 2011 8 R 2 E Reality or Fata Morgana (Mirage)

Possible SEE Failures Observed in 2010 WIC crate failure in TI 8 Observed in 2009 Known problem with moderate x-section QPS Tunnel Card Failures (2 x in 9 L 7 [ions], 2 x in 8 R 8 [inj. ], + others) ISO 150 -> permanent PM trigger SEE confirmed (EMC has same effect) QPS tunnel Card Failures in 9 R 7 & 9 L 7 u. Fip communication lost (2 x) SEE confirmed (seen in CNRAD) CONFIRMED or very LIKELY CRYO tunnel card SEE in 8 L 2 1 Fault in u. Fip (as observed in CNRAD 2010) SEE confirmed TE/EPC power supply burnout in UA 87 Same effect observed in CNRAD SEE is very likely the cause (Streaming through Maze) NOT CONFIRMED (unlikely) VAC power supply burn out In UA 23 between maze and duct (TDI losses + TCDI losses ) SEE rather unlikely Chamonix 2011 PXI power supply burnout in UJ 16 To be confirmed by producer (comparison with CNRAD burnouts) SEE unlikely (early 2010 operation) R 2 E Reality or Fata Morgana (Mirage)

Radiation Issues – 2010 Summary Single Event 2010 Review – Levels in the machine § Only 105 -108 cm-2 “measured” around (UJ, RR, Tunnel) equipments § Only ~0. 1% of nominal integrated luminosity, up to 2% of peak luminosity, ~1% of nominal lost beam, “no” scrubbing (yet) – What is good? § Simulations correctly verified in many places (controlled tests & standard operation) § Not too many failures occurred Chamonix 2011 10 R 2 E Reality or Fata Morgana (Mirage)

Radiation Issues – 2010 Summary Single Event 2010 Review – What is worrying? § Critical areas: previsions beginning of 2010 expected quite some events in critical areas (several candidate/s, within error margins) § Tunnel: 5 -10 SEE events already seen in 2010 causing Machine Stop (mostly mitigated) § ”we are already wrong” … in the bad direction – What can make things worse § equipments can be more sensitive (most is untested)? § we didn’t see “critical” levels (mostly < 1 x 106) § losses can also be higher than expected (e. g. , electron cloud, injection losses, life-time for ions) Chamonix 2011 11 R 2 E Reality or Fata Morgana (Mirage)

2010/2011/2012 Radiation Levels © D. Kramer 2011/2012/Nominal/+++ ? ? ? !!! Amazing !!! Close to ‘Threshold’ Critical Chamonix 2011 12 Dramatic R 2 E Reality or Fata Morgana (Mirage)

Failure Rates 2010/2011/2012++ For 2010 we expected already some failures (estimate of July) HIGH BUT OK Adding real operation, measured radiation levels and x. Sections Refined estimate possible based on: Radiation Levels Rough Failure x. Sec First Rate Estimate Updated Rad. + x. Secs 0. 05 fb-1 immediate dump and access immediate dump Scheduled access Other 2010 SUM MTBF [days] 18 53 33 56 2 1 Operation Assumptio n Expected Rad. Levels Failures Rates Xsect. rescaled 2010 (from Meas. ) 20 7 11 7 Rescaled x. Sections 150 570 220 480 2011 SUM MTBF [days] 98 19 35 30 4 19 10 12 2012 SUM MTBF [days] 166 33 60 52 2 11 6 7 nominal SUM MTBF [days] 2500 440 740 0. 14 0. 8 0. 5 R 2 E Mandate-> radiation induced MTBF <= 1 per week for Ultimate Intensities, losses and luminosities (a long way to go, even with uncertainties) Chamonix 2011 13 R 2 E Reality or Fata Morgana (Mirage)

Failure Rates 2010/2011/2012++ Chamonix 2011 Unknown Equipment Sensitivity 14 Monitor Calibration We’re missing a Factor 100 Sensitivity analysis based on failure x. Sections © D. Kramer Uncertainties: LHC operation & machine behavior, radiation levels, equipment sensitivities R 2 E Reality or Fata Morgana (Mirage)

2011 LHC Energy/2012 Operation Energy greater than 3. 5 Te. V: Only minor impact on radiation levels, thus not an issue from the R 2 E point-of-view Impact of 2012 operation Will lead to a delay of R 2 E mitigation measures (shielding/relocation) Impact on operation not to be excluded Risk of destructive failures Failure rate expected to be (just) acceptable Chamonix 2011 15 R 2 E Reality or Fata Morgana (Mirage)

What We Will Do… Preparation of shielding & relocation measures 2011 experience together with detailed monitoring & scheduled radiation tests (full power-converters) will allow us a further optimization step Monitoring and preparation of patch solutions Our Strategy: Anticipate problems whenever possible Aim to be ready for 2012 shutdown in any case Optimize the long-term solution Chamonix 2011 16 R 2 E Reality or Fata Morgana (Mirage)

To be kept in mind… Tests with (LHC) beam Field-calibration (for detectors) measurements: quench-test location would be ideal for additional loss/radiation -field studies TCDI near UJ 87 would be an additional good spot Scrubbing & Beam-Gas © V. Baglin (So far) very low radiation levels in most of the ARC/DS locations (only specific loss locations [including some surprises] are concerned already ~1 Gy in some cases!) Minor effect on shielded areas Tunnel equipment will be exposed Ion-Operation One month of ions is in some locations worse than one year of nominal! Chamonix 2011 17 R 2 E Reality or Fata Morgana (Mirage)

R 2 E Global Strategy & Dates Panic Study Phase (2008 -2010) Calculations, Early Actions, Strategy Evaluation Struggling Project Proposal (2010) Testing, Evaluation, Mitigation Plan Resources R 2 E Mitigation Project (2011 -2016) Medium Term Shielding & Relocation P 1, P 5, P 7, P 8 (+ small items) Collimation Betatron in IR 3 R&D For Long. Term Power. Converters Rad-Tol Solution SCLs Horizontal/ Vertical Links Evaluation Long-Term Monitoring Analysis, Calibration Testing Component & System Tests Combination Relocation Rad-Tol Equipment SCLs (Civil Engineering) Key Dates (Status 01. 2011): 11. (1 st Test Results) 2010 Operation 08. 11. (CE pre-study) 02. 11. (Cables) 12. 11. (1 st CD Review) 06. 11. (UX/UL@P 1) 06. 12. (CE Study) 03. 11. (Safe. Room 1) 2012 (FGCs Production) 10. 11. (Refine) 2013 (Hori-SCLs) 06. 11. (Safe. Room 2) 2014 (PCs and/or SCLs) 07. 11. (Orders for SD) 2013 (Prototype) Chamonix 2011 2014 (Production) R 2 E Reality or Fata Morgana 18 (Mirage)

Chamonix Valley Next Shutdown 2012/13/ 14/15/… THANK YOU R 2 E Seminar (+Drink) Tent. Date: February 18 th Chamonix 2011 19 R 2 E Reality or Fata Morgana (Mirage)

Backup Chamonix 2011 20 R 2 E Reality or Fata Morgana (Mirage)

R 2 E Mitigation Project Plan LHC 2011 2012 I <=100% L <=100% E=1 2014 -> R 2 E Long-Term LHC Performance • Radiation Tolerant Power-Converter R 2 E Long-Term ? ? ? • Additional shielding upgrades where required • Tunnel equipment upgrades (if needed) LHC Long-Term ? ? ? • Vertical SCLs (LHC Upgrade UJs @P 1/5, R 2 E only @P 1) • Additional shielding (e. g. , RA/UA ducts, TI 2/8, UJ/UA Junctions) • Further radiation tolerant equipment 2016 or I >=100% L >=100% E=1 ? ? ? OR • Horizontal SCLs for RRs @P 7 • Vert. (or Hor. ) SCLs for RRs at P 1/5 ? ? ? OR R 2 E Time Gain Chamonix 2011 OR • Civil Engineering – New Caverns 21 Hor. SCLs X Rad-Tol Power. Converters X Vert. SCLs I X Design Reviews Procurement I <=50% L <=25% E ~= 1/2 2012 or 2013 • Relocation US 85 + Shielding • Shielding: UJ 14/16/56 (23/87) • Relocation UJ 76 (remaining) • Fire Detectors (UJ 14/16/56/76, RR 13/17/53/57) • Shielding: RR 13/17/53/57 (mid/long-term)? • Relocation P 1 (UJ 14/16) • Relocation P 5 (UJ 56) • Betatron Collimation @ Point-3 (? ? ? ) R 2 E Safety & Patch R&D Development I <=30% L <=15% E ~= 1/2 • Fire/ODH Control Racks + EN/EL RTUs (Anticipate) • Preparations for long shutdown Purpose(s) Development 2010 Mitigation Actions Development I <=10% L <=1% E = 1/2 Shutdown Vert. SCLs II (high-curr. ) X LHC Upgrade R 2 E Long-Term R 2 E Reality or Fata Morgana (Mirage)

R 2 E Mitigation Project Building Blocks Calculations Radiation Tests Integration Developments Monitoring Implementation Chamonix 2011 People, Planning & Money R 2 E Reality or Fata Morgana (Mirage)

Building Blocks for the R 2 E Project Integration & Implementation Monitoring & Calculations WG: MC Calculations Monitoring & Benchmarking Radiation Levels in Critical Areas M. Calviani A. L. Perrot EN/EL EN EN/MEF TE Radiation Tolerant Power-Converter WG: R&D & Concept Development Testing (Procurement) Y. Thurel GS/ASE BE IT/CS TE/MPE Pre-Studies Impact on Long. Term J. Osborne GS/SE Project Support BE/OP EN/STI DGS DG TE/VSC FP Organization Planning Safety EN/CV DG/PRJ TE/EPC R 2 E Mitigation Project TE/CRG TE/MSC Super Conducting Links SCLs -WG: R&D & Tests Hor. SCLs Vert. SCLs (Procurement) DGS/RP M. Brugger BE/CO DGS/SEE FP/PI BE/ABP S. Weisz K. Foraz S. Roesler C. Jach Test Facilities PH/DT Collimation-WG: Betatron @ Point-3 A. Ballarino Chamonix 2011 GS PH BE/CO BE/BI Civil Engineering Radiation-WG: Failure Analysis Radiation Tests Patch Solutions & New Developments Radiation Policy G. Spiezia LHC Planning: Shielding Relocation Coordination BE/OP Power Converters Radiation Testing New-Irradiation Facilities WG: PS East-Area Facility M. Moll R. Assmann 23 R 2 E Reality or Fata Morgana (Mirage)

R 2 E Approach Constraints & Mandate: Minimize (Avoid) any risk of radiation induced failure to electronics Foresee a mitigation plan fitting in the planned shutdown periods Optimize with respect to planning and costs Strategy: Monitor and benchmark to refine actions and planning Prepare “Patch-Solutions” where available (not many left!) Shield and Envisage/Prepare/Perform Relocations Study/Pursue Major Long-Term Solutions (R&D for SCLs and Rad-Tol PCs, CE as backup) … cross fingers, review, optimize, cross fingers… Chamonix 2011 24 R 2 E Reality or Fata Morgana (Mirage)

What’s to Be Avoided ACCESS: LHC REPAIR Chamonix 2011 25 R 2 E Reality or Fata Morgana (Mirage)

R 2 E Work 2010/11 x. Mas. Break LHC Point-7: UJ 76 shielding wall; done UJ 56 -TZ 76: relocation of the ODH and fire detection control central; done UJ 56 -TZ 76: RTUs relocation; (done) 24 th January; LHC Point-5: UJ 56 -USC 55: relocation of the fire detection control central; done (reception on the 25 th January) UJ 56 - USC 55: RTUs relocation; (to be done) 31 rd January; LHC Point-8: EN/EL: advance on the cables installation work foreseen during the next long shut down; LHC Point-1: (optional and kept in pipeline) UJ/UL: relocation of the fire detectors Chamonix 2011 26 R 2 E Reality or Fata Morgana (Mirage)

What Would Happen If… Injection lines - High Energy Hadron fluence - Beam on the TED TI 2 – UJ 22/23 R 2 E shielding factor >10 TI 8 - UJ 87/88 shielding factor >10 Installed shielding improves situation by ~<= a factor of 10! Reminder: installed shielding for UJ 23, UJ 87, UJ 76, RR 73/77, UA 63/67 Chamonix 2011 27 R 2 E Reality or Fata Morgana (Mirage)

What Would Happen If… Injection lines - High Energy Hadron fluence - Beam on the TCD UJ 87 – no shielding UJ 87 –R 2 E shielding Installed shielding improves situation by ~<= a factor of 10! Reminder: installed shielding for UJ 23, UJ 87, UJ 76, RR 73/77, UA 63/67 Chamonix 2011 28 R 2 E Reality or Fata Morgana (Mirage)

Failure Rate Estimate Chamonix 2011 29 R 2 E Reality or Fata Morgana (Mirage)

Failure Rate Approach (1) Use the ‘Aging’ Radiation Level Table (2) Check for each equipment in each area (3) Apply the expected failure cross section Type Cross Section Hypothesis & Comments (DCCTs being excluded since redundant and low risk) Risk factor (DCCTs being excluded since redundant and low risk) Rad Status LHC 60 A-08 V LHC 120 A-10 V LHC 600 A-10 V LHC 4 -6 -8 k. A-08 V Inner-Triplet RPLA RPLB RPMB RPH RPxx [1 E-10. . 1 E-11] /cm 2 [5 E-9. . 1 E-11] /cm 2 [5 E-8. . 5 E-9] /cm 2 [2 E-8. . 1 E-9] /cm 2 Not Rad Tested • power part relatively safe, with some SEGR on some Power Mos. Fets • Converter more complex than 60 A (more components) • CPLD in Digital control board only 1 x CPLD, then not adding too high extra failure • FGC cross section @ 1 E-11 is correct Not Rad Tested • power part relatively safe, with some SEGR on some Power Mos. Fets • 5 x CPLD in Digital control board only 1 x CPLD, not adding too high extra failure • FGC cross section @ 1 E-11 is correct • 3 x DC-DCs unknown • AC-DC unknown but high AC voltage range • 48 x Power Mosfets used in 4 QLS Not Rad Tested • power part relatively safe, with some SEGR on some Power Mos. Fets • 8 x CPLD in Digital control board only 1 x CPLD, adding extra failure • FGC cross section @ 1 E-11 is correct • 1 DC-DC • no AC-DC Not Rad Tested • power part relatively safe, with some SEGR on some Power Mos. Fets • 8 x CPLD in Digital control board only 1 x CPLD, adding extra failure • FGC cross section @ 1 E-11 is correct • no AC-DC • Sigma delta + 1 CPLD • Additional Thyristor + 1 DCDC No High Risk. Well tested in CNGS. No high Risk since CERN Design, and very few critical or unknown components. A security hole remains on current lead protection (CPLD based). High Risk since some unknown integrated devices: 5 x CPLD + 1 x AC-DC + 3 x DC-DC in power part. High Risk since some unknown integrated devices + 8 x CPLD + 1 x DCDC in power part High Risk since some unknown integrated devices + 8 x CPLD 1 x DC-DC in power part + Inner-Triplet additional components with DCDC or CPLD Well tested. Converter not Tested under radiation. Estimation provided assuming no unexpected very high sensitive device included in unknown / untested parts, which would completely change the situation. • power part relatively safe, with some SEGR on some Power Mos. Fets • FGC cross section @ 1 E-11 is correct Chamonix 2011 30 (4) Result: failures per equipment/area R 2 E Reality or Fata Morgana (Mirage)

2011 Estimation EXPECTED FAILURES PER YEAR - LHC ALCOVES Failure Mode CV immediate dump and access 1 Pconv Opt Pconv Pess CRYO BPWIC Fire/ODH QPS 29 544 13 35 9 3 immediate dump Scheduled access 8 0. 0 15 0. 0 other 3 0. 0 9 8 CL heaters 14 13 4 0. 4 Pessimistic SUM MTBF [days] Chamonix 2011 VAC SURVEY Collim EN/EL TIMING REM RESET 4 19 10 12 614 19 50 39 1 19 7 9 0. 2 0. 6 7 0. 3 9 3 0. 0 Optimistic 98 19 35 30 IT 12 1. 3 BI RP 0. 1 5 0. 0 0. 9 Guess/Tested Ratio 2. 7 0. 02 0. 6 0. 4 R 2 E Reality or Fata Morgana (Mirage)

2012 Estimation EXPECTED FAILURES PER YEAR - LHC ALCOVES Failure Mode CV immediate dump and access 2 Pconv Opt Pconv Pess CRYO BPWIC Fire/ODH QPS CL heaters 48 912 immediate dump 22 60 14 5 Scheduled access 14 0. 1 26 0. 0 other 7 0. 0 16 14 24 22 SUM MTBF [days] Chamonix 2011 8 0. 7 2. 2 11 6 7 Pessimistic 0. 4 1. 1 12 0. 7 16 5 0. 0 Optimistic 166 33 60 52 IT VAC SURVEY Collim EN/EL TIMING REM RESET 21 2. 4 BI RP 0. 2 8 0. 0 1. 6 Guess/Tested SUM MTBF [days] Ratio 1031 33 86 68 0. 35 11 4 5 2. 6 0. 02 0. 5 R 2 E Reality or Fata Morgana (Mirage)

Relocation and Shielding Improvements No Additional + NOMINAL OPERATION Changes Shielding relocation UJ 14 UJ 16 1666 17 8 RR 13 RR 17 376 75 75 UJ 33 UJ 56 1 3 265 1 3 0 RR 53 RR 57 376 75 75 UA 63 UA 67 4 4 4 UJ 76 85 85 0 RR 73 RR 77 166 33 33 UX 85 b 8 25 2 0 0 2977 0. 12 8 25 2 0 0 594 0. 61 0 0 201 1. 82 UJ/RE 32 US 85 UW 85 UJ 23 UJ 87 SUM MTBF [d] Chamonix 2011 Most systems (QPS, MKS etc) excluded from UA 63/7 list as duct shielding can be added if necessary Only few equipments included in UW 85 list Power converters 120 A in UJ 23/87 classified as “other” SEE => no DUMP 33 R 2 E Reality or Fata Morgana (Mirage)

Nominal Estimation (After Shielding & Relocation) Failure Mode EXPECTED FAILURES PER NOMINAL YEAR - LHC ALCOVES CV immediate dump and access 2 Pconv Opt Pconv Pess CRYO BPWIC Fire/ODH QPS 108 2001 immediate dump 3 0 20 0 Scheduled access 39 0. 2 90 0. 0 other 21 0. 1 64 2 60 IT VAC SURVEY Collim EN/EL TIMING REM RESET 3 9 1. 6 BI RP 0. 1 0. 0 2 0. 0 4 2. 4 0. 0 0. 3 0 Optimistic Failure Mode immediate dump and access immediate dump Scheduled access other Chamonix 2011 CL heaters 34 SUM MTBF [days] 176 2. 1 22 52 29 17 7. 0 13 R 2 E Reality or Fata Morgana (Mirage)

Expected SEE Failure Rates (status 08/10) 2011 LHC Operation: Nominal LHC Operation: Best possible estimate today Uncertainties: LHC operation & machine behavior, radiation levels, equipment sensitivities -> see next slides 2011 will be at the edge (and above) and possibly show first limitations In order not to have problems with nominal LHC operation we would have to be wrong by a factor 500 -1000 ! Chamonix 2011 35 R 2 E Reality or Fata Morgana (Mirage)

Radiation Tests Chamonix 2011 36 R 2 E Reality or Fata Morgana (Mirage)

‘Only’ a selection could be tested! CNRAD, some pictures: The ‘Tower of Pisa’ Chamonix 2011 37 R 2 E Reality or Fata Morgana (Mirage)

Radiation Tests in 2011 CERN: q. CNRAD (Mar-Nov 2011) q Tests on Power converters, QPS equipment, Cryogenics q. H 4 IRRAD (start May) q Tests on Power Converters and EN/EL equipment. Outside CERN: q. PSI - Villigen (2011) q Agreement with PSI to get 1 weekend test per month; q Test of Amplifier, ADC buffers, and ADC for the PC redesign; q Continue calibration of the RADMONs. q. CEA – Valduc (Feb 2011) q Calibration of Pin. Diodes (RADMON) for the Displacement Damage measurements. q. TRIGA– Rome (or Prague facility) (2011) q RADMON memory calibration with a thermal neutron beam. Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)

What Equipment Is/Was Tested Cooling and Ventilation (H. Jena) Siemens S 7 -300, S 7 -200 Schneider Telemecanique Premium Warm Interlock Rack (P. Dahlen) PLC 315 F 2 DP, Ethernet controller 24 DI safety input modules, 2 x DO Relay modules, 2 x 32 DO modules IM 153. 1 - ET 200 M Boolean Processor - FM 352 -5 Ethernet (E. Sallaz) Three Ethernet Switches 3 Com 4400 Chamonix 2011 39 R 2 E Reality or Fata Morgana (Mirage)

What Equipment Is/Was Tested Fire Detectors (S. Grau and Team) 4 Detectors (different types) Collimation (G. Spiezia and Team) Full Rack with Drivers, I/O RIO National Instruments PXI MDC + PRS (ADC, DAC, FPGA card, power supply) Europa crate (custom electronic for LVDTs and Resolvers excitation/acquisition, power supply) Timing & Remote Reset (R. Chery) Chamonix 2011 40 R 2 E Reality or Fata Morgana (Mirage)

Benchmarks Chamonix 2011 41 R 2 E Reality or Fata Morgana (Mirage)

Operation & Normalisation Summary (Protons) In 6. 02 E+15 Dumped 5. 82 E+15 Lost in Machine 1. 99 E+14 Of Lost protons Collisions 2. 33 E+13 Elsewhere 1. 76 E+14 96. 70% 3. 30% 11. 73% 88. 27% Summary (Ions) In 7. 46 E+13 Dumped 6. 36 E+13 Lost in Machine 1. 10 E+13 Of Lost protons Collisions 3. 77 E+10 Elsewhere 1. 10 E+13 Chamonix 2011 85. 25% 14. 75% 0. 34% 99. 66% R 2 E Reality or Fata Morgana (Mirage)

IR 7 Benchmark - Loss Scenario 2010 Summary (Protons B 1) In 3. 53 E+15 Dumped 3. 44 E+15 Lost in Machine 9. 67 E+13 Of Lost protons B 1 Collisions 1. 17 E+13 Elsewhere 8. 50 E+13 97. 26% 2. 74% BLM ratio IR 7 / IR 3 TCSG. A 6 L 7. B 1 / TCSG. 5 L 3. B 1 TCSG. A 6 R 7. B 2 / TCSG. 5 R 3. B 2 Ratio % Loss in IR 7 3. 1 76 5. 6 85 12. 07% 87. 93% Summary (Protons B 2) In 2. 49 E+15 Dumped 2. 38 E+15 Lost in Machine 1. 02 E+14 Of Lost protons B 1 Collisions 1. 17 E+13 Elsewhere 9. 05 E+13 Chamonix 2011 95. 89% 4. 11% 11. 42% 88. 58% R 2 E Reality or Fata Morgana (Mirage)

IR 7 Benchmark - Loss Scenario 2010 Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)

Shielding Benchmark: © C. Adorisio, S. Roesler T. Nakamura, Journal of Nuclear and Radiochemical Sciences, Vol. 4, No. 2, pp. R 15 -R 24, 2003 Chamonix 2011 45 R 2 E Reality or Fata Morgana (Mirage)

IP 8/TI 8 Radiation Detectors 5 RM 07 S 5 RM 05 S © V. Boccone PMIL 8812 Current location not known in full detail of the Radmon and RAMSES positions PMIL 8811 Large Gradient: 4 -5 orders of magnitude PATL 8711 Chamonix 2011 5 RM 06 S 5 RM 08 S 46 R 2 E Reality or Fata Morgana (Mirage)

TED Loss: Rad. Mon Downstream © V. Boccone Chamonix 2011 47 R 2 E Reality or Fata Morgana (Mirage)

UJ 87: Loss on TCDIH. 87904© V. Boccone Chamonix 2011 48 R 2 E Reality or Fata Morgana (Mirage)

Point-8 Application Benchmark FLUKA/RAMSES benchmark Measured dose (m. Sv/h) @10. 6 MHz Ratio (meas/simu) PMI 8501 (UX 85) 24. 0 1. 1 PMI 8511 (UX 85) 120. 0 0. 8 PAT 8511 (US 85) 36. 7 0. 6 Detector US 85 UX 85 UW 85 FLUKA Simulations provide high energy hadron fluence, FLUKA/Rad. Mon benchmark dose and 1 Me. V Si equivalent in the LHCb cavern Detector Ratio according to the Phase-2 shielding implementation (FLUKA exp/measure) proposed in the R 2 E Project 8 LE 10 S 1. 6 8 LE 07 S 2. 0 8 LE 04 S 1. 6 8 LE 08 S 2. 2 Chamonix 2011 Very good agreement with PMIs and PATs RAMSES detectors Rad. Mons set at 3 V more difficult (at low count rates) Significant uncertainties to be considered (thermal neutron contribution, detector geometry, etc…) Uncertainty at least a factor of 2 R 2 E Reality or Fata Morgana (Mirage)

Rad. Mon/FLUKA @ CNRAD © D. Kramer, K. Roeed Very Complex Geometry Large Distances, ‘unknown’ materials, … Area 455 456 off-axis 467 line-of-sight 463 to TSG 4 451 Chamonix 2011 High energy hadron fluence Rad. Mon [cm-2/pot] 4. 3 e-9 1. 8 e-9 3. 7 e-10 1. 7 e-8 1. 8 e-7 Fluka/ [cm-2/pot] Rad. Mon 3. 2 e-9 0. 74 9. 2 e-10 0. 50 3. 0 e-10 0. 81 1. 9 e-8 1. 12 2. 1 e-7 1. 17 50 Dose Rad. Mon [Gy(Si)/pot] 1. 4 e-18 7. 4 e-19 1. 2 e-19 3. 3 e-18 2. 8 e-17 Fluka/ [Gy(Air)/pot] Rad. Mon 1. 9 e-18 1. 36 1. 1 e-18 1. 48 2. 0 e-19 1. 67 3. 9 e-18 1. 18 4. 2 e-17 1. 50 R 2 E Reality or Fata Morgana (Mirage)

Benchmark Simulations for TI 2/8 © V. Boccone Large Gradient Chamonix 2011 51 R 2 E Reality or Fata Morgana (Mirage)

Injection: TED – High-E Fluence © S. Roesler RADMON 1. 2 x 1010 cm-2 8 RM 08 S, 26. 5 cm behind dump 2 cm off beam axis FLUKA 0. 96 x 1010 cm-2 ± 3. 2% (scoring in a volume of 2 x 2 cm 3) protons: 9. 8% neutrons: 34. 1% pos. pions: 21. 9% neg. pions: 22. 4% others: 11. 8% (1. 03 x 1010 cm-2 ± 3. 2% scoring in a volume of 5 x 5 cm 3) Chamonix 2011 52 R 2 E Reality or Fata Morgana (Mirage)

Injection: TED – 1 Me. V N-Equiv. © S. Roesler RADMON 2 x 1010 cm-2 FLUKA 2. 1 x 1010 cm-2 ± 2. 5% (scoring in a volume of 2 x 2 cm 3) protons: neutrons: pos. pions: neg. pions: others: 4. 6% 81. 6% 5. 3% 5. 6% 2. 9% (2. 25 x 1010 cm-2 ± 1. 8% scoring in a volume of 5 x 5 cm 3) Chamonix 2011 53 R 2 E Reality or Fata Morgana (Mirage)

Injection: TED - Dose © S. Roesler RADMON 4. 73 Gy (Si) FLUKA 5. 0 Gy (air) ± 10% (scoring in a volume of 5 x 5 cm 3) Chamonix 2011 54 R 2 E Reality or Fata Morgana (Mirage)

TCDQ Losses 07 -09. 11. 2009 Chamonix 2011 55 R 2 E Reality or Fata Morgana (Mirage)

TCDQ Losses 07 -09. 11. 2009 Chamonix 2011 56 R 2 E Reality or Fata Morgana (Mirage)

TCDQ Losses 07 -09. 11. 2009 • • • ~3 x 1010 cm-2 high-E hadrons for 7 Te. V and 2. 6 x 1013 rough scaling: ~2 x 109 cm-2 at 450 Ge. V this results in ~4 x 105 per 5 x 109 shot We had about 50 (full) shots on the TCDQ -> ~2 x 107 expected 5. 6 x 107 measured at the tunnel location (~30 counts!) In the UA, the monitor is set to 3 V (factor of 10 more sensitive) -> nothing measured -> confirms the expected attenuation factor of ~1000 Chamonix 2011 57 R 2 E Reality or Fata Morgana (Mirage)

Remarks Chamonix 2011 58 R 2 E Reality or Fata Morgana (Mirage)

Remark 0 In all our estimates and predictions we try being as ‘accurate’ as possible (conservative but fairly close to reality), thus uncertainties strictly apply in all directions NO BIG SAFETY MARGINS LEFT Safety factors usually used in the field of SEE estimates are in the order of 10 -100 and more (depending on the application: space missions, airplanes, …) This we can’t do (afford) at the LHC, even after the mitigation actions are applied Chamonix 2011 59 R 2 E Reality or Fata Morgana (Mirage)

Remark 1 Uncertainties go in both directions, thus as we try to be ‘as accurate as possible’, it’s not necessarily conservative! Example: LHCb (UX/US 85: measurements versus ‘expectation’): Updated FLUKA Simulations, ‘old’ detector X(PMI) RADMONS X(PAT) PMI sees up to 30 m. Sv/h at a luminosity of 1031 cm-2 s-1 (expected: 10 m. Sv/h) PAT sees a few m. Sv/h (expected: less than one) Rad. Mons: see equivalent counts, however less statistics (expected: first count only) © M. Calviani, C. Theis Reason: old detector geometry and magnetic field (in work!) Other areas: low-energy neutrons not to be forgotten! (then our estimates would be even less conservative) Chamonix 2011 60 R 2 E Reality or Fata Morgana (Mirage)

Remark 2 In our estimations, we have to be ‘wrong’ by a factor of 5001000 (only in one direction) in order to reach acceptable SEE induced failure rates for the LHC (if nothing is done)! Nominal LHC Operation: We’re looking for a MTBF of 150300 h what concerns ‘acceptable’ SEE induced problems (tunnel equipment not included!) All test results we get, all the analysis from the available early monitoring (some shown today), make this impossible SEE induced problems happened already once or twice (WIC in injection line, QPS during ‘dirty’ injection [unlikely]) Chamonix 2011 61 R 2 E Reality or Fata Morgana (Mirage)

Remark 3 Seeking maximum LHC performance, we’re bound to fit all R 2 E related work in the shutdown planning. Operational periods between long shut-downs will be challenging and require the possibility to react in case of problems Needed: preparation of patchsolutions (equipment level) radiation tolerant developments (R&D for power converters) strong Rad. WG radiation test possibilities at CERN Chamonix 2011 62 R 2 E Reality or Fata Morgana (Mirage)

Left of IR 7 “Bad Ion Fill (1488)” - Beam intensity for ions -> ‘equivalent’ of Ions/Protons to be scaled (208/82) -> ~9 x 101 - At the same time, this corresponds to 68 bunches, thus about 10% of nominal. - Rad. Mons nicely ‘counting’ at uneven MQs (9 -17), below Interconnects - this dataset: RM 9 -11 set to 5 V, RM 13 -17 set to 3 V - this fill: with th. neut. ratio of 1: ~1 E 6 cm-2 (@Q 9), ~1 E 7 cm-2 (@Q 17) >20 Me. V - VERY simplified scaling to nominal (7 Te. V) and 1 month of Ions: ~1 E 10 cm-2 Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)

Left of IR 7 “Good Proton Fill (1450)” - To be compared with ~9 x 1011 as equivalent number of protons of previous ion fil - Again running at about 10% of nominal - Rad. Mons have low statistics (2 counts in critical locations) - Voltage settings are the same as before RM 9 -11 @5 V, RM 13 -17@3 V - VERY simplified scaling to nominal (7 Te. V) and 1 LHC year: ~5 E 9 cm-2 >20 Me. V -2/year) Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage) (nicely agreeing with the simulation estimates: 1 E 9 -1 E 10 cm

Right of IR 7 “Bad Ion Fill (1488)” - Beam intensity for ions -> ‘‘equivalent’ of Ions/protons to be scaled (208/82) -> ~9 x 10 - At the same time, this corresponds to 68 bunches, thus about 10% of nominal. - Rad. Mons nicely ‘counting’ at different locations (mainly Q 9 and Q 11) - this dataset: RM 9 -11 set to 5 V, RM 13 -17 set to 3 V - this fill: with th. neut. ratio of 1: ~7 E 6 cm-2 (@Q 9) , ~1 E 7 cm-2 (@Q 11) >20 Me. V - VERY simplified scaling to nominal (7 Te. V) and 1 month of Ions: ~1 E 10 cm-2 Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)

Right of IR 7 “Good Proton Fill (1450)” - To be compared with ~9 x 1011 as equivalent number of protons of previous ion fill - Again running at about 10% of nominal - Rad. Mons have VERY low statistics (only one has 1 count) - Voltage settings are the same as before RM 9 -11 @5 V, RM 13 -17@3 V - VERY simplified scaling to nominal (7 Te. V) and 1 LHC year: ~4 E 9 cm-2 >20 Me. V (still nicely agreeing with the simulation estimates: 1 E 9 -1 E 10 cm-2/year) Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)

Ions: Preliminary ‘Conclusion’ q Ions (at least during ‘bad fills’) lead to high losses in the DS/ARC > ratio Ion/Proton based on Radmon signal and normalized to proton equivalent is in the order of 600 q Scaling for nominal q Protons: coherent with simulation estimates (bad statistics, can be improved by looking at more fills, ongoing) -> levels might reach up to 1 E 10 cm-2 per year q Ions: similar levels can be ‘achieved’ within one month of ions in case ‘bad fills’ would become ‘nominal’ q QPS crates affected from Q 9 -Q 17, firmware upgrade possibly needed soon (under investigation - R. Denz) q Quickly checked for P 3, and DSs adjacent to experiments -> only little counts so far, levels significantly lower but more statistic/time needed to give some estimates Chamonix 2011 R 2 E Reality or Fata Morgana (Mirage)