LC availability Simulation done for the LC comparison task force. Tom Himel SLAC

Transcription

1 LC availability Simulation done for the LC comparison task force SLAC

2 Committee Members (chair) Paul Czarapata Helen Edwards Markus Huening Nan Phinney Marc Ross 2

3 Contents Brief review of our approach to estimating an acceleratorõs availability Comparison of Warm 2 tunnel, Cold 2 tunnel and Cold 1 tunnel designs Positron source effects 3

4 Overall plan of Attack Write a simulation that given the MTBFs, MTTRs, numbers and redundancies of components, and access requirements for repair can calculate average availability and the integrated luminosity per year. Luminosity is mostly either design or zero in this simulation. Collect data on MTBFs and MTTRs of components in existing machines to guide our budgeting process Make up a reasonable set of MTBFs that give a reasonable overall availability. We allowed 25% downtime total. 10% was kept as contingency and MTBFs were tuned so the simulation gave 15% downtime. Iterate as many times as we had time for (one and a half iterations were done) to minimize the overall cost of the LC while maintaining the goal availability 4

5 The Simulation includes: Effects of redundancy such as 21 DR kickers where only 20 are needed in the cold design or the 3% energy overhead in the warm design Some repairs require accelerator tunnel access, others canõt be made without killing the beam and others can be done hot. Time for radiation to cool down before accessing the tunnel Time to lock up the tunnel and turn on and standardize power supplies Recovery time after a down time is proportional to the length of time a part of the accelerator has had no beam. Recovery starts at the injectors and proceeds downstream. Manpower to make repairs can be limited. 5

6 The Simulation includes: Opportunistic Machine Development (MD) is done when part of the LC is down but beam is available elsewhere for more than 2 hours. MD is scheduled to reach a goal of 1-2% in each region of the LC. The linacs and DRs are modeled in detail down to the level of magnets, power supplies, power supply controllers, vacuum valves, BPMs É Due to time constraints other regions were simulated as monolithic units. Non-hot maintenance is only done when the LC is broken. Extra non-essential repairs are done at that time though. Repairs that give the most bang for the buck are done first. 6

7 The Simulation includes: PPS zones are handled properly (e.g. can access the warm (but not the cold) linac when beam is in the DR. It assumes there is a tuneup dump at the end of each region. (Important design requirement which should not be forgotten.) Kludge repairs can be done to ameliorate a problem that otherwise would take too long to repair. Examples: Tune around a bad quad in the cold linac or a bad quad trim in either damping ring or disconnect the input to a cold power coupler that is breaking down. During the long (3 month) shutdown, all devices with long MTTRÕs get repaired. 7

8 Powersupplies -bend beamline e+ linac broken 3luminosity mult Powersupplies -quad beamline e+ linac broken 253 luminosity mult Powersupplies -cor beamline e+ linac broken 379 luminosity mult PS controler-bend beamline e+ linac broken 3 luminosity mult PS controler-quad beamline e+ linac broken 253 luminosity mult PS controler-cor beamline e+ linac broken 379 luminosity mult Vac Mech device beamline e+ linac broken 6 luminosity mult VacP beamline e+ linac broken 596 luminosity mult The Full Components Sheet 8 VacV beamline e+ linac broken 4 luminosity mult Waterpumps beamline e+ linac broken 3 luminosity mult Water beamline e+ linac broken 3 luminosity mult Flow Switch beamline e+ linac broken 6 luminosity mult Cavities cav ity e+ linac degrade 7152 e+ energy overhead add -12 1E Cavities cav ity e+ linac broken 7152 e+ energy overhead add E Cavity tuner cavity e+ linac broken 7152 e+ energy overhead add Cavity piezo tuner cavity e+ linac broken 7152 e+ energy overhead add powercouplerdisc overhead add LLRF cavity e+ linac broken 7152 e+ energy 28.8 overhead add powercouplerdisc powercoupler coupler e+ linac degrade 7152 e+ energy powercoupler coupler e+ linac broken 7152 e+ energy overhead add powercouplerdisc powercouplerdisc coupler e+ linac disc 7152 e+ energy overhead add E couplerinterlocks coupler e+ linac broken 7152 e+ energy overhead add VacP coupler e+ linac broken 596 e+ energy overhead add As you can ÒseeÓ there 6.3 1cryo vac enclosure cryo segment e+ linac broken 4 luminosity mult insulating vacuump cryo string e+ linac leak 59.6 e+ energy overhead add cryo JT valve cryo string e+ linac broken 59.6 e+ energy overhead add BPMs diagnostic e+ linac broken 300 luminosity mult laserwires diagnostic e+ linac broken 12 luminosity mult Kickerpulser diagnostic e+ linac broken 1 luminosity mult wires diagnostic e+ linac broken 0 luminosity mult is a 0 lot 0 0 of 0 0 data to 0 0 get 4 Kicker diagnostic e+ linac broken 1 luminosity mult Klystrons firstklystron e+ linac broken 5 luminosity mult pulse transformers firstklystron e+ linac broken 5 luminosity mult Modulators firstklystron e+ linac broken 5 luminosity mult pulse cables firstklystron e+ linac broken 5 luminosity mult klys pre-amp firstklystron e+ linac broken 5 luminosity mult VacG/Ctrl firstklystron e+ linac broken 5 luminosity mult right VacP firstklystron e+ linac broken 5 luminosity mult Waterpumps firstklystron e+ linac broken 2 luminosity mult Water firstklystron e+ linac broken 2 luminosity mult Flow Switch firstklystron e+ linac broken 6 luminosity mult Electrical ->0.5 firstklystron e+ linac broken 2 luminosity mult Klystrons klystron e+ linac broken 293 e+ energy overhead add pulse transformers klystron e+ linac broken 293 e+ energy overhead add Modulators klystron e+ linac broken 293 e+ energy overhead add pulse cables klystron e+ linac broken 293 e+ energy overhead add klys pre-amp klystron e+ linac broken 293 e+ energy overhead add VacG/Ctrl klystron e+ linac broken 293 e+ energy overhead add I 0 can guarantee that it 5VacP klystron e+ linac broken 293 e+ energy overhead add Waterpumps klystron e+ linac broken 146 e+ energy overhead add Water klystron e+ linac broken 146 e+ energy overhead add Flow Switch klystron e+ linac broken 438 e+ energy overhead add Electrical ->0.5 klystron e+ linac broken 146 e+ energy overhead add controls backbone sector e+ linac broken 298 luminosity mult timing sec tor e+ linac broken 305 luminosity mult is NOT all correct Electrical -.05<<0.5 Utility power e+ linac broken 305 luminosity mult Bends beamline e+ DR broken 216 luminosity mult Quads beamline e+ DR broken 849 luminosity mult Sextupoles beamline e+ DR broken 312 luminosity mult Cors beamline e+ DR broken 629 luminosity mult 0 1E quad orcor Wigglers beamline e+ DR broken 90 luminosity mult Kickers -injection beamline e+ DR broken 21 e+ DR inj kick add Kickers -extraction beamline e+ DR broken 21 e+ DR extkick add Powersupplies strings beamline e+ DR broken 36 luminosity mult Powersupplies Cors beamline e+ DR broken 629 luminosity mult quad orcor PS controlerex.cor beamline e+ DR broken 36 luminosity mult WeÕve checked it 1.2 many PS controler-cor beamline e+ DR broken 629 luminosity mult quad orcor Vac Mech device beamline e+ DR broken 4 luminosity mult VacP beamline e+ DR broken 2048 luminosity mult VacV beamline e+ DR broken 125 luminosity mult Waterpumps beamline e+ DR broken 6 luminosity mult Water beamline e+ DR broken 6 luminosity mult Flow Switch beamline e+ DR broken 12 luminosity mult times, and an Cavities cav ity e+ DR broken 12 e+ DR RF HV add E LLRF cavity e+ DR broken 12 e+ DR RF HV add powercoupler coupler e+ DR broken 12 e+ DR RF HV add couplerinterlocks coupler e+ DR broken 12 e+ DR RF HV add VacP coupler e+ DR broken 24 e+ DR RF HV add insulating vacuump cryo module e+ DR leak 4 e+ DR RF HV add cryo vac enclosure cryo module e+ DR broken 4 e+ DR RF HV add individual error has 1 BPMs diagnostic e+ DR broken 1251 luminosity mult laserwires diagnostic e+ DR broken 2 luminosity mult Kickerpulser diagnostic e+ DR broken 0 luminosity mult wires diagnostic e+ DR broken 0 luminosity mult Kicker diagnostic e+ DR broken 0 luminosity mult Klystrons klystron e+ DR broken 3 e+ DR RF HV add Modulators klystron e+ DR broken 3 e+ DR RF HV add minor effects klys pre-amp klystron e+ DR broken 3 e+ DR RF HV add VacG/Ctrl klystron e+ DR broken 3 e+ DR RF HV add VacP klystron e+ DR broken 6 e+ DR RF HV add Waterpumps klystron e+ DR broken 3 e+ DR RF HV add Water klystron e+ DR broken 3 e+ DR RF HV add Flow Switch klystron e+ DR broken 6 e+ DR RF HV add Electrical ->0.5 klystron e+ DR broken 3 e+ DR RF HV add controls backbone sector e+ DR broken 3 luminosity mult timing sector e+ DR broken 3 luminosity mult Electrical -.05<<0.5 Utility power e+ DR broken 3 luminosity mult total down e+ DR dummy 0 luminosity mult E totaldown #access permonth e+ DR dummy 0 luminosity mult E #access permonth 8

9 Results

10 Convention vs Undulator e+ Source Undulator e+ source has much more downtime especially during commission period Not due to actual source reliabilities (which werenõt modeled in detail and were assumed to be the same.) Completely due to undulator source needing well tuned high energy electrons. Prevents doing scheduled MD in e.g. e- linac and e+ DR simultaneously During downtime recovery cannot tune e+ system until e- linac is tuned. CanÕt do opportunistic MD in e+ system while e- is down or being tuned. 10

11 Version A (warm 2 tunnel) MTBFs

12 Version B (cold 2 tunnel) MTBFs

13 Warm, downtime by region 13

14 Warm DR+Linac, downtime by system 14

15 Summary The fact that an undulator positron source requires well tuned high energy electrons before positrons can be produced significantly reduces the integrated luminosity of a LC. For example in the warm LC after a few years of running, the luminosity integrated in a year would be 18% less for an undulator positron source than for a conventional one. During commissioning, it would be far worse; a factor of two less luminosity would be integrated. There is not a great difference between warm and cold 2 tunnel designs. Both are very large and complex accelerators where significant effort and expense will be needed to make them reliable enough. If all components had the same reliability, the cold 1 tunnel design would have a downtime of 25% instead of the 15% of the 2 tunnel design. The required improvements to Mean Time To Failures (MTBFs) of many components and the necessary 8% increase in the energy overhead needed to recover the 15% availability are costly. This cost should be compared to the amount saved in building one tunnel instead of two to see if it is truly worthwhile. The effects of Mean Time To Repair (MTTR) and the speed of recovery from a downtime are quite significant. It will be worth considerable design effort to keep these small. Success at this could reduce the requirements on the MTBFs