1 / 54

LEP and FCC- ee Beam Polarization and Energy Calibration

This comprehensive work delves into the beam polarization and energy calibration processes at LEP from 1989 to 2000, presenting groundbreaking observations and detailed analyses. It covers topics such as first observations of transverse polarization, resonant depolarization, spin tunes at the Z peak, and harmonic spin matching techniques. The text offers insights into depolarizing effects, polarization measurement methods, and empirical spin matching practices for enhancing polarization at various energy levels. It also explores the impact of energy spread on polarization and the challenges involved in maintaining high levels of polarization. The study emphasizes the importance of accurate energy calibration for precise measurements of particle properties such as mass and width.

wwatkins
Download Presentation

LEP and FCC- ee Beam Polarization and Energy Calibration

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. LEP and FCC-ee BeamPolarization and Energy Calibration Keil Jowett Wenninger Buon Hofmann Placidi Assmann Koutchouk and manyothers!

  2. LEP (1989-2000) first observation of P in 1990 first resonantdepolarization in 1991 =310 minutes at 46 GeV Spin tune at the Z peak : 103.5 The scan points 101.5 / 103.5 / 105.5 are perfect optimum for Z widthmeast Spin tune for W threshold 183.5

  3. Self polarization: a priori expectations were O(5-20%) depending on imperfections (Koutchouk) At beginning of LEP a layer of nickel in beam pipe caused a strongxycoupling– taken out in 1990 Polarizationmeasurementwith a back scattered frequencydoubledNd-Yag laser (528 nm)

  4. in FCC-eeemittances are muchsmaller and profile shouldbenarrower. displacementupon inversion of circular light is a measure of the vertical polarization

  5. depolarization by a staticharmonicbump and polarizationrise.

  6. Resonantdepolarization, 1991 variation of Rffrequency to eliminatehalfinteger ambiguity

  7. in 1995 wepushed the measurement to «high precision» (100 keV) spin tune

  8. ... and no e+ polarimeter! shouldrevisit the uncertainty and the method to understand how muchbetter wecan do. Also how practicalisit to co-exist ‘polarized single bunches’ with ‘top-off injection’

  9. Spin corrections In order to disturb the experiments as little as possible wecompensated the solenoids vertical BL=10Tm experimental verification

  10. Whatwe know from LEP on depolarizingeffects http://dx.doi.org/10.1063/1.1384062 AIP Conf. Proc. 570, 169 (2001) Spin 2000 conf, Osaka

  11. canbeimproved by increasing the sum of |B|3 (Wigglers) canbeimproved by ‘spin matching’ the sources of depolarizationcanbeseparatedintoharmonics (the integerresonances) and/or into the components of motion: receipes: -- reduce the emittances and vertical dispersion   thiswillbedoneat TLEP to reducebeam size! -- reduce the vertical spin motion n harmonic spin matching -- do not increase the energyspread

  12. examples of harmonic spin matching (I) DeterministicHarmonic spin matching : measureorbit, decompose in harmonics, cancel components near to spin tune.  NO FIDDLING AROUND. This workedverywellat LEP-Z and shouldworkevenbetterat TLEP-Z if orbitismeasuredbetter.

  13. NB high level of polarizationrequires constant adjustments to remain.

  14. examples of harmonic spin matching (II) When all elsefails,empirical spin matching: excite harmonics one by one to measuredirectlytheireffect on polarization and fit for pole in 4-D space. Here 8% polarizationat 61 GeV.

  15. The energyspreadreallyenhancesdepolarization

  16. effect of energyspread on Polarization in a given machine wasstudiedusing the dampingwigglers

  17. These observations were consistent with the hypothesisthat energy spread drives the synchrotron resonances. Wewere able to keep the polarization as long as E < 56 MeV. In a larger ring the energy spread issmaller But the polarization time ismuch longer.

  18. E  Eb2 /  The good news isthatpolarization in LEP at 61 GeV corresponds to polarization in TLEP at 81 GeV  Good news for MWmeasurement (see E. Gianfelice’spresentation)

  19. Polarization in collisions CERN-SL-26-021 v = 101.5 11-1994 y = 0.04

  20. Beambeamdepolarizationis not very large. < 0.65 Alone, and for one exp. itwould have limited the polarization to ~55% Predictabilityislow.

  21. Longitudinal polarization at LEP? schemedeveloped in 1988 see A.B. CERN-PPE-93-125

  22. measuring the polarizationasymmerty: requires (continuous) depolarization of selectedbunches e+ and e- polarimeter bb tune shift not too large.  probably at least factor 4 loss in Luminosity. Did not happenunder pressure from LEP2, then LHC

  23. Energy calibration for physics The main goal was to measure the Z mass and width. in 1993 and 1995 data were taken at threeenergy points withspin tune 101.5/103.5/105.5 whereresonantdepolarizationcouldbeapplied. (To mygreat frustration the solenoid compensation was not applied and wecould not measure the polarizationduring data taking. ) The energymeasurementsweretaken at the end of the physicsfills, whichwerevery long: LEPI wasbeam-beamlimited and the luminosity wouldremain constant -- or eventimproving – for up to 10 hours. The energy calibrations showed important residuals.

  24. Energy calibrations at LEP Intrinsic calibration wasestablished to bebetterthan 100 keV per beam at the time of the measurement. Long termstabilitywasfound to not bebetterthan 5 MeV It wasfound and demonstratedthatitwasaffected by gound motion and electriceffects Ground motion: -- tides -- level of the lake -- rain on the Jura Electric effects: -- trains

  25. stillafter corrections for tidestherewere large fluctuations. a drift during a long MD fill -- beyondtides – was cause of concern for 2 years….

  26. in 1995 NMR probes wereinstalled insidetwomagnets on the ring and the observations werestriking: the fieldrisesduring the fills! in 1996 14 more NMRswereinstalled  two per octant

  27. Measurements of the currentflowing on the LEP beam pipe showed a strange correlation pattern as if current flowedfrom point 1 to point 6 in the two arcs at the same time The culpritwasfound to be the TGVs

  28. by 1999 wehad an excellent model of the energy variations… but wewere not measuring the Z mass and widthanymore – wewerehunting for the Higgsboson!

  29. Final table of errors on the LEP center of mass energies see M. Koratzinos for analysis

  30. Conclusions and outlook for FCC-ee Polarizationwas ‘easy’ to achieve at LEPI and couldbeimproved to ~55% by means of harmonic spin matching. Deterministic (based on orbitmeasurements) or Empirical (based on polarizationitself) Spin bumpswere+ bumps and workedwell (do not generatemuch dispersion) Polarizationwas possible as long as beamenergy spread < ~55 MeV 2. resonantdepolarization (RD) calibration couldbeachieved withprecision of < 100keV on beamenergy 3. for physics use, manyproblemsarosebecause RD was not doneduringphysics -- tides, ground motions, trains etc… had to bediscovered and corrected a posteriori. Time consuming and loss of information. -- in mypersonal opinion thisis a result of excessive separation betweenaccelerator and particlephysicsworlds -- but alsobecauseprocedurewasdelicate and not automatized.

  31. G. Wilkinson

  32. For FCC-ee the energy calibration withresonantdepolarizationisgivennaturally an essential ingredient for the energy calibration needs to keep a few 100 bunchespolarized and not interacting -- sincetop-upwithunpoarizedbeamswould destroy polarization 4. needspolarimeters for both e+ and e- 5. life shouldbe made easier by smallerbeams and modern beam diagnostics 6. continuous monitoring of B field and polarizationneeded 7. probablycomplementary compensation of solenoidsneeded 8. Given the high statistics longitudinal polarizationis not required -- the same information canbeobtainedotherwise 9. list o requiredequipemnt to bbeestablished. 10 possible issue with RD at the Higgs if v isinteger. 11. a complementarymeasurement at similarlevel of precisioncouldbeuseful if simple enough

  33. Use of polarizationwigglersat TLEP

  34. FOREWORD How the sausagewas made... In order to evaluate the effect of wigglers and top-up injection on TLEP polarization performance, I have generatedtwospreadsheets 1. the first one calculates the energypread and polarization time in TLEP assuming a bending radius of 10km for a circumference of 80km, and the presence of the 12 polarizationwigglersthatwerebuilt for LEP as calculated in LEP note 606 (Blondel/Jowett) 2. the second one folds the achivedpolarization performance with top up injection, given the luminosity life time and the regular injection of unpolarizedparticles. The variable parameters are -- B+ : field in the positive pole of the wigglers -- beamenergy -- luminositylifetime -- and of course Jx but I have refrained to playwithit. (one wouldwantit as small as possible)

  35. Energyspread (Jx=1) LEPTLEP beamenergy sigma(E) tau_P sigma(E) tau_P 45 GeV no wiggs 32 MeV 5.5 hrs 18 MeV 167 hrs 45 GeVwigglers 46 MeV 2.4 hrs 58 MeV 12 hrs 55 GeV no wiggs 48 MeV 1.96 hrs 26 MeV 61 hrs 61 GeV no wiggs 59 MeV 1.1 hrs 33 MeV 36 hrs 81 GeV 58 MeV 8.9 hr   annoyingly: withwigglersat TLEP, the energyspreadislargerthanat LEP, for a givenpolarization time. considersomewherebetween 48 and 58 MeV as maximum acceptable for energyspread*). Take 52 MeV for the sake of discussion. Note thatwigglersmakeenergyspreadworsefasterat TLEP (dampingisless)  There is no need for wigglersat 81 GeV. *) The absolute value of energyspread corresponds to an absolute value of spin-tune spread

  36. Hypothetical scenario Insert in TLEP the 12 Polarizationwigglersthathad been built for LEP (B-=B+/6.25) Use formulaegiven in TLEP note 606 to determine as a function of B+ excitation 1. the energypreadE 2. the polarization time P then set an uppperlimit on energyspread... and seewhatpolarization time weget for 10% polarization the time isP eff=0.1 P for 52 MeV energyspreadat TLEP Z wegetP = 15hrs or P eff=90 minutes -- lose 90 minutes of running , thencandepolarize one bunchevery 10 minutes if we have 9 ‘single bunches’ per beam. (willkeep a few more to be sure) Changing the wigglers (e.g. more, weaker) makeslittledifference B- B+ • B-

  37. Jowett: were not easy to use (orbitdistortions) and shouldprobablybebetterdesigned

  38. E(GeV) hrs BWiggler(T)

  39. Synchroton radiation power in the wigglers Synchrotron radiation power by particles of a givenenergy in a magnet of a givenlength scales as the square of the magneticfield. The energylossper passage through a polarizationwigglerwascalculated in LEP note 606 (seenext page). It is 3.22 MeV per wiggler or 38.6 MeV for the 12 wigglers. At LEP the energyloss per particle per turnis 117 MeV/turn in the machine with no wiglers and becomes 156 MeV per turnwith the 12 wigglersat full field. Fromthisitfollowsthat in the machine running at 45 GeV and wigglersat full field the radiation power in the wigglerswould have been 25% of the total power dissipatedaround LEP. In TLEP now, the energyloss per turn in the ring is 36.3 MeV while, in the wigglersat 0.64T, the energyloss in the wigglersisapprox. a quarter of the above or 9.4 MeV. The fraction of energylost in the wigglersisthen 9.4/(36.3+9.4) or 21%. For a total SR power of 100MW, 21 MW will go in the wigglers.

  40. ENERGY LOSS PER PARTICLE PER WIGGLER 3.22MeV at 45 GeV

  41. TOTAL ENERGY LOSS PER TURN PER PARTICLE WITH 12 WIGGLERS

  42. Preliminary conclusions: 1.  the wigglersincreaseenergyspread in TLEP fasterthan in LEP 2.  a workable point canbefound for the ‘energy calibration mode’ at the Z pole or W threshold, assuming no better performance than in LEP for depolarizingeffects. 3.  thingsshouldgetbetterwithlower emittance and lower vertical dispersion. 4. !!!Averycareful design of the SR absorbersisrequired, as the SR power in the wigglersisvery large (20% of the total in the ring) !!! 5. reducing the luminosity for polarizationrunscanbeenvisaged, since the statisticalprecision on mZ and Z isverysmall (<10 keV) and probably smallerthansystematics.

More Related