html5-img
1 / 31

The role of the PS/SPD in the trigger

The role of the PS/SPD in the trigger. Míriam Calvo , Hugo Ruiz. 5 October 2010. Introduction. http://indico.cern.ch/getFile.py/access?contribId=25&sessionId=5&resId=1&materialId=slides&confId=67047 Are the PS/SPD necessary in the LHCb Upgrade ? Now , their role at L0 is :

tashya-blake
Download Presentation

The role of the PS/SPD in the trigger

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. The role of the PS/SPD in thetrigger Míriam Calvo, Hugo Ruiz 5 October 2010

  2. Introduction • http://indico.cern.ch/getFile.py/access?contribId=25&sessionId=5&resId=1&materialId=slides&confId=67047 • Are the PS/SPD necessary in theLHCbUpgrade? • Now, their role at L0 is: • Confirm ECAL clusters as electromagnetic (PS) • Distinguish/e (SPD) • Photonconversion ~40% (with M1) • SPD multiplicity as GEC, luminosity monitor, ... • Fortheupgrade, thecurrent plan isto use aninteractiontriggerwithout PS/SPD. • Whatwouldbelosingbydoingthat? • Stillusefulfor PID? (notcovered in thistalk)

  3. MC samples • MC 2010 • Upgrade ML 2.1 • (Stillincludes M1) • 7 TeV per beam • MagDown, 25 ns • Lumi 2, 5, 10, 20 (·1032 cm-2s-1) • Non-empty BX frequency ~12, 22, 28, 30 MHz • 100k MinBias, 35-50k Bs2PhiGamma, 35-50k Bd2Kstareeevents per luminosity • Strippingselectionsusedforsignalsamples • pT ()>2 GeV • pT (e)>300 MeV

  4. Rate MinBias 4 Lumi 2 (12 MHz of vis. X-sings) Lumi 5 (22 MHz) SPD/PS OFF em SPD/PS ON em SPD/PS ON electron SPD/PS ON photon off em e  Lumi 20 (30 MHz) Lumi 10 (28 MHz)

  5. Use of aninteractiontrigger (provideseedsfor HLT algorithms) to reduce the input of the EFF to 5 MHz. • Reduce electromagneticcandidatesto 1 MHz (whichistheacceptablerate?). • Withoutelectron/photondistinction. • Thisispossiblewith a cut of ETfor ECAL clusters (e.g. 3 GeV at 1033 cm-2s-1). ~50% increase of ratewithout PS/SPD Thus, number of T-stationsconfirmationneeded in HLT Ratio rateem OFF/ON (~sameforanyL) Anotherpoint of view, PS rejects 1/3 of ECAL clusters (hadrons, somephotons?)

  6. Rate vs luminosity Rate ECAL candidates (SPD/PS em OFF) Rateemcandidates (SPD/PS em ON) ET>2GeV 2.5 3 3.5 4 ET>2GeV 2.5 3 3.5 4 i.e.without SPD/PS i.e.with SPD/PS Lowerrateswith more loose ETcuts

  7. Number of candidatesper acceptedevent 7 lumi2 lumi5 SPD/PS ON electron SPD/PS ON photon SPD/PS ON em SPD/PS OFF em lumi20 lumi10

  8. As shown, increasedtriggerrateduetoincreasedrate of visible collisions at highluminosity, as expected. • Butnumber of candidates per eventapproximatelythesame at high ET. Ratio em OFF/ON ET>2GeV 2.5 3 3.5 4 without SPD/PS Increase of thenumber of candidates per acceptedevent <5% without PS/SPD Number of candidates vs L is flat at high ET

  9. Bs

  10. 10 Efficiency, photons Bsfg lumi2 5.2% of generated events accepted by the stripping selection lumi5 5.0% evts off em  e Efficiency of L0 onstrippedevents (pT()>2 GeV) lumi10 lumi20 4.6% evts 4.0% evts

  11. 11 Efficiency, photons Bsfg lumi5 SPD/PS OFF em SPD/PS ON em SPD/PS ON electron SPD/PS ON photon lumi2 photons Convertedphotons Using only ECAL clusters matched to MC signal g(< 60 cm) lumi10 lumi20

  12. PS kills 20% of em (photon) clusters (higherefficiencywith no PS requirement) Ratio efficiencyem OFF/ON Bsfg MC matchedclusters Ratio efficiencyphoton/em SPD identifies as photon 55-60% of keptemclusters (slightlylower at lumi 2·1033, probablyduetohigher pile-up)

  13. Efficiency (photons) vs rate 13 lumi2 lumi5 2 GeV off 2.25 GeV em  Highestefficiencywithout PS/SPD for a givenrate lumi10 lumi20 3 GeV 2.5 GeV 3.75 GeV 3.25 GeV 2.5 GeV

  14. Current L0 photontrigger can beimproved (for non-convertedphotons) if SPD=0 and no requirementonthe PS energy Fordifferent PS energycuts, withoutthe <=2 cellsrequirement off em PS energy in MeV when SPD=0, higherefficiencyif PS>=0 ~ currentvalue (10 MIPs) 

  15. Photon candidates/second vs efficiency 15 lumi2 lumi5 off 1 M em  Lesscandidates per second at a givenefficiencywithout PS/SPD lumi10 lumi20 2.5 GeV 2 GeV 3 GeV

  16. B0K*e+e-

  17. Efficiency, electrons 17 B0 K*e+e- lumi2 lumi5 5.2% of generated events accepted by the stripping selection 5.1% evts em off e  Photontriggers (notrelatedtosignalelectronsclusters) pT(e)>0.3 GeV in stripping lumi20 lumi10 3.6% evts 4.6% evts

  18. Efficiency, electrons 18 B0 K*e+e- SPD/PS OFF em SPD/PS ON em SPD/PS ON electron SPD/PS ON photon lumi2 lumi5 off e em  Using only ECAL clusters matched to MC e (< 60 cm)  photontriggersdisappear lumi20 lumi10

  19. Ratio efficiencyem OFF/ON PS kills 5% of eventswithem (electron) clusters B0 K*e+e- MC matchedclusters SPD identifies as electron 99% of remainingsignalelectronclusters Ratio efficiencyelectron/em

  20. Efficiency (electrons) vs rate 20 lumi2 lumi5 em off e 2.25 GeV Highestefficiencywith PS/SPD for a givenrate lumi20 lumi10 2.25 GeV 2.5 GeV 3 GeV 3 GeV 3.25 GeV 3.75 GeV

  21. Electroncandidates/secondvs efficiency 21 lumi5 lumi2 off 1 M em e Lesscandidates per second at a givenefficiencyfor L0 electronswith PS/SPD lumi10 lumi20 3 GeV 2.25 GeV

  22. Conclusions • Results are similiar at anyluminosity. • Maindifferenceistherate of visible ppcollisions. • Usingthe PS/SPD toconfirmelectromagneticclusters reduces therateby a factor 1/3 for a given ETcut. • Photons: • Higherefficiencywithoutthe PS/SPD (withhigher ETcuttokeepsamerate). • PS rejects 20% of emclusters. • Forphotons, itisbetterthe case withoutany PS requirement. There are photonsthatdepositenergy in the PS belowthecurrentthreshold. • SPD rejects 40% of remainingclusters. • Convertedphotons are triggered as electrons. • Without M1, lessphotonconversions, theperformace of thecurrent L0  shouldbebetter. • Electrons: • PS/SPD electronprovideshighersignalefficiencywith a lower ETcut at a givenrate. • Therate can bereducedwithout PS/SPD by a harder ETcut. • Drawback: relativeloss of efficiencyaround 20% forsignalelectrons. • Acceptableorrecoverablethroughthehadronicpart of thedecay? • WhatabouttheeffectonParticleIdentificationwhenremovingthe PS/SPD?

  23. SPARES

  24. em off electron photon B0 K*e+e- EPS (MeV) of thefour PS cells Threshold at ~10 MIP L0 electronif2 cellsabovethreshold & SPD = 1 feweventsrejected

  25. em off electron photon Convertedphotons 20% eventsrejected Bsfg There are photonsthat do notdepositenoughenergy in PS, thusdiscarded Threshold at ~10 MIP L0 photonif2 cellsabovethreshold & SPD=0

  26. Bsfg Fordifferent PS energycuts Withoutthe <=2 cellsrequirement Higherefficiencyif PS/SPD OFF PS energy in MeV when SPD=0, higherefficiencyif PS>=0

  27. B0 K*e+e- Fordifferent PS energycuts Withoutthe <=2 cellsrequirement Thecurrent L0 electrontriggersettingisclosetotheoptimalone

  28. MinBias # interactions/event lumi2 lumi5 lumi10 lumi20

  29. SelBd2eeKstar Strippingselections # DiLepton (e+e-) DiLeptonForBd2LLKstar.InputLocations = ["StdLooseElectrons"] DiLeptonForBd2LLKstar.DecayDescriptor = "J/psi(1S) -> e+ e-" DiLeptonForBd2LLKstar.DaughtersCuts = {"e+": "(PT>300*MeV) & (MIPCHI2DV(PRIMARY)>1)" } DiLeptonForBd2LLKstar.CombinationCut = "AM<5500*MeV" DiLeptonForBd2LLKstar.MotherCut = "(VFASPF(VCHI2/VDOF)<25)" # Kstar Kstar2KPiForBd2LLKstar.InputLocations = ["StdTightPions", "StdTightKaons"] Kstar2KPiForBd2LLKstar.DecayDescriptor = "[K*(892)0 -> K+ pi-]cc" Kstar2KPiForBd2LLKstar.DaughtersCuts = {"K+": "(PT>350*MeV) & (P>3000*MeV) & (MIPCHI2DV(PRIMARY)>3)", "pi-": "(PT>300*MeV) & (P>3000*MeV) & (MIPCHI2DV(PRIMARY)>3)" } Kstar2KPiForBd2LLKstar.CombinationCut = "(ADAMASS('K*(892)0')<200*MeV)" Kstar2KPiForBd2LLKstar.MotherCut = "(VFASPF(VCHI2/VDOF)<25)" # Bd-> eeKstar PreselBd2Kstaree.InputLocations = ["DiLeptonForBd2LLKstar", "Kstar2KPiForBd2LLKstar"] PreselBd2Kstaree.DecayDescriptor = "[B0 -> K*(892)0 J/psi(1S)]cc" PreselBd2Kstaree.DaughtersCuts = {"K*(892)0": "ALL", "J/psi(1S)": "ALL"} PreselBd2Kstaree.CombinationCut = "(ADAMASS('B0')<1200*MeV)" PreselBd2Kstaree.MotherCut = "(BPVIPCHI2()<64) & (VFASPF(VCHI2/VDOF)<(36.0/4.0)) & (BPVVDCHI2>9) & (BPVDIRA>0.999)" # final selections hardee = "(INTREE( (ID=='J/psi(1S)') & (BPVVD>1.0*mm) ))" hardKstar = "(INTREE( (ABSID=='K*(892)0') & (ADMASS('K*(892)0')<130*MeV) & (BPVIPCHI2()>1.0) & (BPVVDCHI2>1.0) ))" hardB = "(ADMASS('B0')<1000*MeV) & (BPVIP()<0.05*mm)“ from Configurables importFilterDesktop SelBd2eeKstar = FilterDesktop("SelBd2eeKstar") SelBd2eeKstar.InputLocations = ["PreselBd2Kstaree"] SelBd2eeKstar.Code = hardee + " & " + hardKstar + " & " + hardB

  30. Bs2PhiGamma defcombineBs(self, name = "MakeBs2PhiGamma" ): """ Define the Bs """ _stdPhi4Bs = DataOnDemand(Location = "Phys/StdLoosePhi2KK") _phi4BsFilter = FilterDesktop ("PhiFilterFor"+name) _phi4BsFilter.Code = "(MINTREE(ABSID=='K+', MIPCHI2DV(PRIMARY))> %(TrIPchi2Phi)s) & (ADMASS('phi(1020)') < %(PhiMassWinT)s*MeV) & (VFASPF(VCHI2/VDOF) < %(PhiVCHI2)s)" % self.getProps() Phi4Bs = Selection ("Phi2KKFor"+name ,Algorithm = _phi4BsFilter ,RequiredSelections = [_stdPhi4Bs]) _stdgamma = DataOnDemand(Location = "Phys/StdLooseAllPhotons") _gammaFilter = FilterDesktop("GammaFilterFor"+name) _gammaFilter.Code = "(PT> %(photonPT)s*MeV)" % self.getProps() Gamma = Selection ("GammaFor"+name ,Algorithm = _gammaFilter ,RequiredSelections = [_stdgamma]) _Bs2PhiGamma = CombineParticles ( name ,DecayDescriptor = "B_s0 -> phi(1020) gamma" ,CombinationCut = "(ADAMASS('B_s0')<%(BsMassWin)s*MeV)" % self.getProps() ,MotherCut = "(acos(BPVDIRA) < %(BsDirAngle)s) & (BPVIPCHI2() < %(BsPVIPchi2)s)" % self.getProps() ,ReFitPVs = True) Bs2PhiGamma = Selection ( "Sel"+name ,Algorithm = _Bs2PhiGamma ,RequiredSelections = [Gamma, Phi4Bs]) return Bs2PhiGamma

  31. class StrippingB2XGammaConf(LHCbConfigurableUser): """ Definition of B -> X Gamma stripping """ __slots__ = { 'TrIPchi2Phi' : 10 # Dimensionless ,'TrIPchi2Kst' : 10 # Dimensionless ,'PhiMassWinL' : 40 # MeV ,'PhiMassWinT' : 15 # MeV ,'KstMassWinL' : 200 # MeV ,'KstMassWinT' : 100 # MeV ,'KstMassWinSB' : 200 # MeV ,'BsMassWin' : 1000 # MeV ,'B0MassWin' : 1000 # MeV ,'BMassWinSB' : 2000 # MeV ,'BsDirAngle' : 0.02 # radians ,'B0DirAngle' : 0.02 # radians ,'BDirAngleMoni' : 0.06 # radians ,'BsPVIPchi2' : 15 # Dimensionless ,'B0PVIPchi2' : 15 # Dimensionless ,'photonPT' : 2000 # MeV ,'PhiVCHI2' : 15 # dimensionless ,'KstVCHI2' : 15 # dimensionless }

More Related