Triggering on ISR Events Nicolas Berger BaBar Online/DAQ/Trigger Workshop December 2 nd, 2004.

Triggering on ISR Events

Nicolas Berger

BaBar Online/DAQ/Trigger WorkshopDecember 2nd, 2004

ISR Signal• Look for events with

– high-E photon (E* = 3-5.3 GeV)

– Recoiling hadronic system (charged or neutral)

• Produce qq continuum and 1-- resonances (…)

• Use s’ = inv. mass squared of Had system as reference

BG Filter Improvements• 2 New filters in 2003 reprocessing:

• BGFIsr: Select ISR events– Require cluster with E*>2.5 GeV,

-0.75 < cos(*)< 0.93 and one of:– 1 track E/p < 0.7, p*>600 MeV/c OR– 2 tracks, E/p < 0.8, p*>250 MeV/c OR– 2 clusters, E > 100 MeV– Veto if tracks with E/p > 0.8 and p*>4 GeV/c– Veto on topologies

• BGFRadTwoProng: Unbiased 2-prong sample for efficiency studies.

– 2 tracks with p* > 250 MeV– Missing p* > 2 GeV, Pointing in EMC acceptance– Missing M < 2 GeV.

BG Filter Improvements

• BG Filter efficiency vs. s’ for different filter configurations, on 0

+- MC

• Denominator = fiducial events

• BGFIsr gives significant improvement

Only BGFMultiHadron

with BGFNeutralHadron

with BGFTau

with BGFTwoProng with BGFIsr

s’ [GeV]

Eff

icie

ncy

Trigger/Filter Overall Performance

• We use:– L3 Physics =

(L3OutDch || L3OutEmc)– MH = BGFMultiHadron– NH = BGFNeutralHadron– ISR = BGFIsr– = BGFPhiGamma

• Denominator = Events with >0 clusters with – E* > 3 GeV– |cos(*)|<0.95

Filter Level Eff (%)

L1 99.97

L3 99.5

L3 Physics 99.4

BGF 97.2

MH, NH, ISR, 94.4

MH, NH, ISR 93.7

MH, ISR 93.0

ISR 90.1

Trigger/Filter Overall Performance

Events Passing:

L1L3L3 PhysicsBGFMultiHad, NeutralHad, ISR and MultiHad, NeutralHad and ISRMultiHad, ISRISR

s’ [GeV]

L1 Line Breakdown

s’ [GeV]

s’ [GeV]

+

L3 Breakdown

s’ [GeV]

s’ [GeV]

High L3OutEmc efficiency due to auxiliary clusters caused by interaction in EMC

0 +-


L1 99.96

L3 98.9

L3 Physics 98.9

BGF 97.4

MH, NH, ISR, 90.7

MH, NH, ISR 90.2

MH, ISR 90.2

ISR 89.6s’ [GeV]

BGFPhiGamma designed to pick up K+K- but also works for low-mass

0 +- Issue 1: Close Tracks

Fails L1 Fails L3 (2 L3 tracks but misreconstructed)

0 +- Issue 2: Pion showers

Fails BGF Doesn’t fire BGFIsr due to Bhabha veto

0 +- Issue 3: Interactions

Fails L3 , No IP tracks (We probably don’t want this event anyway!)

K0SK0

L


L1 99.97

L3 97.3

L3 Physics 96.0

BGF 71.8

MH, NH, ISR,

61.6

MH, NH, ISR 56.5

MH, ISR 56.3

ISR 56.0

Fails L3

Fails L3

Accounts for almost all of the overall efficiency dip at mass

e+e-


L1 99.95

L3 99.6

L3 Physics 99.5

BGF 57.4

MH, NH, ISR, 9.5

MH, NH, ISR 9.5

MH, ISR 9.1

ISR 7.8

• Almost completely removed by Bhabha Vetos

• Accounts for most of the dip in efficiency at mass.

Break-up by NGoodTracksLoose : NGTL=0


L1 99.0

L3 94.0

L3 Physics 93.6

BGF 58.4

MH, NH, ISR, 48.5

MH, NH, ISR 48.5

MH, ISR 47.5

ISR 47.5

s’ [GeV]

• BGF dominated by BGFIsr, requires 2 mips

• Low eff. At low s’, but not clear how useful these events are anyways.

Break-up by NGTL : NGTL=1


L1 99.96

L3 98.0

L3 Physics 97.9

BGF 88.8

MH, NH, ISR, 86.3

MH, NH, ISR 86.2

MH, ISR 84.1

ISR 84.0

s’ [GeV]



L1 99.99

L3 99.5

L3 Physics 99.5

BGF 98.7

MH, NH, ISR, 93.1

MH, NH, ISR 91.5

MH, ISR 90.4

ISR 90.2s’ [GeV]

BGFPhiGamma efficient on low-mass events for both KK and



L1 100

L3 99.93

L3 Physics 99.93

BGF 99.7

MH, NH, ISR, 99.5

MH, NH, ISR 99.5

MH, ISR 99.4

ISR 92.5

s’ [GeV]

Break-up by NGTL : NGTL>4


L1 100

L3 99.99

L3 Physics 99.99

BGF 99.97

MH, NH, ISR, 99.96

MH, NH, ISR 99.96

MH, ISR 99.96

ISR 94,4

s’ [GeV]

Conclusions

• ISR events are a challenge for the trigger

• The BaBar trigger performs well– Most of the events useful for physics are

selected– For L1 and L3, benefit of the “inclusive”

selection approach.– For BGF, more problems originally but

modifications always possible (if willing to wait ~6 months!)

Triggering on ISR Events Nicolas Berger BaBar Online/DAQ/Trigger Workshop December 2 nd, 2004.

Documents