The BeamCal Simulation Project
This presentation is the property of its rightful owner.
Sponsored Links
1 / 82

The BeamCal Simulation Project Progress Report PowerPoint PPT Presentation


  • 58 Views
  • Uploaded on
  • Presentation posted in: General

The BeamCal Simulation Project Progress Report Keith Drake, Tera Dunn, Jack Gill, Gleb Oleinik , Uriel Nauenberg University of Colorado at Boulder. Beam Calorimeter Studies. Two photon process cross section about 10 5 larger than

Download Presentation

The BeamCal Simulation Project Progress Report

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.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


  • The BeamCal Simulation Project

  • Progress Report

  • Keith Drake, Tera Dunn, Jack Gill,

  • Gleb Oleinik, Uriel Nauenberg

  • University of Colorado at Boulder


Beam Calorimeter Studies

Two photon process

cross section

about 105 larger than

SUSY cross section.

Serious source of

background for SUSY

if not tagged .

Pointed out by our group around 1998


2 Photon Process

e+ e+

Very forward

into BEAMCAL

- -

γ

q , l

Very forward

into BEAMCAL

in the detector

γ

q , l

e- e-


Lumi Cal

-7 mrad ent.

7 mrad exit

Beam Cal


  • Solenoid field keeps the low energy charged particle in the forward direction. Beam hole is at 7 mrad. Need to add an x field component to move low energy charged particles in the 7 mrad direction. Anti-DiD

  • dipole field proposed by Andrei Seryi.


Beamstrahlung e+e- pairs.

Energy deposited in

0.25 x0.25 cm2 cells.


Outside Beam Pipe

Integrated

Energy into the Beam Pipe is ~TeV


Shower in Beamcal from 2 γ process alone

side view

head-on view

core of shower

Radius of 25 mm

Radius of 25 mm


  • head on view

side view

beamstrahlung

electron from 2γ process


Beamstrahlung Energy Deposition in a Tile

Tile at start of BeamCal

Tile 3.0 cm in


Beamstrahlung Energy Deposition in a Tile

Tile 6.0 cm in

Tile 4.7 cm in


  • Consequences of Beamstrahlung Energy

  • Deposition

  • Energy Deposition is not Gaussian until one

    reaches a depth > 3 cms.

  • Distribution is very wide and hence affects

    energy resolution if we subtract an average

    value.

  • This problem is seen in the study how to measure

    the electron/positron energies. Resolution.


Electron from 2 photon

overlayed

Beamstrahlung

Beamstrahlung

with average subtracted


Clustering Cuts in Depth and Energy, Example 1

Beamstrahlung +

Electron from 2-photon

Beamstrahlung Alone

Cut 30 mm

Work in Progress

Cut 10 MeV


Clustering Cuts in Depth and Energy, Example 2

Beamstrahlung +

Electron from 2-photon

Beamstrahlung Alone


Clustering Cuts in Depth and Energy, Example 3

Beamstrahlung +

Electron from 2-photon

Beamstrahlung Alone


Energy observed as a function of distance from center of BeamCal

φ

r

Exit Beam Pipe

Entrance Beam Pipe

Δφ = 5 deg

Δr = 1 mm


Energy Deposition versus r and φ

Φ=0.0

Φ=30.0

30.0

Effect of Exit Beam Pipe

Effect of Exit Beam Pipe

Distance from center of beamcal (mm)

Distance from center of beamcal (mm)


Energy Deposition versus r and φ

Φ=150.0

Φ=180.0

150.0

180.0

Effect of Entrance Beam Pipe

Effect of Entrance Beam Pipe

Distance from center of beamcal (mm)

Distance from center of beamcal (mm)


sensors

Within a radius of 25mm

no clustering

Full detector

Full Detector

with clustering


Fraction of Energy Observed within a

Radius of 0.5 cm of Electron Path

No clustering cuts

With clustering cuts


Fraction of Energy Observed within a

Radius of 1.0 cm of Electron Path

No clustering cuts

With clustering cuts


Fraction of Energy Observed within a

Radius of 2.0 cm of Electron Path

With clustering cuts

No clustering cuts


Fraction of Energy Observed within a

Radius of 3.0 cm of Electron Path


  • Energy Resolution of High Energy Electrons

{E(measured)- E(expected)} /E(expected)

Best Possible Resolution

No Beamstrahlung effects

included


  • Energy Resolution of High Energy Electrons

{E(measured)- E(expected)} /E(expected)

Energy Resolution with

Beamstrahlung Included.

Average Substracted. Only energy deposited in a 25 mm

radius from maximum. Sum energy over full Beamcal thickness.

Effect of Beamstrhalung fluctuation in resolution clearly has an effect


  • Energy Resolution of High Energy Electrons

{E(measured)- E(expected)} /E(expected)

Energy Resolution. Beamstrahlung

Included and Average Substracted.

Measurements from 3.0 cm in and

Including only cells with more than

10 MeV energy deposited.


  • Energy resolution of the reconstruction of the electrons from 2-photon events including the effects of beamstrahlung

Preliminary results

Preliminary results

Using Bayesian

Statistical Analysis


Case of no Beamstrahlung

Included in analysis

Case of Best Analysis with

Beamstrahlung Included.

Exit Beam Pipe

Region

Exit Beam Pipe

Region


  • Reason for Resolution Tail

Measured Energy loss due to cuts


  • Work to be Done

  • Understand the low energy tails of the energy resolution

    distribution. Develop a better scan.

  • What is the missing Pt distribution of 2 photon events given

    the resolution.

  • Study the resolution of new geometries.


  • Work to be Done

  • Optimize signal to Background.

  • Check all our Calculations.

  • Find other analysis techniques that reduce the

    beamstrahlung fluctuations and hence improve the

    signal resolution.

  • Study the effect of this analysis on SUSY signal.

    Missing Pt limits.


  • Study of a Scintillator Calorimeter

  • We are simulating a scintillator based calorimeter where the tiles

  • are offset in alternate layers. We are making now a great deal of progress.


γ s are well separated

5 cm

100 GeV

100 GeV

π0 mostly low p


ALCPG

Geometrical Arrangement

5 cm

Scintillator

Panels

Worst separation

1.25 cm

·

·

·

·

φ

·····

0 0.5 1.0

0.25 0.75

θ

Best separation


Track Following into the Calorimeter

Low momentum curlers


Cluster Correlation with Charged Tracks Success Probability

Energy cluster in the calorimeter associated with

a charged track interaction

Success

Conclude that it is a neutral cluster

Failure


ALCPG

  • The Chi-Square Structure

  • μi = average photon energy deposited in ith tile

  • σi = standard deviation in the energy deposition

  • Hij = σi σj

  • χ2 = Σ (xi – μi )Hij (xj – μj )

  • where xi is the energy deposited by the shower being tested in the ith tile.

9

-1

i,j =1


  • We are now in the middle of trying to separate

  • photon clusters by means of the chi-square

  • method. Hard problem. Crucial aspect of

  • pattern recognition and calorimeter resolution.


Fitted γ direction from shower energy distribution

Z vs R

50 GeV

20 GeV


  • Study of the Characteristics of

  • Silicon Photomultipliers


  • New Silicon Photo-Detectors

Bias Voltage

~40 volts

Photonique, SA

Pulsar, Russia

+

Moscow Eng.

Physics Inst.

2mm

Scintillator

performance


2mm scint., cosmic rays

0 < t < 200 nsec

20 < t < 70 nsec

1 photo elec.

Similar resolution

5 photo elec.

Need to Improve Resolution

8 photo elec.


Sipm attached to the center of surface of the scintillator tile


New Nat. Inst. PHA


  • Special Budgetary Issue

  • ILC R&D in DOE (Paul Grannis) awarded us a

  • $20 K late award that could not be sent to CU

  • but could only be deposited at SLAC because of

  • the timing.

  • I request that I use these funds for BaBar work

  • and be allowed to used BaBar funds deposited

  • in Colorado for ILC R&D work.


  • Because ILC R&D funds become available in

  • ~ July the funds cover 2 years of BaBar work

  • 1 year of differential housing costs for Nagel

  • $5 K

  • 1 year travel costs from Colorado to SLAC

  • $5 K

  • The total award from ILC R&D is $53 K


  • The University has contributed to my research

  • a total of $38 K towards support of a Research

  • Associate since I have become Chair of the

  • Boulder Faculty Assembly and my research time

  • is now limited. I propose to use these and the

  • ILC R&D funds towards the Research

  • Associate if DOE accepts the ILC-BaBar fund

  • exchange.


The Calorimeter

  • Modules

  • Each module = ~12 tons


Removal of Charged Track Hits

Extrapolation of charged tracks

Charged hits removed

Track

Interact

Jason Gray, Jiaxin Yu


Pattern Recognition of Showers

Chris Geraci


Chi Square Separation, 1st order

Sarah Moll

Joseph Proulx

100 GeV π0

Odd layers

100 GeV π0

Even layers


2 Photon Process

Very forward

into BEAMCAL

e+ e+

- -

γ

q , l

in the detector

γ

q , l

Very forward

into BEAMCAL

e- e-

Discussion in Beam Cal section at end


  • Study the efficiency to observe the electron and positron of the two photon process above the beamstrahlung background

  • Essential to remove this background in the

  • study of Supersymmetry in the dynamic region

  • of low Pt. Needed to measure the masses.


Testing GEANT 4.0

Evidence for multiple scattering

air in beam pipe

No field, 50 GeV muons

No field, 50 MeVmuons


50 MeV, solenoid on, forward

50 MeV, no field, forward

Solenoid field shrinks mult. Scat.


Beamstrahlung Distribution with Solenoid + Anti-DiD

Beam pipes

Entrance Exit

No beam pipes

Difficultregion to detect

2 photon process


  • GEANT 4.0 seems to be working

  • properly We have fixed various bugs in

  • collaboration with SLAC team.

  • All Simulation is work in progress.


  • Hardware Studies

  • Keith Drake, Elliot Smith


  • Long Term Tests of Scint. Fiber Stability


  • Latest Pulse Distribution from

  • Photonique/Russia

6 photoelectrons

10 photoelectrons


Pulse National Inst.

.

.

.

.

.

.

3 pe

.

1 pe

8 pe

Scan every 2.5 nsec


Our Measurements

6 photo-elec.

Blue LED,

10nsec, 2.5 volts drive

50 nsec gate

noiseless SiPD


1st 10 nsec

2nd 10 nsec

3rd 10 nsec

4th 10 nsec


  • Cosmic Rays in a 1 cm Thick Scintillator

20 < t < 70 nsec

0 < t < 200 nsec

poorer resolution

1 pe

2pe

4pe


1cm scint

10 < t < 20

30 < t < 40

40 < t < 50

60 < t < 70


  • Pulse Height versus Temperature

  • Gain of SiPD Increases ~x4

Noisy SiPD

Could not detect peaks

Room temp

-300 C

Needs to be repeated with

New SiPDs; noiseless


2 mm scint

10 < t < 20

30 < t < 40

40 < t < 50 nsec

60 < t < 70 nsec


  • By time tagging we are observing the

  • photons arriving in time sequence. Possibility

  • to use this to improve resolution. Need beam tests

  • to check this hypothesis.


  • Work still in progress. Comparison with Russian

  • plots indicate 100 Mhz x 4 (measuring pulse height

  • every 2.5 nsec) not enough resolution.

  • National Instruments has just released a 2

  • Gigahertz unit. Using our trick of x4 will allow us to

  • scan every 0.125 nsec. A demo is on its way here to

  • check whether 8 bits resolution is good enough.


  • Conclusions

  • A new revolution in photo-detection. A lot of

  • improvements still possible. A lot of work to be

  • done in this area. If one is bold and reckless one

  • may say that “It may revolutionize calorimetry

  • resolution.”


  • Conclusions

  • Our simulation work with the undergraduates is

  • moving ahead. A lot of work needs to be done

  • still. Need manpower help. Most of the pieces are

  • in place to study Z and W mass resolution.


We helped with the organization

of the Linear Collider Workshop

and the various ALCPG meetings

up to but not including Vancouver


  • Login