Документ взят из кэша поисковой машины. Адрес оригинального документа : http://zebu.uoregon.edu/~uochep/talks/talks05/torrence_mdi.pdf Дата изменения: Tue Sep 20 00:19:23 2005 Дата индексирования: Tue Oct 2 11:51:07 2012 Кодировка: ISO8859-5 Поисковые слова: m 45

Interaction Regions, Backgrounds, IP Beam Instrumentation

Eric Torrence University of Oregon
Eric Torrence 1/27 August 2005


Machine Detector Interface

Machine

and Cost!
Detector Physics

Machine Detector Interface is a complicated optimization problem. Need best configuration for the physics program Example MDI topics З З З З З З Crossing Angle - Final Focus Layout Final Doublet L* Machine Backgrounds IP Beam Instrumentation (Energy, Polarization) Luminosity Spectrum Determination Instrumentation in Forward Region Way too much information to cover. Couldn't even attend all relevant sessions...
Eric Torrence 2/27 August 2005


ILC Working Group 4 Beam Delivery and Interaction Region see also A. Seryi WG4 Summary th August 19

Eric Torrence

3/27

August 2005


Interaction Region Layout From Cartoon at KEK

1st ILC Workshop November 2004

to Design at RHUL
BDIR WS London June 2005

Full optics for all beamlines, 2 mRad and 20 mRad designs explored in detail, up/downstream instrumentation present for both IRs. Working now on refinements, evaluating performance of designs
Eric Torrence 4/27 August 2005


Crossing Angle Comparison 20 mRad crossing angle З З З З З З З З З Separate incoming and extraction beamlines More amenable to high luminosity? Cleaner downstream diagnostics? May be compatible with and > 1 TeV Expect good operational margins and flexibility Reliance on crab cavity Reduced detector hermeticity Need to correct solenoid crossing (DID or other) Somewhat higher pair backgrounds

but

2 mRad crossing angle З З З З З Lots of recent work (could still be improved?) Improves 20 mRad issues (crab, hermeticity, DID) Downstream instrumentation more difficult? More particle losses downstream, closer to IP More constrained design, problems with high Lumi Derived from A. Seryi, other opinions exist... Not at all obvious which is best in the big picture continue to develop and understand both...
Eric Torrence 5/27 August 2005


Extraction-Line Losses From A. Seryi
L

20 mrad IR
Low ~2W High ~1000 W

vs
High Lumi

L

2 mrad IR
preliminary & qualitative High Very High To be evaluated Low ~1W Low ~1W

Nominal

High Lumi

Very Low <1W

Very Low <1W

500 GeV

1 TeV

E

Nominal

500 GeV

1 TeV

E

Numbers in Watts show losses on SC FD Magnets Personal Opinion: Current high Lumi parameter sets may be unrealistic, but they probably give a good indication of where the machine wants to go... Remember: ILC Lumi = 10,000 x SLC Lumi achieving this will be a major (the major?) challenge of ILC
Eric Torrence 6/27 August 2005


20 mRad Detector Integrated Dipole e+e With 20 mRad crossing-angle З Polarization vector is rotated (difficult for precision) З Orbit bump causes synchrotron radiation - limits y (30% lumi) 20 mRad, no correction e-e
- -

Detector Integrated Dipole (DID) offers a good machine solution align field along incoming beam direction But likely causes difficulties for detector: Non-uniform solenoid field Background issues (redirecting pairs) Concepts asked to evaluate if significant impact. Tracking issues under study, LDC (Busser) indicates higher TPC backgrounds with current masking design. Significance not clear.
Eric Torrence 7/27 August 2005


Alternate Crossing Angles Advanced design for 20 mRad final doublet Now thinking about reducing crossing angle by reducing material between coils B. Parker, BNL

Angles down to 10-12 mRad will be studied as alternates to baseline Likely to improve 20 mRad issues, incompatible with Head-on Redux Complicated magnets and potentially large losses in 2 mRad has prompted a renewed look at head-on options: RF kicker (Y. Iwashita) and SLC-style separator (L. Keller) Needs large effort to become realistic alternative
Eric Torrence 8/27 August 2005


Backgrounds

see also T. Maruyama talk th August 17

Eric Torrence

9/27

August 2005


Background Sources IP Backgrounds (Good backgrounds) З З З З З З З Disrupted primary beam - extraction line losses Beamstrahlung (BSL) photons e+e- pairs from BSL s pair backsplash from final doublet Hadrons from BSL or Neutrons from e+e- pairs Radiative Bhabhas These scale with Luminosity: transport away, shield detectors. More reliable simulations Machine Backgrounds (Bad backgrounds) З З З З З Synchrotron radiation Neutron back-shine from dump Muon production Collimator scraping Beam Gas These don't scale with Lumi: avoid near IP. Highly dependent upon assumptions Tedious to evaluate all in detail, but clearly important for detector and IR conceptual designs!
Eric Torrence 10/27 August 2005


New LDC Pair Simulations

LDC Studies (K. Busser) Investigate pair backgrounds in VXD and TPC for crossing angle options w/ realistic DID field

Initial Conclusions З Realistic DID field changes pair hit pattern З Pairs (and junk) from incoming hole channeled into vertex detector L1 З Pairs hitting LumiCal edge scatter photons to TPC З Opening LumiCal reduces effect somewhat Details of fields and geometry very important... TPC backgrounds worse in 20 mRad + DID, but is this significant?
Eric Torrence 11/27 August 2005

TPC Hits/BX

20 mRad + DID

20 mRad

2 mRad

Adjust


Background Tolerance Estimates Full simulations and physics studies are slow. What can be understood from "rule of thumb" background tolerances? W. Kozanecki (Saclay), et. al. Different tolerances: damage, pile-up, pattern recognition, physics performance Working assumption: 1% occupancy in tracking detectors. Conservative, but need realistic x10 safety factor.

Example from Witold's Talk LDC Vertex occupancy vs. layer comparing DID/crossing angle (pairs) Data from K. Busser Interpretation (tolerance) by Witold See talk Wed. Aug. 24th for many more plots like this

~ 173 hits mm-2 tr -1 (but: hot spots!?)

Tol ( = 6 hits mm-2 / 50 Еs)

Eric Torrence

12/27

August 2005


More Background Tolerance General conclusions (Witold) З З З З З 1% occupancy/readout window threshold useful for comparisions VXD (SiD, LDC) at or below 1%, GLD well below 1% TPC well below 1%* SiD tracker pileup appears to be 5-10 greater than stated tolerance High luminosity and low power parameters cause trouble

*Warning: Correlated hits (i.e.: tracks) and hot-spots may drastically change these conclusions. True impact/tolerance can only be evaluated by detector experts with detailed studies in concept groups.

Background Comments General limits useful to guide IR design, but also need the details. Each detector concept group must "take ownership" of their background estimations and work with WG4 to more towards realistic IR designs. Work generally started in all concepts. This is a lot of work, and adequate resources must be available in all concept groups to tackle all relevant background sources.

Eric Torrence

13/27

August 2005


Vertex Tolerances Radius (cm)

SiD 12 mm beampipe (Maruyama)

Minimize R1

Distance from IP (cm) Direct pairs kept away from VXD by solenoid field, but tolerances are often tight (few mm) May limit initial machine operations. Solenoid may not always run at nominal field. Detectors willing to sacrifice layer 1?
Eric Torrence 14/27 August 2005


Variation with Parameter Sets Transverse Momentum vs. Theta C. Rimbault, LAL Orsay

Nominal

Low Q

Large Y

Low P

High Lumi

Nominal Edge High Lumi and Low Power problematic Unwise to push too hard here?
Eric Torrence 15/27 August 2005


Beam Instrumentation and Forward Region

see also K. M?nig's talk th August 17

Eric Torrence

16/27

August 2005


IP Instrumentation
0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 490 492 494 496 498 500 502 504 Root(s) (GeV)

( -, - ), ( +, + ), ... dL -----dE L

0

s Fundamental IP Beam Instrumentation Goal Spin-dependent absolute collision energy spectrum Typical Components З З З З З Beam Energy Beam Energy Width Beam Polarization Absolute Luminosity Differential Luminosity Spectrum

Mixture of beam-based and physics-based measurements
Eric Torrence 17/27 August 2005


Beam Instrumentation Design Upstream and Downstream spectrometer and polarimeter designs now exist for both 2 mRad and 20 mRad schemes 20 mRad Extraction Line
10 meters 10 cm
BVX9 z=100.59 m BVY1 z=120.59 m

Polarimeter Chicane
25 GeV

Synchrotron Stripe Detector z=143.69 m BVY3 z=152.59 m QFX4 z=156.59 m

BVY2 z=132.59 m 3 mrad energy stripe

Energy Chicane
BVX1 z=59.69 m BVX2 z=61.99 m BVX6 BVX4 z=70.49 m z=75.09 m BVX5 BVX3 z=68.19 m z=72.79 m BVX8 z=83.59 m BVX7 z=81.29 m

2 mrad energy stripe 45.59 GeV

Stripe Detector

Synchrotron

0 .75 m ra d

Compton IP

250 GeV

BPM

BPM

25.1 GeV

35.7 GeV

12.0 cm

0. 75 mr ad
2 mrad energy stripe 3 mrad energy stripe

1 7. 8 cm

z=64.29 m

Low Field BLEX

z=66.89 m

BLEX

BLEX BLEX z=77.39 m z=79.99 m WEX1 z=78.69 m

WEX1 z=65.59 m Wiggler

Ken Moffeit - LCWS

Synchrotron Stripe Detector

Shielding

Cerenkov Detector

Upstream probably cleaner, measures incoming beam parameters Downstream more challenging, but gives sensitivity to collision details Both needed to reach ultimate accuracy of a precision machine Complimentary systematics and control over collision uncertainties
Eric Torrence 18/27 August 2005


Beam Instrumentation R&D 100 ЕRad

RF BPM Triplets ~10 meters

1 mm

Upstream spectrometer needs BPM resolution and stability at sub-100 nm level for several hours BPM Tests at ATF (nanoBPM program)

+/- 50 nm drifts

Many discussions this meeting on advancing spectrometer designs. Tests starting in ESA this fall.
Eric Torrence 19/27 August 2005


Physics Reference Reactions Bhabha acolinearity З Best input for lumi spectrum shape З Strong requirements on performance of forward tracking and calorimetry? 200 mRad Е + Е - "Radiative Returns" З Potentially best measure of s correct for any collision bias З Actually used at LEPII serious detector systematics e+e- Е ЕЕ 1 2 e+e- Еe+e-

Also t-channel WW for polarization monitoring Stringent detector requirements Need precise tracking to ~150 mRad 0. 1 % per event ( Z limit), absolute angle known to 10-4 Forward tracking system must be given same care and effort as precision luminosity measurements. See K. M?nig LDC talk Will likely determine ultimate precision on masses due to s ! Need to do a good job here to reach m ~ 50 MeV
Eric Torrence 20/27 August 2005


Putting it all Together

0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 490 492 494 496 498 500 502 504 Root(s) (GeV)

s e+eЕe+ee+e- Е e+e- Е ЕЕ

1 2

1

2 e+e- Еe+e-

Need combined analysis putting together all pieces and extracting physics result S. Boogert working on efficient MC generation from beam-beam simulations Goal: close the loop and show required precision can be achieved
Eric Torrence 21/27 August 2005


Forward Detectors LDC Forward design (2 mRad, L* = 4.05 m)

Luminosity Monitor ~ 20-50 mRad - Outside pair backgrounds Also shields pair backsplash from lower angles З Pair Monitor ~ 5-20 mRad - Front surface, machine diagnostics З Tagger ~ 5-20 mRad - Cal. to veto electrons as SUSY background Could be all one, or several different detectors
Eric Torrence 22/27 August 2005


Precision Luminosity Series of talks on precision luminosity - W. Lohmann et. al. Ambitious goal of 10-4 (below 10-3 probably OK)

Extensive studies made for head-on Rad-hard detectors considered Initial studies for 20 mRad indicate larger backgrounds - need increased inner radius. Detailed evaluation of 20 mRad boost on detector geometry needed Ultimate OPAL precision based on phi symmetry - D. Strom May be much more difficult in crossing angle
Eric Torrence 23/27 August 2005


Far-Forward Tagging

Electron hermeticity key for SUSY and other analyses Challenge: Separate 250 GeV electron from 10s TeV pair backgrounds per crossing Backgrounds to stau analysis Zhang (Orsay)

Several studies done on performance ~~ ~~ for Е Е and Rely on huge suppression of background ~ 99.9% probably required! Modest acceptance hole is OK if you know where it is - reduced efficiency

1 fb

Larger hole leads to Bhabha backgrounds (1 Bhabha per 2 BX) Pair background rates problematic

Eric Torrence

24/27

August 2005


Forward Pair Distributions Pair Energy in BeamCal (L* = 4m, B = 4T) - P. Bambade - Orsay 2 mRad Nominal 2 mRad Low Q 2 mRad High Lumi

Larger energy deposition at larger angles impacts 20 mRad + DID Nominal ultra-efficient electron ID 2 mRad qualitatively better Can quantify 2 mRad - 20 mRad difference under certain assumptions/models ~~ e.g.: Factor of 1.8 in M ( - x 0 ) reach - Bambade Differences from machine parameter sets very significant Low Q (or similar) would be most beneficial if physics demanded Hard to judge how much weight this carries in global optimization
Eric Torrence 25/27 August 2005


MDI Structure

There is no MDI working group!
More of an avenue of communication between the accelerator, detector, and physics groups Most closely tied to ILC WG4 - IR layout issues + community on detector/physics side Global issues (e.g.: 1 or 2 IRs, parameters) also important

WWS Interim MDI Panel - through Snowmass M. Woods, P. Bambade, T. Tauchi Needs to be expanded/reformulated to include concept representatives, WG4, and guidance from GDE MDI Communication Examples З Urgent MDI questions for concepts - necessary to complete conceptual design З Vertex session Tuesday with questions for WG4 Communication goes both ways...
Eric Torrence 26/27 August 2005


Summary

З Wide range of topics covered at Snowmass under WG4, MDI, and IP Beam Instrumentation З Key features of IR conceptual design in place Baseline crossing angles: 2 mRad and 20 mRad Intermediate 10-12 mRad to be pursued as alternate З Detector backgrounds depend on details of detector technology and IR geometry Large effort from concepts needed here! З Beam instrumentation design proceeding Detailed evaluation of performance starting З Physics reference processes also needed Stringent detector requirements, part of benchmarks З Studies of forward detector performance continuing

Working towards ever better understanding of MDI optimization process

Eric Torrence

27/27

August 2005