The BoNuS Detector A Radial-Drift GEM TPC Howard Fenker TPC R&D Meeting LBL, March 24, 2005

The BoNuS Detector

A Radial-Drift GEM TPC

Howard Fenker TPC R&D MeetingLBL, March 24, 2005

This work was partially supported by DOE Contract No. DE-AC05-84ER40150 under which the Southeastern Universities Research Association (SURA) operates the Thomas Jefferson National Accelerator Facility (Jefferson Lab).

Barely

Off-shell

Structure

Nucleon

Motivation

• Purpose– Provide almost-free

neutron target to improve our understanding of neutron structure.

Method

• Measure slow protons– Identify spectator

protons to tag e-d events in which the neutron was struck.

– Measurement of proton momentum vector tells the initial state of the neutron.

np

e

before

e

p

?

after

n

Spectator Proton Characteristics

• Angular distribution is isotropic. Backwards proton almost certain to be a spectator.

• Momentum distribution favors low values.

Relative Ionization Yields

Spectator tracks are 20x - 50x minimum ionizing.

-> Detector can be made almost insensitive to lighter particles.

How to do it?

• Spectator must escape target

• Low density detector media.

• Minimal insensitive material

• Acceptance– Large– Symmetric about the target

• Detector sensitive to spectators, insensitive to background.

Big Picture

• Track secondary e- in CLAS.

• Locate e- interaction point in target.

• Link pspectator with electron vertex (need z ~8mm).

Narrow Target

Optical properties of the CEBAF beam allow the use of a very narrow target.

Spectator protons can escape the target and be detected.

Conventional TPC

Advantages for BoNuS:•Cylindrical•Very low mass. •Most energy dE/dx is in sensitive materials.•Many measurements of each track.

Conventional TPC Radial

TPC

•Shorter drift path.•Tolerates non-uniform B-field.

STAR Radial TPC

BoNuS Detector Concept

Why GEM Readout?

• Uniform acceptance

• Reduced mechanical strength required– Less massive components

• It is interesting!– First use of GEMs at JLab– First use of curved GEMs anywhere

Availability of GEMs• Both CERN and 3M have produced high quality GEM foils.• CERN’s priority is the internal program; they have limited

capacity.• 3M’s priority is Return on Investment: mass production is

possible. • Tech Etch (Plymouth, MA) applying for R&D Funding.• We (and others) have tested both 3M and Tech-Etch GEMs.• Tech-Etch: Have supplied working custom GEMs for the

BoNuS detector.

Problems with GEMs

• Gain Uniformity– Seen to vary ~10% over a 10cm x 10cm CERN GEM

• Study of GEM Characteristics for Application in a Micro-TPC B. Yu, V. Radeka, G. C. Smith, C. L. Woody, and N. N. Smirnoff

• Gain Stability over time– Short term- “charge-up”– Long term- gain drift– Rate dependence

Short-term: “Charge-Up”

From “A Comparative Study of GEM Foils from Different Manufacturers”, Bob Azmoun (BNL), G. Karagiorgi(FIT), C. Woody (BNL)

Long-term: Gain Drift

Rate Dependence of Gain

• Gain variations are a potentially serious problem for dE/dx measurements– We are considering various calibration options

• Not so serious for tracking.

• NEVERTHELESS… we proceed.

Sample Event: Proton ID by dE/dx & Curvature

100 MeV/c pion 100 MeV/c proton

drift

xpad #

xpad #

drift

20mm x 19mm 20mm x 19mm

20mm x 19mm 20mm x 19mm

box size represents Q(x,t)

Cosmic Tracks Proton Tracks

Production Model: Exploded View

Production Model

BoNuS in CLAS

Detector Parameters

• Geometric Acceptance– Sensitive over 296 deg. In phi, 20cm in Z.

• Momentum Acceptance– Protons from ~70 MeV/c

• Proton Identification (next slide)

• Vertex Z resolution <~ 10mm• Track Momentum Resolution dP/P <~30%• Track E information from dE/dx• Rate & Timing to handle ~2000 events/s

Detector Development

• Prototype – flat, ‘standard’ GEMs• Test• Prototype – curved, ‘standard’ GEMs• Test• Prototype – flat, custom GEMs• Test• Production – curved, custom GEMs• June test run in CLAS

Prototypes

Prototype Construction• Curved

Prototype Test Fit

Drift Region Cathode

Field Cage Electrodes

GEM HV Connections

(GEMs and Readout Board are not shown)

ULTEM® Frame Parts

GEMs CAN be Curved

GEMs CAN be Curved…

Curved GEM E-Fields

• 60mm radius

• Negligible change in E-field

• Curving the GEMs should not be an issue.

LOCAL FIELD DEFORMATION NEAR CURVED GEM IS NEGLIGIBLE

…and Curved GEMs do work.

Operation of a Curved-GEM Radial TPC, at least with the modest resolution requirements of BoNuS, has been successful.

The challenge is construction.

Electronics Development

• Commercial CAMAC Modules – 8 ch. (5/03)

• Test the whole concept– Cosmic / Source / TUNL Run1 (11/03)

• Milestone: choose ALTRO readout (12/03)

• ALTRO Test Board – 16 ch. – Overcome signal polarity issue– Cosmic / Source / TUNL Run 2 (5/04)

• ALTRO Production System – 128 ch. (12/04)– Cosmic / Source / TUNL Run 3 (3/05)

• Final ALTRO System – 3200 ch. (6/05)

BoNuS Signal Readout System

ALICE and BoNuS Readout System

BoNuS Carrier Card

ALICE FECALICE RCU

RibbonsToRTPC

pRTPC w/ Inverter/Driver Cards

RibbonsToReadoutSystem

Detector System Tests

• Readout – 128 channel tests – Analog Electronics

• Prototype test results. – Software– Straight tracks– dE/dx…

dE/dx Analysis from TUNL

dE/dx Analysis from TUNL

dE/dx Analysis from TUNL

Software/Analysis Challenges• DAQ: Interface w/CLAS data structure• Event visualization – OK and improving• Track Fitting - Straight Tracks OK and

improving

• Need to incorporate– Vdrift(R) (ongoing)– B-Field & its Non-Uniformity– Lorentz angle

RHIT vs. Time Bin(Vdrift varies with R)

Curved e- DriftN.B.: non-uniform B

Bonus Detector Outlook

• Detector Parts Delivery 4/1• Detector Assembly 4/1 - 5/1• Electronics Fabrication now - 5/1• Assemble Detector, Target, Support, Readout

5/1-5/15• Checkout: 5/15 – 5/30• Hall-B Install & Engineering Test 5/31 – 6/8• Physics Run October 20 – December 22,

2005!

BoNuS Detector

• Much done.– Used GEMs

– Curved GEMs.

– Made TPC and RTPC.

– Developed readout.

• Plenty to do!