GLAST Large Area Telescope: Science Requirements and Instrument Design Concepts Steven Ritz

GLAST LAT Project DOE/NASA CD3-Critical Design Review, May 12, 2003

Gamma-ray Large

Area Space

Telescope

GLAST Large Area Telescope:

Science Requirements and

Instrument Design Concepts

Steven Ritz

Goddard Space Flight Center

LAT Instrument Scientist steven.m.ritz@nasa.gov

S. Ritz Document: LAT-PR-01967-01 Section 03 Science Requirements and Instrument Design Concepts 1


Outline

 Context: flow of requirements, definitions, design drivers

 Instrument overview

 Simulations

 Performance calculation results review and status

 Design modifications since PDR

 Failure modes modeling and science impact

 Summary



From Science Requirements to Design

• Flow of instrument requirements from science goals and requirements was developed as part of the LAT proposal.

• LAT will meet or exceed requirements in GLAST Science

Requirements Document (433-SRD-0001).



Simplified Flow

Science Requirements Document (SRD)

Mission System

Specification (MSS)

Mission Assurance

Requirements (MAR)

Interface

Requirements

Document (IRD)

LAT

Performance

Specifications

Design Trade

Study Space

LAT Subsystem

Requirements



Aside: some definitions

Effective area

(total geometric acceptance) • (conversion probability) • (all detector and reconstruction efficiencies)

Point Spread Function (PSF)

Angular resolution of instrument, after all detector and reconstruction algorithm effects. The 2-dimensional 68% containment is the equivalent of ~1.5



(1-dim error) if purely Gaussian response. The non-Gaussian tail is characterized by the 95% containment, which would be 1.6 times the 68% containment for a perfect Gaussian response

.

2000 lower

68%

Histogram of Data

95% upper

5 y 0 1000

0

0 1 2 3 4 5

5

5 0 x

5



Science Performance Requirements Summary

From the

SRD:

Parameter

Peak Effective Area (in range 1-10 GeV)

SRD Value

>8000 cm 2

Energy Resolution 100 MeV on-axis <10%

Energy Resolution 10 GeV on-axis

Energy Resolution 10-300 GeV on-axis <20%

Energy Resolution 10-300 GeV off-axis (>60 º) <6%

PSF 68% 100 MeV on-axis

PSF 68% 10 GeV on-axis

<3.5

°

<0.15

°

PSF 95/68 ratio

PSF 55 º/normal ratio

<10%

<3

<1.7

Field of View >2sr

Background rejection (E>100 MeV)

Point Source Sensitivity(>100MeV)

Source Location Determination

GRB localization

<10% diffuse

<6x10 -9 cm -2 s -1

<0.5 arcmin

<10 arcmin



Experimental Technique

• Instrument must measure the direction, energy, and arrival time of high energy photons (from approximately 20 MeV to greater than 300 GeV):

- photon interactions with matter in GLAST energy range dominated by pair conversion: determine photon direction clear signature for background rejection

limitations on angular resolution (PSF) low E: multiple scattering => many thin layers high E: hit precision & lever arm



Pair-Conversion Telescope

Energy loss mechanisms: anticoincidence shield e + e

– conversion foil particle tracking detectors calorimeter

(energy measurement)

• must detect 

-rays with high efficiency and reject the much larger (~10 4 :1) flux of background cosmic-rays, etc.;

• energy resolution requires calorimeter of sufficient depth to measure buildup of the EM shower. Segmentation useful for resolution and background rejection.


GLAST LAT Project

Science Drivers on Instrument Design

Effective area and PSF requirements drive the converter thicknesses and layout. PSF requirements also drive the sensor performance, layer spacings, and the design of the mechanical supports.



DOE/NASA CD3-Critical Design Review, May 12, 2003

Background rejection requirements drive the ACD design (and influence the calorimeter and tracker layouts).

Field of view sets the aspect ratio (height/width)

Energy range and energy resolution requirements bound the thickness of calorimeter e + e

–

Electronics

On-board transient detection requirements, and on-board background rejection to meet telemetry requirements, are relevant to the electronics, processing, flight software, and trigger design.

Time accuracy provided by electronics and intrinsic resolution of the sensors.

Instrument life has an impact on detector technology choices.

Derived requirements (source location determination and point source sensitivity) are a result of the overall system performance.



IRD and MSS Constraints Relevant to LAT

Science Performance

• Lateral dimension < 1.8m

Restricts the geometric area.

• Mass < 3000 kg

Primarily restricts the total depth of the CAL.

• Power < 650W

Primarily restricts the # of readout channels in the TKR

(strip pitch, # layers), and restricts onboard CPU.

• Telemetry bandwidth < 300 kbps orbit average

Sets the required level of onboard background rejection and data volume per event.

• Center-of-gravity constraint restricts instrument height, but a low aspect ratio is already desirable for science.

• Launch loads and other environmental constraints.



Overview of LAT



• Precision Si-strip Tracker (TKR) 18

XY tracking planes. Single-sided silicon strip detectors (228 m m pitch) Measure the photon direction; gamma ID.

• Hodoscopic CsI Calorimeter(CAL)

Array of 1536 CsI(Tl) crystals in 8 layers.

Measure the photon energy; image the shower.

• Segmented Anticoincidence Detector

(ACD) 89 plastic scintillator tiles.

Reject background of charged cosmic rays; segmentation removes self-veto effects at high energy.

• Electronics System Includes flexible, robust hardware trigger and software filters.

ACD

[ surrounds

4x4 array of

TKR towers] e + e

–

Tracker

Calorimeter

Systems work together to identify and measure the flux of cosmic gamma rays with energy 20 MeV - >300 GeV.



Tracker Optimization

• Converter thickness profile iterated and selected.

• Resulting design: “FRONT”: 12 layers of 3% r.l. converter

“BACK”: 4 layers of 18% r.l. converter followed by 2 “blank” layers

• Large A eff with good PSF and improved aspect ratio for BACK.

• Two sections provide measurements in a complementary manner:

FRONT has better PSF, BACK enhances photon statistics.

TKR has ~1.4 r.l. of material.

Combined with ~8.4 r.l. CAL provides ~9.8 r.l. total.



Design Performance Validation:

LAT Monte-Carlo Model

LAT design based on detailed

Monte Carlo simulations.

Detailed detector model includes gaps, support material, thermal blanket, simple spacecraft, noise, sensor responses…

Integral part of the project from the start.



Background rejection



Effective area and resolutions



Trigger design



Overall design optimization

Simulations and analyses are all C++, based on standard HEP packages.

Instrument naturally distinguishes gammas from backgrounds, but details matter.

gamma ray proton



Monte Carlo Modeling Previously Verified in

Detailed Beam Tests

High-level performance parameters Detailed detector characteristics

(e.g., PSF) (e.g., hit multiplicities)

10

GLAST Data

68% Containment

95% Containment

(errors are 2



)

1

Monte

Carlo

0.1

10

1

10

2

10

3

Energy (MeV)

10

4

1997 SLAC beam test

(photons, positrons)

Demonstrate silicon conversion telescope principle

Published in NIM A446

1999-2000 SLAC beam test

(photons, positrons, protons) flight-scale tower

Published in NIM A474


GLAST LAT Project background event candidate: gamma event candidate:

DOE/NASA CD3-Critical Design Review, May 12, 2003

LAT Balloon Flight

Purpose of balloon test flight: expose prototype LAT tower module to a charged particle environment similar to space environment and accomplish the following objectives:

 Help validate the basic LAT design at the single tower level.

 Demonstrate the ability to take data in the isotropic background flux of energetic particles in the balloon environment.

 Record events for background flux analysis.

All Objectives met by Balloon Flight on

August 4, 2001 (3 hrs at 38 km float)

All subsystems performed properly.

Trigger rate <1.5 kHz, well below BFEM 6 kHz capability.



Calibration Strategy

• Every LAT science performance requirement has a defined test.

• LAT energy range and FOV are vast. Testing will consist of a combination of simulations, beam tests, and cosmic ray induced ground-level muon tests.

– beam tests verify the simulation and sample performance space; analysis using the simulation is used to verify the full range of performance parameters.

• For science performance verification, beam tests can be done with a few towers together. Full-LAT tests are functional tests.

55º, 1 GeV

 side view front view

See LAT-TD-00440 and LAT-MD-00446



Implemented Maximum Background Fluxes

total orbit-max fluxes used for trigger rate calculations

Integrates to ~10 kHz/m 2

• LAT-TD-00250-01 Mizuno et al

• Note by Allan Tylka 12 May 2000, and presentations by Eric Grove

• AMS Alcaraz et al, Phys Lett B484(2000)p10 and Phys Lett

B472(2000)p215

• Comparison with EGRET A-Dome rates provides a conservative ceiling on the total rate.



Implemented Average Background Fluxes

Integrates to ~4.2 kHz/m 2 orbit-avg fluxes used for downlink and final background rejection calculations



Instrument Triggering and Onboard Data Flow

Level 1 Trigger On-board Processing

Hardware trigger based on special signals from each tower; initiates readout

Function: • “did anything happen?”

• keep as simple as possible full instrument information available to processors.

Function: reduce data to fit within downlink

Hierarchical filter process: first make the simple selections that require little CPU and data unpacking.

x x x

• TKR 3 x • y pair planes in a row workhorse

 trigger

OR

• CAL:

LO – independent check on TKR trigger .

HI – indicates high energy event disengage use of ACD.

Upon a L1T, all towers are read out within 20 m s

• subset of full background rejection analysis, with loose

• complete event information cuts

• only use quantities that

 are simple and robust

 do not require application of sensor

Total L3T Rate: <25-30 Hz> calibration constants

• signal/bkgd tunable, depending on analysis cuts:

: cosmic-rays ~ 1:~few

(average event size: ~8-10 kbits)

On-board science analysis:

Instrument Total L1T Rate: <4 kHz> ** transient detection (AGN flares, bursts)

**4 kHz average without throttle (1.3 kHz with Spacecraft throttle); peak L1T rate is approximately 12 kHz without throttle and 3.8 kHz with throttle).



On-board Filters

• select quantities that are simple to calculate. Intelligent use of

ACD information to preserve acceptance of high-energy events. Filter scheme is tunable.

• Filters use

– ACD info: match simple tracks to selected hit ACD tiles, count # hit tiles at low energy

– CAL info: energy deposition pattern consistent with downwardgoing electromagnetic interactions.

– TKR info: remove low-energy particles up the ACD-TKR gap by projecting track to CAL face and selecting on XY position; for very low CAL energy, require TKR hit pattern inconsistent with single prong.

• filter first designed using LAT simulation/reconstruction, then pieces implemented by FSW group. Now wrapping existing

FSW code for use in LAT simulation/recon packages for verification and optimization for science.



On-board Filters Design Results

• After all selections, average background rate is 17 Hz.

16.5 Hz total rate composition:

5 Hz line

2 Hz line

1 Hz line chime albedo albedo CRe albedo p  e+e-

Additional margin available: much of the residual rate is due to high-energy proton and electron events with CAL E>5GeV -- if apply ACD selections onboard to higher energy, rate can be cut in half (to 8 Hz), with ~5% reduction in Aeff at 10 GeV.



Ground-based Background Rejection Analysis

• Evaluation of the performance parameters (Aeff, PSF, etc.) requires the final background rejection analysis.

• Different science analyses can optimize final background rejection selections differently.

– The most stringent requirement driver is the extragalactic diffuse flux.

• Exercises the Science Analysis Software infrastructure:

– generate 50M background events, with a unique tag for each. Machinery to cull lists of events, single event display, analysis, etc.

– Also generate adequate sample of extra-galactic diffuse photon flux and pass it through the analysis chain.



Summary of Ground-Based Background

Rejection Analysis

• The background rejection analysis meets the requirements and is continuing to evolve, becoming more sophisticated, effective, and efficient.

• Current selections using subsystem info:

– TKR: use information from track fits, hit pattern inconsistent with single track at low energy, PSF cleanup cuts, location cuts, loose consistency between track multiple scatter and CAL energy.

– CAL: require xtal hit patterns (shower shapes) to be consistent with downward-going EM showers; more precise track-CAL energy centroid matching.

– ACD: very low energy events require quiet ACD; hidden shower rejection.



Background Rejection Results

• Requirement: <10% contamination of the measured extragalactic diffuse flux for

E>100 MeV

• Residual background is 5% of the diffuse (6% in the interval between 100 MeV and 1 GeV).

Important experimental handle: large variation of background fluxes over orbit – compare diffuse results over orbit.

• Below 100 MeV [no requirement], without any tuning of cuts for low energy, fraction rises to 14%.

This will improve.

• Peak effective area: 10,000 cm 2

(at ~10 GeV).

• At 20 MeV, effective area after onboard selections is 630 cm 2 .

Different physics topics will require different (and generally less stringent) background rejection on the ground.

Diffuse flux, after cuts, scaled to generated background log(E) corrected (MeV)

100 MeV 1 GeV 10 GeV 100 GeV



Energy Resolution

• Energy corrections to the raw visible CAL energy are particularly important for

– depositions in the TKR at low photon energy (use TKR hits)

– leakage at high photon energy (use longitudinal profile)

18%

Normal-incidence

100 MeV



(require <10%)

>60° off-axis 300 GeV



(require <6%)

4.3%

9% uncorrected corrected

E(MeV) corrected E(GeV)



PSF

68% containment

95% containment

100 MeV, normal incidence

FRONT

68% containment radius:

100 MeV

Requirement: <3.5°

3.37° FRONT

4.64° Total

95/68 ratio

Requirement: <3

FRONT: 2.1

(BACK: 2.6)

10 GeV

Requirement: <0.15°

0.08° FRONT

0.115° Total

NOTE: With older version of

TKR reconstruction software, the 95/68 ratio is not <3 everywhere (particularly at high energy). This is a software issue, not an instrument design issue.

TKR reconstruction software under upgrade to improve hit association and more precise vertexing.



Field of View

• Defined as the integral of effective area over solid angle divided by peak effective area:

1.2

10 GeV, uniform flux, all cuts

10 GeV, uniform flux, all cuts

1 cos( q

)

0.8





180

)

0.6

0.4

LAT

FOV

0.2

0 cos( q

) q

0.2

0 10 20 30 40 50 60 70

0 10 20 30 40 50 60 70 deg

Favorable LAT aspect ratio provides large FOV

TKR

TKR

Requirement: >2 sr

LAT current: 2.4 sr

CAL

CAL



Science Performance Requirements Summary

From the SRD:

Parameter

Peak Effective Area (in range 1-10 GeV)

SRD Value

>8000 cm 2

Energy Resolution 100 MeV on-axis

Energy Resolution 10 GeV on-axis

<10%

<10%

Energy Resolution 10-300 GeV on-axis <20%

Energy Resolution 10-300 GeV off-axis (>60 º) <6%

PSF 68% 100 MeV on-axis

PSF 68% 10 GeV on-axis

<3.5

°

<0.15

°

PSF 95/68 ratio

PSF 55 º/normal ratio

<3

<1.7

Field of View

Background rejection (E>100 MeV)

Point Source Sensitivity(>100MeV)

Source Location Determination

GRB localization

>2sr

<10% diffuse

<6x10 -9 cm -2 s -1

<0.5 arcmin

<10 arcmin

Present Prediction

10,000 cm 2 at 10 GeV

9%

8%

<15%

<4.5%

3.37

° (front), 4.64° (total)

0.086

° (front), 0.115° (total)

2.1 front, 2.6 back (100 MeV)

1.6

2.4 sr

6% diffuse (adjustable)

3x10 -9 cm -2 s -1

<0.4 arcmin

(ignoring BACK info)

5 arcmin

(ignoring BACK info)



Updates to the Simulation/Reconstruction

• Simulation and reconstruction packages completely rewritten since

PDR:

– support flight, with more robust architecture

– very large effort, validation proceeding

– re-evaluating instrument performance using new reconstruction.

Some major improvements already apparent:

100 MeV, normal incidence (<5 º), front: 68% containment – 2.4º

98% containment – 7.1º

VERY PRELIMINARY

– support for misalignments

– hooks into hardware databases

(bad channels, etc.) and ability to introduce hardware failures easily

• Now wrapping onboard filter into recon to have high-fidelity assessment of performance (expected complete July)

• Support for science verification and calibration planning (see I&T talk)

• Support for Data Challenges (end-to-end testing, see SAS status talk)



Design Modifications Since PDR

• Most changes are in the mechanical/thermal system and are outside the instrument FOV. Negligible impact on particle propagation.

• Deleted geographic information in TKR trigger signal (used only for optional ACD hardware throttle). Now using much simpler throttle, with whole tower-tile associations.

– Very small impact on instrument triggered FOV when throttle is turned on:

– Even smaller after analysis cuts cos (q)

45° 65°



Failure Modes Impacts on Science Performance

• Support for reliability calculations and design evaluation

• Can analyze identical events with/without failure turned on.

Difficult part is putting awareness into the reconstruction (as we would if failure really happens).

• Highest priority has been ACD: loss of a single tile. Potential impact on trigger rates and downlink volume – could we operate? Needed primarily for micrometeoroid shield reliability requirements.

• TKR: impact on trigger efficiency of loss of a layer within a tower:

• thick converter layer: 0.6% loss of triggered Aeff

• thin converter layer: <0.1% loss of triggered Aeff

• Effects of loss of a tower



Impact of ACD Tile Failures

Main issue here is the micrometeoroid shield reliability analysis. Can we tolerate loss of a single tile?

YES

Damage none

After L1T throttle

1.8 kHz center top tile 1.9 kHz corner top tile 1.9 kHz

After filters [Hz] After simple mitigation [Hz]

17

27

24

N/A

19

17

Simple mitigation when a top tile is non-functional: require found track NOT to start in silicon layer directly underneath. Loss of effective area is small. Additional actions available, if needed.



Impact of Loss of a Tower

• Three kinds of towers, by symmetry

– corner, edge, core

• Requirement: <20% degradation in 5 years

 peak Aeff>6,400 cm 2 ; FOV>1.6 sr

• Use simulation to study loss of a TKR tower (#6, core) on triggered Aeff:

– 1 GeV normal incidence: 6% loss

– 1 GeV 40 ° off axis: 4% loss

• Difficult to estimate all impacts accurately – an actual loss will trigger more intense study of mitigations. Rough estimates of impact:

• corner: peak Aeff ~9,000 cm 2 , FOV ~ 2.4 sr

• edge: peak Aeff ~7,500 cm 2 , FOV ~2.1 sr (asymmetric)

• core: peak Aeff ~7,500 cm 2 , FOV ~1.9 sr (asymmetric)



LAT Has Three Tower Location Types

Loss of a corner is not as bad as loss of an edge is not as bad as loss of a core.

core edge corner

Simplifying assumption: “tower loss” means both CAL and TKR. ACD is still intact.



Summary

• LAT will meet or exceed science requirements from the SRD.

• Much work done over the past year on simulation and reconstruction. Moving forward:

– first filter integration complete: July

– new instrument response functions, using improved reconstruction algorithms: September (collaboration science meeting, data challenge start)

• In addition to design optimization and performance evaluation, detailed simulation is being used as a system engineering tool,

e.g., supporting FMEA.

• See following talks for design details.



Backup Slides



EGRET A-dome Rates

(from D. Bertsch, EGRET team)

SAA

A-dome has an area of ~6 m 2 , so orbit max rate (outside

SAA and no solar flares) corresponds to

~16 kHz/m 2

This represents a conservative upper-limit for us, since the

A-dome was sensitive down to 10’s of keV.

Note peak rate is at

(24.7,260)



Onboard Filter Development

• Filter designs done with the full simulation and ground-based reconstruction, in consultation with FSW group. Demonstration of principles, included in science performance evaluations.

• FSW implemented most of the filter design for benchmarking on the flight processor.

– filtering is hierarchical. Most important to implement the selections that are run first (highest rate, largest multiplier on CPU demand). More cycles/event available for remaining event sample after each step.

• FSW implementation is being wrapped for inclusion in the simulation/recon packages.

– very early functional testing of the flight algorithms, with high fidelity. Examine details (e.g., existing track finding) using full set of SAS tools, event display, etc.

– detailed evaluation of the filter effects on the science performance

– opportunity for a tuning iteration and optimization of the final set of selections



Summary of Filter and Status

Primary Info

ACD

ACD-TKR

CAL

CAL

CAL

TKR-CAL

TKR

TKR

TKR-CAL

TKR

TKR

Design Selection

Tile counts (energy dependent)

Track match with tile

Simple energy selections

Layer ratios

Simple topologies

Track match with energy centroid

Skirt only cut

Simple hit pattern inconsistent with single prong at low energy

Minimal #tracks and CAL E, or make additional demands

Earth direction

TKR hits consistent with a track near

CAL if E>0

FSW Status

DONE

DONE

DONE

DONE

DONE

DONE

DONE



Detector Choices

• TRACKER single-sided silicon strip detectors for high hit efficiency, low noise occupancy, resolution, reliability, readout simplicity. Noise occupancy requirement primarily driven by trigger.

• CALORIMETER hodoscopic array of CsI(Tl) crystals with photodiode readout for good resolution over large dynamic range; modularity matches TKR; hodoscopic arrangement allows for imaging of showers for leakage corrections and background rejection pattern recognition.

• ANTICOINCIDENCE DETECTOR segmented plastic scintillator tiles with wavelength shifting fiber/phototube readout for high efficiency (0.9997 flows from background rejection requirement) and avoidance of ‘backsplash’ self-veto.



SRD LAT Requirements (I)



SRD LAT Requirements (II)



SRD LAT Requirements (III)



Testing Trigger Efficiencies

Two kinds of inefficiencies

LOCAL GLOBAL at least 2 methods to measure:

• spatial distributions of L1T’s

• compare TKR hits with TKR trigger pattern using both

• TKR triggers

• CAL-LO triggers (independent sample!) example: global trigger drops every

10 th trigger. How would we know?

Two types of global inefficiencies:

Time-dependent monitor pulsar fluxes over time

Constant

At least 3 methods to measure:

• count prescales

• periodic triggers

Test this on the ground in LAT using cosmic-ray induced

(both hard and soft)

• use a sensor with a counter independent muons of trigger system

(e.g., ACD tile) to generate heavily prescaled triggers.



On-board Science Requirements

• Requirements on transient recognition and localization onboard are modest.

• Requirements specify the characteristics of the flux (#photons, energy, time period) to be detected:

– 10 arcmin(3 arcmin goal) GRB localization accuracy onboard for any burst that has >100 detected photons with E>1 GeV arriving within 20 seconds.

– 5 second (2 second goal) notification time to spacecraft relative to time of detection of GRB

– NO requirements on AGN flare detection onboard in SRD.

NO mission minimum.



Broad Implications for Onboard Reconstruction

• With 100 photons, the mean per-photon pointing error for E>1 GeV onboard is comfortably LARGE. Rudimentary gamma direction reconstruction onboard is anyway needed for earth albedo gamma rejection.

The crude precision needed to meet the SRD requirements is not a driver.

• The expected (non-transient) gamma rate (E>20 MeV) is a few Hz; the residual rate for downlink (independent of science analysis onboard) is ~30 Hz. Recognizing that >100 photons with E>1 GeV have arrived within a 20 second interval from one direction should not require precision reconstruction or background rejection better than that needed to meet telemetry requirements.

• Simple (low resource) prototype algorithms exist, but not yet in FSW environment. Attempt first to reuse filter’s track finding once embedded into recon environment to study performance.



Additional Goals (no requirement)

• AGN flare localization < 2 ° for flares that occur within the LAT

FOV at high gal. lat., with

D

F > 2 x 10 -6 ph cm -2 s -1 (E > 100 MeV) that last for more than 1000 secs.

• For these flare parameters, notify spacecraft <1 min after recognition of flare.

• This recognition and localization goal is not easy:

• Assuming 30 Hz event rate (residual photon candidate rate after standard onboard filter), on average there will be ~5 photons per square degree bin in the FOV after 1000s.

Change in flux corresponds to ~5 detected photons.

• => Meeting the goal is not possible without additional onboard filtering (additional factor 10 needed --- not crazy to consider).

• This is a GOAL, and is not a driver. First approach should be to extend GRB onboard analysis to longer timescales and see how well we do. A simple energy cut will also help here.

• The science justifies making an effort! However, it is lower priority than other onboard software tasks.



On-board Filters

• select quantities that are simple to calculate and that do not require individual sensor calibration constants. Filter scheme is flexible – current set is basis for flight implementation.

• order of selections to be optimized. Grouped by category for presentation purposes:

– ACD info: match track to hit tile, count # hit tiles at low energy outside tile boundary inside tile boundary

Background

100 MeV

 no tile hit

Rate after

ACD selections is 180 Hz orbit-avg

(360 Hz orbit-max)

[cm] [cm]



On-board Filters (II)

– CAL info: most of the residual rate at this point is due to albedo events and other upward-going energy events.

Require track-CAL energy centroid loose match, fractional energy deposit in front layer reasonably consistent with downward EM energy flow. If no CAL energy, require track pattern inconsistent with single-prong.

Rate after CAL selections is ~80

Hz orbit-avg (130 Hz orbit-max)

– TKR info: low-energy particles up the ACD-TKR gap easily dealt with: project track to CAL face and require XY position outside this band; for low CAL energy, require TKR hit pattern inconsistent with single prong.

X (cm)



Effects of Rocking: Albedo Gammas

albdeo gamma

L1T with Throttle back full background flux front

As we rock, the spike spreads in q, f

: cos ( q

)

At zenith, earth horizon is at 113 degrees. Study what happens when observatory rocks to 35 and 60 degrees off zenith.

f f

Back Front Back Front cos( q) cos( q)

35 degree rock 60 degree rock



Albedo Gamma Rates

zenith rock 35 °

L1T rate [Hz] L1T rate with

Throttle [Hz]

250 190

260 200

After filters

[Hz]

2

3

After fiducial cut [Hz]

2

(no cut)

3

(no cut) rock 60 ° 300 250 8 1 (<45 °)

3 (<53 °)

Notes:

• rates for other backgrounds will be reduced somewhat by the same angle cut, not taken into account here.

• small incremental L1T rate not a problem

• calculating the gamma direction only happens at a relatively low rate, if needed (after other filters), so incremental CPU load not a problem.

• can reduce the downlink contribution to whatever we need with a tighter fiducial cut.



Pointing Knowledge

 Context: requirements and their origins

 Overview of pointing knowledge elements

 Details

 Experimental technique

 Intrinsic LAT measurement characteristics

 LAT mechanical/thermal distortion effects

 Observatory/instrument coordinates to sky coordinates

 Resulting pointing knowledge and source localization

 Calibration techniques and stability timescales

 Precision of corrections and uncorrectable effects

 Sequence of calibrations on orbit

 Required supporting ground-based calibrations



Context: Requirements Summary

• Observatory pointing knowledge: 10 arcsec

– The single photon direction measurement ability of the LAT is not very precise, on average, but the best photon LAT will ever measure will have an error of ~ 30 arcsec.

– The pointing knowledge should be better than the best single photon error to avoid throwing away information.

The single photon accuracy does not, by itself, set the scale of the pointing knowledge, however, since LAT will measure many photons from each source. (see following slide on localization)

• The mission assigned suballocations as follows:

– LAT internal: 7 arcsec

– Spacecraft (both GN&C and thermal/mechanical): 6 arcsec

– Calibration residual: 4 arcsec

– Jitter: 1 arcsec



LAT Source Localizations

SRD 0.5 arcmin 1  radius for 10 -7 source

10 arcsec,

1  radius

Systematics will dominate here note!

figure is from LAT proposal.

Will be updated.

There is no critical value (original SRD requirement was a factor 2 more stringent).



Overview of Pointing Knowledge Elements

Source localization

Ancillary packet info

Single photon direction measurement (sky coords)

LAT-SC calibration

Correction algorithm

Uncorrected effects

Single photon direction measurement

(observatory/instrument coordinates)

Internal LAT mech/thermal effects

High-level LAT measurement characteristics

Intrinsic low-level LAT measurement characteristics

Experimental technique



Experimental Technique and Science Drivers on Instrument Design

Effective area and PSF requirements drive the converter thicknesses and layout. PSF requirements also drive the sensor performance, layer spacings, and drive the design of the mechanical supports.

 Background rejection requirements drive the ACD design (and influence the calorimeter and tracker layouts).

Field of view sets the aspect ratio (height/width)

Energy range and energy resolution requirements bound the thickness of calorimeter e + e

–

Electronics

On-board transient detection requirements, and on-board background rejection to meet telemetry requirements, are relevant to the electronics, processing, flight software, and trigger design.

Time accuracy provided by electronics and intrinsic resolution of the sensors.

Instrument life has an impact on detector technology choices.

Derived requirements (source location determination and point source sensitivity) are a result of the overall system performance.



Tracker/Converter Issues

Expanded view of converter-tracker:



Some lessons learned from simulations

At low energy, measurements at first two layers dominate due to multiple scattering

X

Y

At 100 MeV, opening angle ~ 20 mrad

All detectors have some dead area: if isolated, can trim converter to cover only active area X

Y

Low energy PSF completely dominated by multiple scattering effects: q

0

~ 2.9 mrad

(scales as (x

0

) ½ )

/ E[GeV]

High energy PSF set by hit resolution/lever arm

X

Y

PSF

At low energy, direction measurement is dominated by planes closest to conversion point.

~1/E

Roll-over and asymptote ( q

0 and q

D

) depend on design

At higher energies, more planes contribute information:

Energy # significant planes

100 MeV

1 GeV

10 GeV

2

~5

>10

E



Intrinsic LAT Measurements Characteristics

•

Low-level

–

Strip pitch: 228 m m

– Plane spacing: ~3 cm

– Tower size: ~60 cm tall, ~37 cm wide

– Geometric characteristics:

• hit resolution/plane spacing: ~2 mrad (410 arcsec)

• hit resolution/tower height: 0.1 mrad (21 arcsec)

– thus, on average, a tower rotation of 7 arcsec will have a very small impact on the strip hit distribution in a single event.

• stack diagonal angle: 32 degrees

•

High-level

–

Photon direction determined by reconstruction software algorithms.

• Different categories of events and corresponding measurement quality.

• Kalman filter is used. Optimal algorithm to handle both multiple scattering and geometric effects.

• At low energy, first measurements dominate due to multiple scattering.

Even for cross-tower events, it is the measurements in the tower containing the conversion that dominate the direction measurement.

•

At high energy, it is the tower with the largest track path length that dominates the direction measurement.

To good approximation, the individual tower misalignment is also the photon location misalignment. This also represents the pessimistic case.



Overview of Pointing Knowledge Elements




LAT-SC calibration


Uncorrected effects









Converting to Sky Coordinates

• Information from ancillary data packet

– Attitude information is conveyed to LAT from spacecraft every 200ms. Includes both attitude and angular velocities.

• maximum angular velocity is 30 deg/minute, or 360 arcsec/update, so velocity info is necessary with a timestamp better than ~0.5ms. OK.

• angular acceleration terms are negligible (<0.25 arcsec)

• LAT-SC calibration residual

– The error from the calibration is carried forward, but can be updated over time (see later slide on on-orbit calibrations).



Calibration Techniques and Stability Timescales

• Internal LAT alignment using cosmic rays (straight trajectories)

– standard technique for particle trackers

– no external measurements or references needed

– ground: muons from cosmic ray airshowers

– on-orbit: proton cosmic rays

– statistics set by detail needed

– individual sensors require higher statistics. Stable.

– tower-to-tower lower statistics needed. Varies on thermal timescales.

• LAT to Observatory

– done on orbit. Observe known sources. Precision/statistics source-dependent. Brighter sources provide ~10 arcsec localization knowledge in ~1 month.

– after initial alignment observation, all-sky survey data can be used for continuous update/refinement

=> effects that are slow compared to ~2 months are automatically calibrated out.



Sequence of On-Orbit Calibrations

• LAT SVAC Plan (LAT-MD-00446).

• First, LAT internal alignment using CR’s.

• Then, initial two-week observations to know LAT-SC alignment to better than 15 arcsec (easily sufficient for most science topics).

– optimization of initial observing strategy (source selection, optimal orientation, etc.) under investigation.

• Then, proceed with sky survey and use known sources to reduce the error over year 1.



Supporting Ground Measurements (I)

• Captured in LAT survey plan,

SVAC plan, I&T survey procedures documents.

• Demonstration of technique:

– introduce misalignments into the simulation and reconstruction

– implement muon flux to simulate the measurements, with LAT pointed horizontally.

• Remaining systematic offsets under study, but are already sufficiently small.

• The ground-based tower-to-tower alignment information doesn’t carry forward significantly to flight. Launch loads will change the alignment, and it will anyway be re-established using cosmic rays on orbit.



Supporting Ground Measurements (II)

• The main purpose of the ground-level muon measurements is to test functionality and as part of the mechanical/thermal alignment requirements testing and analysis.

• LAT ground-level muon measurements will be made with LAT

Zaxis parallel to the ground (“sideways”).

• Verified rate of tracks crossing tower pairs with the simulation: m m tower pairs tower pairs

Current estimate: sufficient statistics in one day to do tower-to-tower verification measurement.

Good match to LAT thermal timescale.



Alignment Summary




LAT-SC calibration


Uncorrected effects









Loss of corner tower

• easiest case to estimate

• Aeff loss ~10%

• FOV loss very small.

core edge corner good experimental handle on-orbit: can pretend any other working corner tower is off and check for incremental background leakage and other systematic effects.



Loss of single edge tower

• corner closest to dead edge tower also significantly impacted FOV.

For purposes of conservative estimate, assume only 3x4 LAT left.


• FOV loss ~10%.

core edge corner good experimental handle on-orbit: can pretend any other working edge tower is off and check for incremental background leakage and other systematic effects.



Loss of single core tower

• most difficult (and painful) case.

For purposes of conservative estimate, assume whole quadrant compromised. Looks like 2 LATs, each 1x2, with overlap.


• FOV loss ~15%. (~35% loss in 1-dim in the bottom left and top right quadrants) core edge corner good experimental handle on-orbit: can pretend any corresponding tower is off and check for incremental background leakage and other systematic effects.
