1. Balloon Flight Engineering Model Balloon Flight Results
LAT - Balloon Flight Team
GSFC, SLAC, SU, Hiroshima, NRL, UCSC, Pisa
(Led by D. Thompson, G. Godfrey, S. Williams)
T. Kamae on behalf of the GLAST/LAT Collaboration
CONTENTS
Rationale and Goals
Preparation
Balloon Flight and Operations
Instrument Performance
Results VS. Geant4
Lessons Learned
Conclusions

2. Rationale: Why a Balloon Flight?
NASA Announcement of Opportunity: "The LAT proposer must also demonstrate by a balloon flight of a representative model of the flight instrument or by some other effective means the ability of the proposed instrument to reject adequately the harsh background of a realistic space environment. … A software simulation is not deemed adequate for this purpose.”
Planning the balloon flight: Identify specific goals that were practical to achieve with limited resources (time, money, and people), using the previously-tested Beam Test Engineering Model (BTEM) as a starting point.
Fig.1: Beam Test Engineering Model (’99) (BTEM), a prototype GLAST/LAT tower. The black box to the right is the anticoincidence detector (ACD), which surrounds the tracker (TKR). The aluminum-covered block in the middle is the calorimeter (CAL). Readout electronics were housed in the crates to the left.

3. Goals of the Balloon Flight
Validate the basic LAT design at the single tower level.
Show the ability to take data in the high isotropic background flux of energetic particles in the balloon environment.
Record all or partial particle incidences in an unbiased way that can be used as a background event data base.
Find an efficient data analysis chain that meets the requirement for the future Instrument Operation Center of GLAST.

4. Preparation:What Was Needed for this Balloon Flight?
A LAT detector, as similar as possible to one tower of the flight instrument - functionally equivalent. BTEM
Rework on Tower Electronics Module. Stanford U
Rework on Tracker. UCSC
Rework on Calorimeter. NRL
Rework on ACD. GSFC
External Gamma-ray Target (XGT). Hiroshima, SLAC
On-board software. SLAC, SU, NRL
Mechanical structure to support the instrument through launch, flight, and recovery. GSFC, SLAC
Power, commanding, and telemetry. NSBF, SU, SLAC, NRL
Real-time commanding and data displays. SU, SLAC, NRL
Data analysis tools. SLAC, GSFC, UW, Pisa
Modeling of the instrument response. Hiroshima, SLAC, KTH

5. Preparation: BFEM Integration at SLAC

6. Preparation: BFEM Transportation to Goddard

7. Preparation: Pre-shipment Review on July 16, 2001

8. Preparation: Pre-Launch Review
Real-time event display. A penetrating cosmic ray is seen in all the detectors.
Pre-launch testing at National Scientific Balloon Facility, Palestine, Texas. August, 2001.

9. Balloon Team at Palestine Texas

10. Flight and Operation: Launch on August 4, 2001
First results (real-time data): trigger rate as a function of atmospheric depth. The trigger rate never exceeded 1.5 KHz, well below the BFEM capability of 6 KHz.
The balloon reached an altitude of 38 km and gave a float time of three hours.

11. Flight and Operation: Onboard DAQ and Ground Electronics Worked

12. Flight and Operation: Onboard DAQ and Ground Electronics Worked

13. Instrument Performance: All Subsystems Performed Properly
External Targets (4 plastic scint)
to test direction determination
and measure interaction rate.
4 million L1T in 1 hour level flight and 100k events down linked. Many more in ascending part of flight and in HD.
ACD (13 scint. tiles)
to detect charged particles and heavy ions (Z>=2).
Tracker (26 layers of SSD) to measure charged tracks 200um and reconstruct gamma ray direction.
CAL (CsI logs)
To image EM energy deposition.

14. Instrument Performance: All Subsystems Performed Properly
Level-1 Trigger Rate (L1T)
Level Flight Data Geant4
(Default Cosmic-Ray Fluxes)
All /sec /sec
“Charged” /sec /sec
“Neutral” /sec /sec
Number of Events Recorded
Events through Downlink Events in Hard Disk
Ascending k (R53) + 109k (R54) M (R53+R54)
Level Flight k (R55)

15. Instrument Performance: ACD Threshold and Efficiency
Anti-Coincidence Detector Pulse height distr. for stiff charged particles shows clean separation of the peak from noise.
Scinti. Eff. > 99.96%
if cracks are filled with scintillator tapes.

16. Instrument Performance: CAL’s Energy Measurement and Imaging
Calorimeter
Pulse height distr. for stiff charged particles shows a single-charge peak and a peak due to alpha particles.
Alpha particles are seen.
proton
alpha
Imaging capability demonstrated

17. Instrument Performance: CPU Reboot and DAQ Livetime

18. Instrument Performance: Dead Time of DAQ as Predicted
68us
Will be = 20us in LAT

19. Instrument Performance: Tracker and XGT Association
Cosmic ray interaction in 4 External Targets (plastic scintilators)
Gamma ray produced in XGT (20 identified)
Hadronic shower produced in XGT (416 recorded)

20. Instrument Performance: Mechanical Stability Proven
BFEM has experienced ~7g shocks

21. Instrument Performance: Mechanical Stability Proven
Launch
Landing
Recovery

22. Results: Reconstruction of Events
Gamma-ray event:
Two tracks are seen in the tracker and calorimeter. Pattern recognition of an inverted “V” will allow us to selected gamma-rays from cosmic-ray background.
Charged parrticle event: The track passes through the ACD (top), the tracker, and the calorimeter.
Note: Tracker has no Si strips in the upper right
corner
“Difficult” event:
Particle and gamma –ray splashes deposit energy in ACD, Tracker, and Calorimeter.

23. Results VS Geant4: Simulation of Cosmic Ray Events
Proton spectrum
e-/e+ spectra
gamma spectrum
e-/e+
Gammas prod. by cosmic
protons in the atmosphere
Primary protons passing into the Earth magnetic field and secondary protons prod. by primary
protons in the atmosphere

24. Results VS Geant4: Charged Particles Flux and Angular Distribution
Default fluxes and angular distributions: protons, muons, and electrons
Data is higher than
the model flux! g
e-/e+
Geant4
prediction
muons
protons
90 deg.
Cosine of cosmic-ray direction
Downward

25. Results VS Geant4: “Charged” Particle Distribution
“Charged” particle hit distribution: default fluxes and angular distributions
Data is higher than
the model flux near the Calorimeter
Data
Geant4
prediction
Calorimeter side
Top of Tracker
Tracker layer number

26. Results VS Geant4: “Neutral” Particle Distribution
“Neutral” particle hit distribution: gammas and under-the-ACD electrons
Geant4
prediction
Data is higher than
the model flux above the Super GLAST layers
Data
Calorimeter side
Top of Tracker
Tracker layer number

27. Lessons Learned
Test instrument in the flight environment as much as possible:
Leak in the pressure vessel (Was very expensive for BFEM)
Two (xy) layer sets left out of L1T (Little side-entering muons on ground)
Importance of a well-tune Instrument and CR Simulator:
A strong team assigned for LAT simulation
Simulators for every steps of Integration and Testing
Detection of a small delicate fault in L1T after tuning the Geant4 simulator:
Constant monitoring of the LAT DAQ and filtering process

28. Conclusions Goals of the balloon flight were achieved.
BFEM successfully collected data using a simple three-in-a-row trigger at a rate that causes
little concern when extrapolated to the full flight unit LAT.
There seems little doubt that gamma-ray data can be extracted from the triggers and that the background can be rejected at an acceptable level.
Through the data analysis, we gained confidence in our ability to simulate the instrument and
the cosmic ray background.
Balloon flight offered a first opportunity for the LAT team to deal with many of the issues
involved in a flight program.
Lessons learned drawn from BFEM experiences will be fed back to the enitre LAT team and
that will make the flight unit development slightly easier.

29. Who Was Involved in this Balloon Flight?
D. J. Thompson, R. C. Hartman, H. Kelly, T. Kotani, J. Krizmanic, A. Moiseev, J. F. Ormes, S. Ritz, R. Schaefer, D. Sheppard, S. Singh, NASA Goddard Space Flight Center
G. Godfrey, E. do Couto e Silva, R. Dubois, B. Giebels, G. Haller, T. Handa, T. Kamae, A. Kavelaars, T. Linder, M. Ozaki1, L. S. Rochester, F. M. Roterman, J. J. Russell, M. Sjogren2, T. Usher, P. Valtersson2, A. P. Waite, Stanford Linear Accelerator Center (KTH, ISAS)
S. M. Williams, D. Lauben, P. Michelson, P.L. Nolan, J. Wallace, Stanford University
T. Mizuno, Y. Fukazawa, K. Hirano, H. Mizushima, S. Ogata, Hiroshima University
J. E. Grove, J. Ampe3, W. N. Johnson, M. Lovellette, B. Phlips, D. Wood, Naval Research Laboratory
H. f.-W. Sadrozinski, Stuart Briber4, James Dann5, M. Hirayama, R. P. Johnson, Steve Kliewer6, W. Kroger, Joe Manildi7,G. Paliaga, W. A. Rowe, T. Schalk, A. Webster, University of California, Santa Cruz
M. Kuss, N. Lumb, G. Spandre, INFN-Pisa and University of Pisa

30. Good Teamwork was the Key for our Success
Integration
Command and Data Flow
Responsibilities
Payload Responsibilities